Social psychology

BRIEF REPORT

Social Heuristics and Social Roles: Intuition Favors Altruism for
Women but Not for Men

David G. Rand and Victoria L. Brescoll
Yale University

Jim A. C. Everett
Oxford University

Valerio Capraro
Center for Mathematics and Computer Science,

Amsterdam, the Netherlands

Hélène Barcelo
Mathematical Sciences Research Institute, Berkeley, CA

Are humans intuitively altruistic, or does altruism require self-control? A theory of social heuristics,
whereby intuitive responses favor typically successful behaviors, suggests that the answer may depend
on who you are. In particular, evidence suggests that women are expected to behave altruistically, and
are punished for failing to be altruistic, to a much greater extent than men. Thus, women (but not men)
may internalize altruism as their intuitive response. Indeed, a meta-analysis of 13 new experiments and
9 experiments from other groups found that promoting intuition relative to deliberation increased giving
in a Dictator Game among women, but not among men (Study 1, N � 4,366). Furthermore, this effect
was shown to be moderated by explicit sex role identification (Study 2, N � 1,831): the more women
described themselves using traditionally masculine attributes (e.g., dominance, independence) relative to
traditionally feminine attributes (e.g., warmth, tenderness), the more deliberation reduced their altruism.
Our findings shed light on the connection between gender and altruism, and highlight the importance of
social heuristics in human prosociality.

Keywords: altruism, prosociality, intuition, dual process, gender

Supplemental materials: http://dx.doi.org/10.1037/xge0000154.supp

Humans often choose to help others. Yet such prosociality
typically requires us to expend time, effort, and money. What
makes us willing to do so? Recently, there has been considerable
interest in exploring human prosociality using a dual process
perspective (for a review, see Zaki & Mitchell, 2013), where
decisions are conceptualized as resulting from competition be-
tween cognitive processes that are automatic, fast, and intuitive,
versus those that are controlled, slow, and deliberative (Kahneman,

2003; Sloman, 1996). Does prosociality require deliberative self-
control, or do prosocial impulses get reined in by the calculus of
self-interest?

The Social Heuristics Hypothesis (SHH) has been proposed as a
theoretical framework for answering this question (Peysakhovich
& Rand, 2015; Rand et al., 2014). The SHH adds an explicitly dual
process lens to theories regarding the adoption of typically advan-
tageous behaviors (e.g., theories based on “spillover” effects [Ki-
yonari, Tanida, & Yamagishi, 2000], norm internalization [Chudek
& Henrich, 2011], and consequences of interdependence in one’s
social interaction experiences [Van Lange, De Bruin, Otten, &
Joireman, 1997]). The SHH posits that the social strategies which
are typically successful in daily life become automatized specifi-
cally as intuitive responses. Deliberation can then override these
intuitions and adjust one’s behavior in light of the details of the
specific decision at hand.

In particular, the SHH argues that a key component of deliber-
ation is the consideration of strategic concerns and payoff maxi-
mization, which favors self-interested behavior. As a result, delib-
eration is predicted to sometimes undermine prosocial intuitions,
but not to push selfish intuitions toward prosociality. A mathemat-
ical model of dual-process agents playing Prisoner’s Dilemma
games formalizes this prediction (Bear & Rand, 2016): among all

This article was published Online First February 25, 2016.
David G. Rand, Departments of Psychology and Economics, and School

of Management, Yale University; Victoria L. Brescoll, School of Manage-
ment, Yale University; Jim A. C. Everett, Department of Experimental
Psychology, Oxford University; Valerio Capraro, Center for Mathematics
and Computer Science, Amsterdam, the Netherlands; Hélène Barcelo,
Mathematical Sciences Research Institute, Berkeley, CA.

Funding by the John Templeton Foundation and the National Science
Foundation (Grant 0932078000) is gratefully acknowledged. We also
thank the authors of the studies included in our meta-analysis for sharing
their data.

Correspondence concerning this article should be addressed to David G.
Rand, Department of Psychology, Yale University, Box 208205, New
Haven, CT 06520-8205. E-mail: [email protected]

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

Journal of Experimental Psychology: General © 2016 American Psychological Association
2016, Vol. 145, No. 4, 389 –396 0096-3445/16/$12.00 http://dx.doi.org/10.1037/xge0000154

389

possible strategies, the dual-process strategies that perform best
(and thus are favored by evolution, social learning, and/or strategic
decision-making) use deliberation to switch from cooperation to
defection in one-shot anonymous settings. Conversely, strategies
that use deliberation to switch from defection to cooperation
(under any circumstances) never perform well, and are always
disfavored.

By this account, where typically successful strategies are intu-
itive, intuition should favor cooperation for most people: in the
context of daily life, most important interactions (e.g., with co-
workers, friends, and family) are repeated. Thus, because cooper-
ation is non-zero-sum, cooperating can be in one’s long-run self-
interest: cooperating with others today can induce others to
cooperate with you in the future (Rand & Nowak, 2013). Con-
versely, when interacting in settings where future consequences
are not enough to incentivize cooperation (e.g., one-shot anony-
mous laboratory experiments), it is never in one’s self-interest to
cooperate; and, therefore, deliberation should favor selfishness. As
predicted by this account, experimentally promoting intuition rel-
ative to deliberation via time pressure or a conceptual priming
exercise has been found to increase cooperation on average in
one-shot anonymous interactions (Cone & Rand, 2014; Lotz,
2015; Protzko, Ouimette, & Schooler, 2015; Rand, Greene, &
Nowak, 2012; Rand, Newman, & Wurzbacher, 2015); for a meta-
analysis, see Rand et al. (2014).

The implications of the SHH for altruism (unilaterally giving
resources to others), however, remain unclear. On the one hand, a
narrow read of the SHH suggests that, like deliberation, intuition
should disfavor altruism: transferring money to someone and then
having them transfer it back does not make one better off than just
keeping the original money, and so altruism (unlike cooperation) is
not advantageous even in repeated games. On the other hand, a
broader interpretation of the SHH suggests that intuition may favor
altruism in a similar way to what has been observed with cooper-
ation. If being selfish in the context of zero-sum interactions is
seen negatively by others, it may create reputational costs in the
context of other (non-zero-sum) interactions. If so, then altruism
could be payoff maximizing in the long run. However, it may not
be the case that all people are harmed from being seen as selfish,
such that moderators may exist for whether altruism is advanta-
geous in daily life (and thus favored by intuition).

A particularly compelling candidate for such moderation is
gender. Specifically, we might expect women, but not men, to
have altruism as their intuitive social response for two reasons.
First, a large body of work suggests that the behavior of men and
women is governed by stereotypes concerning their social roles;
and in particular that women are expected to be communal and
unselfish, whereas men are expected to be agentic and independent
(Eagly, 1987). When women behave in ways that are perceived as
insufficiently communal, they are not only liked less, but they are
also less likely to be helped, hired, promoted, paid fairly, and given
status, power, and independence in their jobs (Heilman &
Okimoto, 2007). Thus, women are subject to much stronger ex-
pectations that they will behave altruistically (Heilman & Chen,
2005). Furthermore, recent work has found that women are well
aware of these gender stereotype-based behavioral prescriptions,
and their concern over encountering backlash effects from violat-
ing these stereotypes helps explain, in part, a range of behaviors
that systematically vary by gender (Brescoll, 2011). As a result,

behaving altruistically in accordance with others’ expectations is
typically advantageous for women.

Second, the fact that women disproportionately occupy roles
that either mandate self-sacrificing and altruistic behavior (e.g.,
mother) or, at the very least, require a great deal of other-oriented,
communal behavior (e.g., nurse; Eagly, 1987), may cause women
to habituate to being altruistic. And even women who do not
explicitly occupy such family or work roles may acquire altruistic
intuitive social responses because female peer groups are markedly
more communal and egalitarian than male peer groups, and thus
make self-sacrificing, unselfish behavior socially adaptive (Mac-
coby, 1998). Taken together, consideration of both the expecta-
tions of others and the behaviors one engages in regularly point to
intuition favoring altruism for women more so than men.

In this paper, we experimentally investigated the role of intu-
ition and deliberation in altruism, and the potential moderating role
of gender. In Study 1, a meta-analysis of 22 giving studies where
cognitive processing was manipulated revealed the predicted in-
teraction between cognitive processing mode and gender: promot-
ing intuition increased altruism in women but had no significant
effect in men. Study 2 then investigated the mechanism behind this
effect by examining whether identification with gender norms
moderated the sex differences found in Study 1.

Study 1

Method

In Study 1, we conducted a meta-analysis (N � 4,366) of new
and existing studies looking at the effects of experimentally ma-
nipulating the use of intuition versus deliberation on giving in the
Dictator Game (DG). In the DG, participants unilaterally decide
how to divide actual money between themselves and an anony-
mous recipient. Across studies, we used the percentage of the
endowment given to the recipient as our measure of altruism.

To minimize file-drawer effects, we began by including all data
each of the authors of the present study had ever collected where
cognitive processing was manipulated in a zero-sum dictator game
(including failed pilots, experiments with problematic design fea-
tures, etc.). We had 13 such experiments (all previously unpub-
lished), each of which manipulated cognitive processing using
either time constraints or conceptual priming. To avoid selection
effects, we included participants who disobeyed the time con-
straints.

Reducing the amount of time subjects have to decide shortens
the window of opportunity for deliberation to rein in intuition,
leading to more intuitive decisions (Wright, 1974). Therefore, in
the time constraint experiments, reliance on intuition was in-
creased by asking subjects to make their giving decision in less
than a specified number of seconds (time pressure) and was
reduced by asking subjects to wait and think for at least a specified
number of seconds before deciding (time delay). The conceptual
priming conditions, on the other hand, used a writing exercise at
the outset of the experiment to induce more or less intuitive
decision making (Shenhav, Rand, & Greene, 2012). Reliance on
intuition was increased by asking subjects to write about a time in
their life where following their intuition worked out well, or where
carefully reasoning through a problem worked out poorly. Reli-
ance on intuition was decreased by asking subjects to write about

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

390 RAND, BRESCOLL, EVERETT, CAPRARO, AND BARCELO

a time in their life where following their intuition worked out
poorly, or where carefully reasoning through a problem worked
out well.

We also included data from other labs by doing a comprehen-
sive database search and requesting raw data (including gender)
for experiments in which cognitive processing was manipulated in
dictator games with the standard setup of (a) a single anonymous
recipient, and (b) a decision space ranging from completely selfish
to completely altruistic. For cognitive process manipulations, in
addition to time pressure and conceptual priming, we also included
experiments that used cognitive load (where participants had to
engage in a more or less cognitively demanding task, such as
holding a 7-digit vs. 3-digit number in working memory, while
completing the DG).1

In total, we received data sets for nine additional experiments
(Benjamin, Brown, & Shapiro, 2013; Cornelissen, Dewitte, & War-
lop, 2011; Evans, 2014; Hauge, Brekke, Johansson, Johansson-
Stenman, & Svedsäter, 2014; Kinnunen & Windmann, 2013). Thus,
Study 1 aggregated data from 22 experiments, for a total of 4,366
participants (52.7% female, Mage � 29.8). (Please see online supple-
mental material Table S1, which provides details for each experi-
ment).

Results and Discussion

As predicted, random-effects meta-analysis of DG giving found
a significant interaction between gender and cognitive processing
mode, effect size 5.5 percentage points, 95% CI [2.6, 8.5], Z �
3.66, p � .0001 (see Figure 1). There was no evidence of publi-
cation bias (Egger’s test, t � �.28, p � .79; Begg’s test, z � �.31,
p � .76), or of heterogeneity in the true size of this interaction
across studies, chi2(21) � 16.04, p � .77. Furthermore, metare-
gression found no significant difference in interaction effect be-
tween studies run online versus in the physical laboratory, t � .22,
p � .83 (interaction effect: 6.1 percentage points in lab, 5.4
percentage points online); and no significant differences in inter-
action effect size across methods of manipulating cognitive pro-
cessing, F(2, 19) � .4, p � .68 (interaction effect: cognitive load,
6.4 percentage points; conceptual priming, 6.1 percentage points
with; time constraints, 2.7 percentage points).

Examining simple effects showed a significant positive effect of
promoting intuition among women (see Figure 2), effect size 3.8
percentage points, 95% CI [1.9, 5.7], Z � 3.87, p � .0001;
resulting in on average 10.8% more giving in the high intuition
condition relative to the low intuition condition. Conversely, there
was no significant effect among men (see Figure 3), effect
size �2.0 percentage points, 95% CI [�4.2, .001], Z � 1.87, p �
.062. Again, there was no evidence of heterogeneity in effect size
across studies (women, chi2(21) � 13.1, p � .91; men, chi2(21) �
16.4, p � .75).

Study 2

Study 1 showed an interaction between gender and cognitive
processing mode: intuition favored unilaterally helping others for
women, but not for men. Although this effect was consistent with
our predictions based on the SHH and the differential daily-life
value of altruism for women versus men, the goal of Study 2 was
to provide more direct evidence for social norms as the proposed

mechanism. Specifically, if sex differences in the relationship
between intuition, deliberation, and altruism are driven by social
norms associated with men and women’s social roles, then indi-
vidual differences in the extent to which people adopt such sex role
norms should moderate this relationship. In particular, for women,
explicit self-identification with traditionally masculine attributes
was predicted to influence deliberative responses but not intuitive
responses.

Women who identify with masculine attributes are still per-
ceived by the world as women, and thus are subject to the altruistic
expectations placed upon women (making altruism typically ad-
vantageous for them). Since intuitive responses are not within
one’s conscious control, but instead implement typically advanta-
geous behaviors, the intuitive responses of even women who
explicitly adopt masculine attributes should be as altruistic as
women who explicitly identify with feminine attributes; and wom-
en’s intuitive responses should be more altruistic than men’s,
regardless of whether women identify more with masculine versus
feminine attributes.

When deliberating, however, explicit gender role identification was
predicted to influence women’s altruism: women who identified more
with masculine attributes were predicted to shift in the direction of
men (i.e., to become less altruistic), because altruism is disfavored by
both (a) masculine gender roles (which involve power, dominance,
and independent self-interest) and (b) deliberation’s general tendency
to make people consider strategic self-interest.

Women who explicitly identify with feminine attributes, on the
other hand, deliberatively embrace traits that are consistent with
altruism, and are in conflict with the self-interested effects of
strategic deliberation. Thus, deliberation should not affect their
behavior in the DG. For the same reason, men who explicitly
identify with feminine attributes should also not override their
selfish intuitions to become more altruistic when deliberating.
Finally, for men who identify with masculine attributes, their
intuitive and deliberative responses are in alignment, both favoring
relative selfishness, which again leads to no predicted effect of
deliberating. Study 2 directly tested these hypotheses.

Study 2 also explored a second potential moderator, how
strongly participants felt that gender norms were enforced in their
lives, which was unsuccessful for reasons that we believe were
practical, rather than theoretically informative, in nature (see on-
line supplemental material section 2 for details).

Method

Study 2 took advantage of the fact that three of our experiments
from Study 1 (K, L, and M), in which participants completed the
conceptual priming manipulation described in Study 1 and made a
single dictator game decision (total N � 1,831; 51.5% female,
Mage � 35.0 years), also included questions about self-
identification with male and female sex roles (the short-form of the

1 We did not include ego depletion, based both on evidence that ego
depletion may not function in the same way as other cognitive process
manipulations (Inzlicht & Schmeichel, 2012), and the fact that we were
only able to obtain data from one depletion study with a total of 54
participants (Halali, Bereby-Meyer, & Ockenfels, 2013); including these
data does not qualitatively alter our key results.

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

391INTUITION FAVORS ALTRUISM FOR WOMEN BUT NOT FOR MEN

Bem Sex-Role Inventory [BSRI; Bem, 1977]).2 Participants indi-
cated the extent to which each of 20 attributes (10 traditionally
masculine, 10 traditionally feminine) described them (from 1 �
never or almost never true to 7 � always or almost always true).3

We then classified participants’ gender role identification as mas-
culine versus feminine using a median split on the sum of all
masculine items minus the sum of all feminine items (Hoffman &
Borders, 2001).

Results and Discussion

Study 2 reproduced the pattern observed in the meta-analysis: an
analysis of variance (ANOVA) predicting DG giving based on cog-
nitive processing mode, gender, and study demonstrated an interac-
tion between cognitive processing mode and gender, F(1, 1819) �
3.85, p � .050, effect size 4.4 percentage points, such that dictator
giving was significantly greater among women when intuition was
promoted (M � .40, SD � .23) relative to deliberation (M � .37,
SD � .24), t(941) � 2.08, p � .038; but cognitive process did not
significantly affect giving among men (intuition: M � .32, SD � .27;
deliberation: M � .33, SD � .27), t(886) � .58, p � .55. No other
terms were significant (p � .15 for all), except for a significant main
effect of gender, F(1, 1819) � 29.5, p � .001, effect size 6.3
percentage points, such that women (M � .39, SD � .24) gave more
than men (M � .33, SD � .27). In particular, because the three-way

interaction between gender, cognitive processing mode, and study
was not significant, F(2, 1819) � .46, p � .63, we collapsed across
study in our subsequent analyses.

To test for moderation, we conducted an ANOVA predicting DG
giving using gender, cognitive processing mode, and sex role self-
identification (0 � feminine, 1 � masculine).4 In addition to signif-
icant main effects of gender F(1, 1823) � 18.53, p � .0001, such that
women were more altruistic than men, and sex role self-identification,
F(1, 1823) � 13.57, p � .0002, such that feminine participants were
more altruistic than masculine participants, we observed the predicted
significant three-way interaction between gender, cognitive process-

2 The order (i.e. whether the moderator questions came before or after
the conceptual priming task and DG) was randomized. Our analyses
collapsed over order, rather than analyzing the effect of order, because a
substantial difference in attrition rates between the orders prevented valid
causal inference about order effects.

3 Traditionally masculine attributes: willing to take a stand; defends own
beliefs; independent; has leadership abilities; strong personality; forceful;
dominant; aggressive; assertive; willing to take risks. Traditionally femi-
nine attributes: affectionate; warm; compassionate; gentle; tender; sympa-
thetic; sensitive to needs of others; soothe hurt feelings; understanding;
loves children.

4 Our results are qualitatively equivalent when using a continuous mea-
sure of sex role self-identification, but for ease of calculating and display-
ing simple effects of cognitive processing mode, we used the median split.

Overall

M

V

C

G

N

J

Q

R

K

D

P

Study

I

O

L

H

F

U

ID

A

B

S

E

T

5.52 (2.57, 8.48)

6.46 (-2.74, 15.65)

10.18 (-0.44, 20.79)

7.53 (-8.12, 23.19)

20.05 (2.97, 37.14)

3.60 (-16.92, 24.13)

-0.59 (-19.78, 18.59)

2.10 (-9.26, 13.45)

17.39 (1.62, 33.17)

1.73 (-5.89, 9.35)

6.23 (-13.96, 26.41)

-3.41 (-21.75, 14.93)

8.03 (-4.24, 20.29)

13.16 (-0.78, 27.10)

6.37 (-1.67, 14.41)

-6.31 (-25.29, 12.67)

18.54 (-1.88, 38.96)

-7.86 (-42.01, 26.29)

ES (95% CI)

-0.94 (-17.36, 15.48)

-5.50 (-21.35, 10.36)

6.65 (-15.00, 28.29)

3.07 (-16.14, 22.29)

0.51 (-34.41, 35.42)

100.00

10.34

7.76

3.57

2.99

2.07

2.38

6.78

3.51

15.05

2.15

2.60

%

5.81

4.50

13.53

2.43

2.10

0.75

W eight

3.24

3.48

1.87

2.37

0.72

5.52 (2.57, 8.48)

6.46 (-2.74, 15.65)

10.18 (-0.44, 20.79)

7.53 (-8.12, 23.19)

20.05 (2.97, 37.14)

3.60 (-16.92, 24.13)

-0.59 (-19.78, 18.59)

2.10 (-9.26, 13.45)

17.39 (1.62, 33.17)

1.73 (-5.89, 9.35)

6.23 (-13.96, 26.41)

-3.41 (-21.75, 14.93)

8.03 (-4.24, 20.29)

13.16 (-0.78, 27.10)

6.37 (-1.67, 14.41)

-6.31 (-25.29, 12.67)

18.54 (-1.88, 38.96)

-7.86 (-42.01, 26.29)

ES (95% CI)

-0.94 (-17.36, 15.48)

-5.50 (-21.35, 10.36)

6.65 (-15.00, 28.29)

3.07 (-16.14, 22.29)

0.51 (-34.41, 35.42)

100.00

10.34

7.76

3.57

2.99

2.07

2.38

6.78

3.51

15.05

2.15

2.60

%

5.81

4.50

13.53

2.43

2.10

0.75

W eight

3.24

3.48

1.87

2.37

0.72

0-30 -20 -10 0 10 20 30
[Cognitive process X gender] interaction effect size (percentage points)

Figure 1. Effect size (ES) for interaction between gender and cognitive processing mode for each experiment
in Study 1. See online supplemental materials Table S1 for key. Error bars indicate 95% confidence intervals
(CI). Gray squares indicate weight placed on each study by random-effects meta-analysis.

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

392 RAND, BRESCOLL, EVERETT, CAPRARO, AND BARCELO

ing mode, and sex role self-identification, F(1, 1823) � 5.23, p �
.022, effect size 11.1 percentage points.5 This three-way interaction
was such that there was a significant two-way interaction between
cognitive processing mode and self-identification among women,
F(1, 939) � 5.52, p � .019, effect size 7.4 percentage points, but not
among men, F(1, 884) � 1.00, p � .32, effect size �3.8 percentage
points. Decomposing the significant two-way interaction among
women, we found a significant negative effect of promoting deliber-
ation for women who self-identified as masculine, t(371) � 3.19, p �
.002, effect size 7.8 percentage points, and no significant effect of
promoting deliberation for women who self-identified as feminine,
t(570) � .19, p � .85, effect size .00 percentage points. As can be
seen in Figure 4, men gave a comparatively smaller amount regardless
of cognitive processing mode or identification with masculine versus
feminine attributes, whereas women gave a larger amount unless they
both identified as masculine and were induced to deliberate. We also
note that these results were not driven entirely by the feminine sex
role items, many of which were directly related to prosociality/
altruism: a median split on just the 10 masculine items also reveals a
significant negative effect of deliberation among more masculine
women, t(429) � 2.35, p � .019, effect size 5.4 percentage points,
and no significant effect of deliberation among less masculine
women, t(514) � .67, p � .50, effect size 1.4 percentage points.

Thus, Study 2 found that although intuition favored altruism
among women, those women who explicitly saw themselves as

occupying traditionally masculine sex roles became more selfish
when deliberating. Conversely, men’s comparatively selfish intu-
itive response was unaffected by deliberation.

General Discussion

What roles do intuition and deliberation play in altruism? Here
we have explored this question using economic games and found
that the answer depends on who you are. Study 1 meta-analyzed 22
DG studies and revealed an interaction whereby promoting intu-
ition relative to deliberation made women significantly more likely
to give, but had no significant effect on giving among men. Study
2 then demonstrated moderation by sex role identification, such
that deliberation specifically undermined the altruistic intuitions of
women who saw themselves as masculine.

Our results tie together two distinct lines of theory: one regard-
ing gender differences in altruism, and another regarding social
heuristics and the basis of intuition. Women disproportionately
occupy social roles that require communal and even self-
sacrificing behavior: thus, failing to behave communally results in
negative consequences for women more so than men. The SHH

5 Note that this result is robust to Bonferroni correcting for also having
tested a second moderator (described in the supplemental materials), given
that p � .025.

Overall

C

B

Study

I

D

P

V

L

J

F

ID

K
T

H

G

M
S

O
Q

U

E

R

A

N
3.77 (1.86, 5.68)

8.48 (-4.20, 21.17)

1.74 (-8.50, 11.97)

3.00 (-4.49, 10.48)

9.09 (-6.76, 24.94)

-0.68 (-11.60, 10.24)

6.16 (0.59, 11.72)

4.57 (-0.78, 9.93)

-1.98 (-17.26, 13.30)

9.05 (-7.93, 26.02)

ES (95% CI)

1.77 (-3.68, 7.22)
2.78 (-24.37, 29.92)

-9.15 (-24.14, 5.85)

8.53 (-2.34, 19.39)

3.82 (-2.52, 10.16)
3.76 (-8.25, 15.77)

5.89 (-2.76, 14.54)
5.43 (-0.34, 11.19)

-15.68 (-43.14, 11.78)

-2.35 (-14.14, 9.44)

2.85 (-7.02, 12.72)

-2.22 (-15.44, 11.00)

10.67 (-2.19, 23.53)
100.00

2.26

3.48

%

6.50

1.45

3.05

11.76

12.70

1.56

1.26

W eight

12.25
0.49

1.62

3.09

9.05
2.52

4.87
10.96

0.48

2.62

3.74

2.08

2.20
3.77 (1.86, 5.68)

8.48 (-4.20, 21.17)

1.74 (-8.50, 11.97)

3.00 (-4.49, 10.48)

9.09 (-6.76, 24.94)

-0.68 (-11.60, 10.24)

6.16 (0.59, 11.72)

4.57 (-0.78, 9.93)

-1.98 (-17.26, 13.30)

9.05 (-7.93, 26.02)

ES (95% CI)

1.77 (-3.68, 7.22)
2.78 (-24.37, 29.92)

-9.15 (-24.14, 5.85)

8.53 (-2.34, 19.39)

3.82 (-2.52, 10.16)
3.76 (-8.25, 15.77)

5.89 (-2.76, 14.54)
5.43 (-0.34, 11.19)

-15.68 (-43.14, 11.78)

-2.35 (-14.14, 9.44)

2.85 (-7.02, 12.72)

-2.22 (-15.44, 11.00)

10.67 (-2.19, 23.53)
100.00

2.26

3.48

%

6.50

1.45

3.05

11.76

12.70

1.56

1.26

W eight

12.25
0.49

1.62

3.09

9.05
2.52

4.87
10.96

0.48

2.62

3.74

2.08

2.20

0-20 -10 0 10 20
Cognitive process effect size (percentage points)

Figure 2. Effect size (ES) for simple effect of promoting intuition among women for each experiment in Study
1. See online supplemental materials Table S1 for key. Error bars indicate 95% confidence intervals (CI). Gray
squares indicate weight placed on each study by random-effects meta-analysis.

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

393INTUITION FAVORS ALTRUISM FOR WOMEN BUT NOT FOR MEN

therefore predicts that men and women would develop different
intuitions regarding altruism. And indeed, this is what our data
show. Our results therefore support a core tenant of the SHH—that
intuitive responses implement social behaviors that are typically
advantageous.

In contrast to the current work on altruism, the SHH predicts
that cooperation will be intuitive regardless of one’s gender:
cooperating inherently has the possibility to be long-run payoff-

maximizing because it is non-zero-sum, and thus does not rely on
expectations related to social roles. Consistent with this prediction,
a follow-up paper inspired by the current studies found no gender
moderation of the relationship between intuition and cooperation,
which was positive for both women and men (Rand, 2016).

The fact that deliberation only works against altruism in our
data, rather than sometimes making men more altruistic by over-
riding their selfish intuitions, is also consistent with the SHH. The
SHH posits that a key component of deliberation is the consider-
ation of what choice is payoff maximizing, which is always self-
ishness in our experiments (because they involve one-shot anon-
ymous interactions)—and such payoff-maximizing considerations
work against any deliberative motivations to give (such as having
a conscious desire to be communal).

Although we explicitly rely on social norms in our theorizing
about the gender difference in intuitive altruism, the ultimate
origins of the distribution of men and women into different social
roles could be biological in nature (Preston, 2013). Specifically,
women’s capacity for reproduction and men’s greater physical size
and strength (Wood & Eagly, 2002) along with the evolutionary
advantages to women of occupying roles that require a longer-term
investment in caring for offspring (Buss, 1995) may explain why
women end up occupying roles that require communal and self-
sacrificing behavior in the first place and thus why altruism may
become the intuitive social response for women.

Overall

I

L

P
T

H

F

ID

G

N

K

M

Q

O

Study

S
V

D

A

C

B

E

J

R

U

-2.04 (-4.18, 0.10)

-5.03 (-14.74, 4.68)

-1.80 (-7.79, 4.20)

2.73 (-12.01, 17.46)
2.27 (-19.69, 24.24)

-2.84 (-14.47, 8.80)

-9.49 (-20.84, 1.85)

ES (95% CI)

-11.52 (-24.71, 1.66)

7.07 (-8.93, 23.07)

0.04 (-5.29, 5.36)

-2.64 (-9.30, 4.02)

3.33 (-6.46, 13.11)

-7.27 (-18.20, 3.66)

-2.88 (-20.89, 15.12)
-4.02 (-13.05, 5.02)

2.87 (-9.63, 15.36)

-1.28 (-11.02, 8.45)

0.95 (-8.23, 10.13)

7.24 (-4.87, 19.34)

-5.42 (-20.59, 9.75)

-1.38 (-12.98, 10.22)

-14.55 (-26.85, -2.24)

-7.82 (-28.12, 12.48)

100.00

4.86

12.74

2.11
0.95

3.38

3.56

W eight

2.63

1.79

16.15

10.33

4.78

3.83

%

1.41
5.61

2.93

4.83

5.44

3.13

1.99

3.40

3.02

1.11

-2.04 (-4.18, 0.10)

-5.03 (-14.74, 4.68)

-1.80 (-7.79, 4.20)

2.73 (-12.01, 17.46)
2.27 (-19.69, 24.24)

-2.84 (-14.47, 8.80)

-9.49 (-20.84, 1.85)

ES (95% CI)

-11.52 (-24.71, 1.66)

7.07 (-8.93, 23.07)

0.04 (-5.29, 5.36)

-2.64 (-9.30, 4.02)

3.33 (-6.46, 13.11)

-7.27 (-18.20, 3.66)

-2.88 (-20.89, 15.12)
-4.02 (-13.05, 5.02)

2.87 (-9.63, 15.36)

-1.28 (-11.02, 8.45)

0.95 (-8.23, 10.13)

7.24 (-4.87, 19.34)

-5.42 (-20.59, 9.75)

-1.38 (-12.98, 10.22)

-14.55 (-26.85, -2.24)

-7.82 (-28.12, 12.48)

100.00

4.86

12.74

2.11
0.95

3.38

3.56

W eight

2.63

1.79

16.15

10.33

4.78

3.83

%

1.41
5.61

2.93

4.83

5.44

3.13

1.99

3.40

3.02

1.11

0-20 -10 0 10 20
Cognitive process effect size (percentage points)

Figure 3. Effect size (ES) for simple effect of promoting intuition among men for each experiment in Study
1. See online supplemental materials Table S1 for key. Error bars indicate 95% confidence intervals (CI). Gray
squares indicate weight placed on each study by random-effects meta-analysis.

20%

25%

30%

35%

40%

45%

Masculine Feminine Masculine Feminine

D
G

T
ra

ns
fe

r

Sex Role Identification

Women Men

Intuition
Deliberation

**

Figure 4. Average dictator game (DG) giving in Study 2 by gender,
cognitive processing mode, and sex role self-identification. Error bars
indicate 95% confidence intervals. �� p � .01.

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

394 RAND, BRESCOLL, EVERETT, CAPRARO, AND BARCELO

The size of the effects we observed was determined not only by the
actual magnitude of the influence that intuition and deliberation have on
altruism, but also by the effectiveness of the cognitive processing manip-
ulations used. These manipulations were at best only partially successful
in making participants rely more heavily on intuition versus deliber-
ation—those in the “intuition-promoting” conditions were un-
doubtedly still able to engage in substantial degrees of delibera-
tion, despite the experimental manipulation. As a result, our
overall estimates of the observed effects are likely underestimates
of the effect size of intuitive processing on altruism one might
observe with more powerful manipulations, or in more strongly
valenced real-world interactions. Thus, we argue that the size of
the observed effects is less important that their direction.

Relatedly, it is unclear the extent to which these manipulations
acted by reducing deliberation versus amplifying intuition. Thus,
future research should use different manipulations to better under-
stand the nature of participants’ baseline responses, as well as
including baseline responses (i.e., no manipulation). Furthermore,
future research should examine whether the gender differences in
intuitive altruism we observed in the context of interpersonal
giving extend to other forms of altruistic behavior, such as chari-
table giving (Small, Loewenstein, & Slovic, 2007) and “extreme”
altruism (Rand & Epstein, 2014), and to intergroup contexts (given
evidence that tribal instincts for parochial altruism are stronger in
men; Van Vugt, De Cremer, & Janssen, 2007).

In sum, we provide evidence that promoting intuition relative to
deliberation increases altruistic giving in women but not men.
These findings extend our understanding of gender and prosocial-
ity, and advance a model of intuitive decision-making based on
social heuristics.

References

Bear, A., & Rand, D. G. (2016). Intuition, deliberation, and the evolution
of cooperation. Proceedings of the United States of America National
Academy of Sciences, 113, 936 –941. http://dx.doi.org/10.1073/pnas
.1517780113

Bem, S. L. (1977). On the utility of alternative procedures for assessing
psychological androgyny. Journal of Consulting and Clinical Psychol-
ogy, 45, 196 –205. http://dx.doi.org/10.1037/0022-006X.45.2.196

Benjamin, D. J., Brown, S. A., & Shapiro, J. M. (2013). Who is “behav-
ioral”? Cognitive ability and anomalous preferences. Journal of the
European Economic Association, 11, 1231–1255. http://dx.doi.org/10
.1111/jeea.12055

Brescoll, V. L. (2011). Who takes the floor and why: Gender, power, and
volubility in organizations. Administrative Science Quarterly, 56, 622–
641. http://dx.doi.org/10.1177/0001839212439994

Buss, D. M. (1995). Psychological sex differences. Origins through sexual
selection. American Psychologist, 50, 164 –168. http://dx.doi.org/10
.1037/0003-066X.50.3.164

Chudek, M., & Henrich, J. (2011). Culture-gene coevolution, norm-
psychology and the emergence of human prosociality. Trends in Cog-
nitive Sciences, 15, 218 –226. http://dx.doi.org/10.1016/j.tics.2011.03
.003

Cone, J., & Rand, D. G. (2014). Time pressure increases cooperation in
competitively framed social dilemmas. PLoS ONE, 9, e115756. http://
dx.doi.org/10.1371/journal.pone.0115756

Cornelissen, G., Dewitte, S., & Warlop, L. (2011). Are social value
orientations expressed automatically? Decision making in the dictator
game. Personality and Social Psychology Bulletin, 37, 1080 –1090.
http://dx.doi.org/10.1177/0146167211405996

Eagly, A. H. (1987). Sex differences in social behavior: A social-role
interpretation. Mahwah, NJ: Erlbaum.

Evans, A. M. (2014). Intuition in the Dictator Game: Identifying social and
selfish heuristics. Manuscript in preparation.

Halali, E., Bereby-Meyer, Y., & Ockenfels, A. (2013). Is it all about the
self? The effect of self-control depletion on ultimatum game proposers.
Frontiers in Human Neuroscience, 7, 240. http://dx.doi.org/10.3389/
fnhum.2013.00240

Hauge, K. E., Brekke, K. A., Johansson, L.-O., Johansson-Stenman, O., &
Svedsäter, H. (2014). Keeping others in our mind or in our heart?
Distribution games under cognitive load. University of Gothenburg
Working Papers in Economics, (600).

Heilman, M. E., & Chen, J. J. (2005). Same behavior, different conse-
quences: Reactions to men’s and women’s altruistic citizenship behav-
ior. Journal of Applied Psychology, 90, 431– 441. http://dx.doi.org/10
.1037/0021-9010.90.3.431

Heilman, M. E., & Okimoto, T. G. (2007). Why are women penalized for
success at male tasks?: The implied communality deficit. Journal of
Applied Psychology, 92, 81–92. http://dx.doi.org/10.1037/0021-9010.92
.1.81

Hoffman, R. M., & Borders, L. D. (2001). Twenty-five years after the Bem
Sex-Role Inventory: A reassessment and new issues regarding classifi-
cation variability. Measurement and Evaluation in Counseling and De-
velopment, 34, 39 –55.

Inzlicht, M., & Schmeichel, B. J. (2012). What is ego depletion? Toward
a mechanistic revision of the resource model of self-control. Perspec-
tives on Psychological Science, 7, 450 – 463. http://dx.doi.org/10.1177/
1745691612454134

Kahneman, D. (2003). A perspective on judgment and choice: Mapping
bounded rationality. American Psychologist, 58, 697–720. http://dx.doi
.org/10.1037/0003-066X.58.9.697

Kinnunen, S. P., & Windmann, S. (2013). Dual-processing altruism. Fron-
tiers in Psychology, 4, 193. http://dx.doi.org/10.3389/fpsyg.2013.00193

Kiyonari, T., Tanida, S., & Yamagishi, T. (2000). Social exchange and
reciprocity: Confusion or a heuristic? Evolution and Human Behavior,
21, 411– 427. http://dx.doi.org/10.1016/S1090-5138(00)00055-6

Lotz, S. (2015). Spontaneous giving under structural inequality: Intuition
promotes cooperation in asymmetric social dilemmas. PLoS ONE, 10,
e0131562. http://dx.doi.org/10.1371/journal.pone.0131562

Maccoby, E. E. (1998). The two sexes. Cambridge, MA: Harvard Univer-
sity Press.

Peysakhovich, A., & Rand, D. G. (2015). Habits of virtue: Creating norms
of cooperation and defection in the laboratory. Management Science.
Advance online publication. http://dx.doi.org/10.1287/mnsc.2015.2168

Preston, S. D. (2013). The origins of altruism in offspring care. Psycho-
logical Bulletin, 139, 1305–1341. http://dx.doi.org/10.1037/a0031755

Protzko, J., Ouimette, B., & Schooler, J. (2015). Believing there is no free
will corrupts intuitive cooperation. Retrieved from SSRN 2490855.

Rand, D. G. (2016). Cooperation (unlike altruism) is intuitive for men as
well as women. Retrieved from SSRN 2722981. http://ssrn.com/
abstract�2722981

Rand, D. G., & Epstein, Z. G. (2014). Risking your life without a second
thought: Intuitive decision-making and extreme altruism. PLoS ONE, 9,
e109687. http://dx.doi.org/10.1371/journal.pone.0109687

Rand, D. G., Greene, J. D., & Nowak, M. A. (2012). Spontaneous giving
and calculated greed. Nature, 489, 427– 430. http://dx.doi.org/10.1038/
nature11467

Rand, D. G., Newman, G. E., & Wurzbacher, O. (2015). Social context and
the dynamics of cooperative choice. Journal of Behavioral Decision
Making, 28, 159 –166. http://dx.doi.org/10.1002/bdm.1837

Rand, D. G., & Nowak, M. A. (2013). Human cooperation. Trends in
Cognitive Sciences, 17, 413– 425. http://dx.doi.org/10.1016/j.tics.2013
.06.003

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

395INTUITION FAVORS ALTRUISM FOR WOMEN BUT NOT FOR MEN

Rand, D. G., Peysakhovich, A., Kraft-Todd, G. T., Newman, G. E.,
Wurzbacher, O., Nowak, M. A., & Greene, J. D. (2014). Social heuris-
tics shape intuitive cooperation. Nature Communications, 5, 3677. http://
dx.doi.org/10.1038/ncomms4677

Shenhav, A., Rand, D. G., & Greene, J. D. (2012). Divine intuition:
Cognitive style influences belief in God. Journal of Experimental Psy-
chology: General, 141, 423– 428. http://dx.doi.org/10.1037/a0025391

Sloman, S. A. (1996). The empirical case for two systems of reasoning.
Psychological Bulletin, 119, 3–22. http://dx.doi.org/10.1037/0033-2909
.119.1.3

Small, D. A., Loewenstein, G., & Slovic, P. (2007). Sympathy and cal-
lousness: The impact of deliberative thought on donations to identifiable
and statistical victims. Organizational Behavior and Human Decision
Processes, 102, 143–153. http://dx.doi.org/10.1016/j.obhdp.2006.01.005

Van Lange, P. A., De Bruin, E. M., Otten, W., & Joireman, J. A. (1997).
Development of prosocial, individualistic, and competitive orientations:
Theory and preliminary evidence. Journal of Personality and Social
Psychology, 73, 733–746. http://dx.doi.org/10.1037/0022-3514.73.4.733

Van Vugt, M., De Cremer, D., & Janssen, D. P. (2007). Gender differences
in cooperation and competition: The male-warrior hypothesis. Psycho-
logical Science, 18, 19 –23. http://dx.doi.org/10.1111/j.1467-9280.2007
.01842.x

Wood, W., & Eagly, A. H. (2002). A cross-cultural analysis of the behavior
of women and men: Implications for the origins of sex differences.
Psychological Bulletin, 128, 699 –727. http://dx.doi.org/10.1037/0033-
2909.128.5.699

Wright, P. (1974). The harassed decision maker: Time pressures, distrac-
tions, and the use of evidence. Journal of Applied Psychology, 59,
555–561. http://dx.doi.org/10.1037/h0037186

Zaki, J., & Mitchell, J. P. (2013). Intuitive prosociality. Current Directions
in Psychological Science, 22, 466 – 470. http://dx.doi.org/10.1177/
0963721413492764

Received June 19, 2015
Revision received January 18, 2016

Accepted January 20, 2016 �

Psychological Bulletin Call for Papers: Replication and Reproducibility:
Questions Asked and Answered via Research Synthesis

The primary mission of Psychological Bulletin is to contribute a cohesive, authoritative, theory-
based, and complete synthesis of scientific evidence in the field of psychology.

The editorial team is currently interested in the contribution of meta-analysis and systematic reviews
to answering questions about the replication and reproducibility of psychological findings. This call
is designed to encourage authors to consider the degree to which multi-study primary research
papers, systematic replications efforts (e.g., Open Science Framework and Many Labs Replica-
tions), and meta-analysis can contribute to substantial psychological knowledge.

The editors of Psychological Bulletin announce interest in publishing Replication and Reproduc-
ibility: Questions Asked and Answered via Research Synthesis. Papers may be published as a
special issue or as a series of special sections. Manuscripts will be evaluated and may be published
as they come in but must be submitted by June 30, 2017 for consideration for the special
sections/issue.

We invite authors to submit single papers that address a clearly defined question related to
replication and reproducibility of psychological findings using systematic-review or meta-analytic
perspectives. Purely methodological papers, such as tests or criticisms of particular approaches, or
commentaries on replication and reproducibility, are not appropriate for Psychological Bulletin.
Primary-level research reports are also outside of the interest domain of Psychological Bulletin.
Examples of possible manuscripts include but are not limited to meta-analyzing replication efforts,
comparing evidence from different replication methods, synthesizing models of replication deter-
minants, meta-analyzing predictors of questionable research practices and retraction, and comparing
methods to gauge publication bias.

For guidance in proposing a submission, interested authors may contact the Editor, Dolores
Albarracín ([email protected]), or Associate Editor, Blair T. Johnson ([email protected]
uconn.edu). Alternatively, authors are free to submit unsolicited manuscripts and should indicate
that the submission targets this call in the cover letter.

T
hi

s
do

cu
m

en
t

is
co

py
ri

gh
te

d
by

th
e

A
m

er
ic

an
P

sy
ch

ol
og

ic
al

A
ss

oc
ia

ti
on

or
on

e
of

it
s

al
li

ed
pu

bl
is

he
rs

.
T

hi
s

ar
ti

cl
e

is
in

te
nd

ed
so

le
ly

fo
r

th
e

pe
rs

on
al

us
e

of
th

e
in

di
vi

du
al

us
er

an
d

is
no

t
to

be
di

ss
em

in
at

ed
br

oa
dl

y.

396 RAND, BRESCOLL, EVERETT, CAPRARO, AND BARCELO

The Journal of Social Psychology, 155: 559–575, 2015
Published with license by Taylor & Francis
ISSN: 0022-4545 print / 1940-1183 online
DOI: 10.1080/00224545.2015.1084985

INVITED ARTICLE

Why Seemingly Trivial Events Sometimes Evoke Strong
Emotional Reactions: The Role of Social Exchange

Rule Violations

MARK R. LEARY
KATE J. DIEBELS

KATRINA P. JONGMAN-SERENO
XUAN DUONG FERNANDEZ

Duke University

ABSTRACT. People sometimes display strong emotional reactions to events that appear dispropor-
tionate to the tangible magnitude of the event. Although previous work has addressed the role that
perceived disrespect and unfairness have on such reactions, this study examined the role of perceived
social exchange rule violations more broadly. Participants (N = 179) rated the effects of another per-
son’s behavior on important personal outcomes, the degree to which the other person had violated
fundamental rules of social exchange, and their reactions to the event. Results showed that percep-
tions of social exchange rule violations accounted for more variance in participants’ reactions than
the tangible consequences of the event. The findings support the hypothesis that responses that appear
disproportionate to the seriousness of the eliciting event are often fueled by perceived rule violations
that may not be obvious to others.

Keywords: anger, emotional intensity, interpersonal interactions, social exchange rules

EMOTION THEORISTS CONCUR THAT EMOTIONS are fundamentally functional, having
evolved to facilitate animals’ effectiveness in dealing with potential threats and opportunities.
Yet, people sometimes display very strong emotional reactions to events that, from an outsider’s

© Mark R. Leary, Kate J. Diebels, Katrina P. Jongman-Sereno, and Xuan Duong Fernandez
This is an Open Access article. Non-commercial re-use, distribution, and reproduction in any medium, provided the

original work is properly attributed, cited, and is not altered, transformed, or built upon in any way, is permitted. The
moral rights of the named authors have been asserted.

Address correspondence to Mark R. Leary, Duke University, Department of Psychology and Neuroscience, P.O.
Box 90085, Durham, NC 27708, USA. E-mail: [email protected]

560 THE JOURNAL OF SOCIAL PSYCHOLOGY

perspective, far exceed the response necessary to deal with the situation, if indeed any response
is needed at all. In many instances, seemingly trivial events provoke strong, sometimes explosive
reactions that seem disproportionate to the seriousness of the precipitating event.

For example, people become extremely angry during “friendly” discussions in which the out-
come of the debate has no tangible consequences, lash out at those who question or criticize
them in inconsequential ways, become enraged in response to another person’s mildly inconsid-
erate behavior, and overreact to other people’s annoying but immaterial eccentricities. People
often react strongly to such behaviors in ways that not only create awkward encounters and
conflicts (Cunningham, Shamblen, Barbee, & Ault, 2005) but also fuel domestic violence and
child abuse (Mann, 1988); righteous indignation and moral outrage (Tannenbaum, Uhlmann, &
Diermeier, 2011); culture-of-honor violence (Cohen, Nisbett, Bowdle, & Schwarz, 1996); road
rage (Neighbors, Vietor, & Knee, 2002); crimes of passion (Scarpa & Raine, 2000); and extreme
collective reactions to seemingly inconsequential slights against one’s national, ethnic, or reli-
gious group (Sageman, 2008). The present article examines one set of factors that lead people to
experience stronger emotions than seem warranted by the seriousness of the precipitating event.

Of course, the line between proportionate, normative reactions and disproportionate, extreme
reactions is not always clear. Indeed, assessing whether a reaction is reasonable and propor-
tionate given the nature of a particular precipitating event is often impossible because people’s
emotions are influenced not only by the immediate situation but also by events unknown to the
observer, real and imagined implications of the event that are not immediately obvious to others,
and the degree to which the event portends future circumstances that are relevant to one’s con-
cerns (Frijda, 1986). At the extremes, disproportionate responses are perhaps easy to spot (e.g.,
physically attacking a driver who delayed for five seconds after a red light turned green or burning
down one’s boss’ house after an unfavorable performance review), but often they are not.

Thus, our interest was in parsing the factors that influence the strength of people’s emotional
reactions to examine those that lead people to respond more strongly than their own assessment
of the event’s tangible outcomes would suggest. By tangible, we mean outcomes that would be
expected to inherently affect a person’s well-being or quality of life. Tangible outcomes include
not only physical and material outcomes (such as outcomes involving physical harm, money, or
possessions) but also events that inherently affect hedonic states (such as engaging in pleasurable
activities, being able to do something that one wants to do, or losing an important relationship).

As we use the term, tangible does not include symbolic or signifying outcomes that, by them-
selves, do not influence well-being. Of course, even when other people’s actions have no tangible
consequences at the moment, they may signal the possibility that undesired tangible outcomes
may occur in the future. For example, people may be upset when treated unfairly even when
the unfair treatment does not tangibly affect their outcomes (Allen & Leary, 2010) because they
are trying to deter future mistreatment that might, in fact, disadvantage them. Others have made
this point in the context of reactions to unfairness and disrespect (e.g., Cohen et al., 1996; Lind
& Tyler, 1988), but we offer a broader perspective on why people sometimes react strongly to
events that have no tangible implications for them.

Two explanations of reactions that exceed the seriousness of the immediate precipitating
event—triggered displaced aggression (Marcus-Newhall, Pedersen, Carlson, & Miller, 2000;
Pedersen, Gonzales, & Miller, 2000) and cumulative stress (e.g., the straw that broke the camel’s
back)—involve general effects of frustration, arousal, or stress that lower the threshold at which
people respond to negative events. Although these effects clearly increase the likelihood that

LEARY ET AL. 561

people will respond to minor provocations, our focus is specifically on the features of events that
evoke strong emotional reactions. What are the characteristics of situations that “push people’s
buttons” even when nothing tangible is at stake?

Four general explanations of such reactions have been offered that involve unfairness, dis-
respect, loss of self-esteem, and rejection. First, people become angry when they perceive that
others have treated them unfairly (Bembenek, Beike, & Schroeder, 2007; Brebels, De Cremer,
& Sedikides, 2008). People sometimes react to unfairness even when it does not matter (Allen
& Leary, 2010), and perceptions of procedural justice and fairness are enhanced when people
have an opportunity to voice their views even when their input cannot affect the outcome (Lind,
Kanfer, & Earley, 1990; Tyler, 1987; Tyler, Rasinski, & Spodick, 1985). However, in most studies
of reactions to unfairness, participants believed that others’ unfair behaviors had tangible implica-
tions for their well-being, either at present or in the future, so previous work has not distinguished
the effects of tangible versus intangible outcomes. Furthermore, as we will see, the phenomenon
seems to encompass a broader set of undesired behaviors than simply unfairness.

A second explanation asserts that disproportionate reactions are sometimes provoked by signs
of disrespect (Cohen et al., 1996; Miller, 2001; Stephenson, Martsolf, & Draucker, 2011). People
feel entitled to respectful treatment (Bies & Moag, 1986; Miller, 2001) and sometimes inter-
pret violations of interpersonal codes of conduct or “psychological contracts” (Robinson, Kraatz,
& Rousseau, 1994; Rousseau, 1995) as disrespect even when the violations are inconsequen-
tial (Cropanzano & Byrne, 2000; Folger & Skarlicki, 1998). Cohen et al. (1996, Cohen, 1998)
suggested that people who live in a “culture of honor,” such as gang members and men in the
American south, are particularly sensitive to signs of disrespect, but the general effect of disre-
spect on anger and aggression is widespread, if not universal (Bettencourt & Miller, 1996; Miller,
2001; Scher, 1997; Vidmar, 2000). As with unfairness, disrespectful behaviors are sometimes
associated with objectively harmful outcomes, but often the person’s reaction is fueled by the dis-
respectful action itself rather than any associated costs (Lind, Kulik, Ambrose, & de Vera-Park,
1993).

A third explanation considers the role of diminished self-esteem. When people hold favorable
views of themselves—and particularly when their views are inflated, unstable, or uncertain—
negative interpersonal evaluations can lead to negative emotions and aggression toward people
who threaten those views (Baumeister, Smart, & Boden, 1996). The theory suggests that anger
and aggression protect people’s self-evaluation (Fast & Chen, 2009; Kernis, Granneman, &
Barclay, 1989), although the process by which becoming excessively angry might protect
self-esteem has not been adequately explained.

Finally, interpersonal rejection can lead to strong emotional reactions (Buckley, Winkel, &
Leary, 2004; Leary, Twenge, & Quinlivan, 2006; Twenge & Campbell, 2003). Many school
shootings were perpetrated by students who felt ostracized by their classmates (Leary &
Jongman-Sereno, 2015; Leary, Kowalski, Smith, & Phillips, 2003), and rejection is a lead-
ing cause of domestic violence and spousal homicide (Barnard, Vera, Vera, & Newman, 1982;
Brown, James, & Taylor, 2010). In many cases, the tangible consequences of a particular rejection
experience may be trivial, yet people react strongly (Gonsalkorale & Williams, 2007).

Each of these explanations has merit. Yet, even assuming that unfairness, disrespect, threat-
ened egotism, and rejection fuel emotional reactions that appear disproportionate to the precipi-
tating event, an integrative explanation is needed. The hypothesis guiding the present research is
that people react disproportionally to the tangible implications of interpersonal events when they

562 THE JOURNAL OF SOCIAL PSYCHOLOGY

perceive that other people have violated the rules that underlie social exchange. Because these rule
violations can evoke strong emotional reactions even when they have no tangible consequences,
people’s reactions often appear stronger than the seriousness of the tangible outcomes seems to
warrant.

Many theories from a variety of perspectives have suggested that interdependent interactions
and relationships carry fundamental rules regarding how the individuals are expected to treat
one another (Emerson, 1976; Gouldner, 1960; Hall, 2011; Homans, 1961). Interdependent inter-
actions are mutually beneficial and rewarding only when all individuals abide by these rules
of social exchange. A number of such rules have been proposed. For example, theorists have
suggested the existence of social exchange rules involving reciprocity, dependability, honesty,
fairness, cooperation, rationality, and some minimal level of concern for others’ needs, among
others (Baron, 1993; Bies & Moag, 1986; Blau, 1960; Buss et al., 1990; Cosmides & Tooby,
1992; Cottrell, Neuberg, & Li, 2007; Cropanzano & Mitchell, 2005; Robinson, 1996; Taylor,
Tracy, Renard, Harrison, & Carroll, 1995). Reactions to violations of some such rules—such as
reciprocity, honesty, and fairness (or justice)—have been studied extensively (e.g., Cropanzano
& Byrne, 2000; Eisenberger, Lynch, Aselage, & Rohdieck, 2004; Gouldner, 1960; Miller, 2001;
Skitka & Crosby, 2003), whereas others have received only passing mention.

People are highly sensitive to violations of these social exchange rules because such violations
signal that the perpetrator is a poor partner for social exchange whose behavior has the potential
to disadvantage the individual, either in the present encounter or in the future (Cosmides & Tooby,
1992). According to the social exchange rule violation hypothesis, social exchange rules are so
important that violations of these rules can provoke strong emotional reactions even when the
other person’s actions have few, if any, tangible consequences. Even violations with no tangi-
ble consequences whatsoever provide diagnostic information that another person should not be
trusted as a social exchange partner. Thus, the social exchange rule violation hypothesis explains
why people’s reactions often seem disproportionate to the tangible consequences of the eliciting
event.

Because trust is a central feature of interdependent group living and necessary for cooperative
relationships and collective action, people highly value trustworthiness (Buss et al., 1990; Cottrell
et al., 2007) and react strongly when others behave in ways that undermine trust. The enforcement
of social exchange rules is the mechanism by which people insist on being treated appropriately
and sanction untrustworthy relational partners. In fact, evolutionary psychologists have suggested
that social exchange rules are so important to well-being that human beings have evolved cog-
nitive mechanisms to detect, at fairly low thresholds, violations of certain social exchange rules
(Cosmides & Tooby, 1992, 2008). The greatest attention has been devoted to a “cheater detection”
system that appears to monitor instances in which other people take benefits to which they are
not entitled (Cosmides, 1989; Delton, Cosmides, Guemo, Robertson, & Tooby, 2012). Although
the existence of an evolved cheater detection system has been controversial (Cosmides & Tooby,
2008; Fodor, 2000), people are certainly quite sensitive to violations of social exchange rules.

One function of strong reactions may be to put those who violate fundamental rules of social
exchange on notice to deter future violations. This explanation has been discussed previously with
respect to aggressive reactions to unfairness and disrespect (e.g., Cohen et al., 1996; Felson, 1978;
Miller, 1993; Toch, 1992), but the reaction seems broader, encompassing violations of social
exchange rules more generally. In fact, from the standpoint of the social exchange rule violation
hypothesis, some of the other explanations described earlier (unfairness, disrespect, threatened

LEARY ET AL. 563

egotism, and interpersonal rejection) might be interpreted as specific cases of a general tendency
to react to violations of social exchange rules. In fact, we suspect that these specific cases evoke
reactions that are disproportionate to the event’s tangible consequences primarily when they are
interpreted as an exchange rule violation. The social exchange rule violation hypothesis provides
a parsimonious, overarching perspective that integrates findings across a number of disparate
domains and offers insights into the conditions under which people react to seemingly trivial
events.

To summarize, people’s emotional reactions to undesired events are affected by two sets of
factors. The most obvious involves tangible outcomes of the events—outcomes that directly and
incontrovertibly influence their well-being or quality of life. The second set of factors involves
the perceived violation of social exchange rules, which, as noted, may influence emotion even
in the absence of tangible consequences. Such violations may portend negative outcomes in the
future but their present impact is often minimal. Although previous theorists have suggested that
both sets of factors influence emotional reactions, we are not aware of any research that has
systematically examined their separate effects. Nor has previous research that has examined the
effects of specific violations—such as unfairness or disrespect—done so in a way that accounts
for perceived tangible outcomes.

This study tested the hypothesis that the strength of people’s emotional reactions to unde-
sired events can be predicted by both their judgment of the event’s tangible outcomes and the
degree to which they perceive that others have violated social exchange rules. We predicted that
perceived violations of exchange rules are associated with strong reactions, particularly anger,
above and beyond the effects of the tangible outcomes or implications of the precipitating event,
if any. In addition, we were interested in documenting the relative strength of these two effects.
To examine the independent effects of tangible consequences and social exchange rule viola-
tions, we solicited reports of situations in which participants experienced negative emotions as a
result of other people’s behavior and asked them to rate their emotional reactions to the event, the
extent to which it had tangible negative consequences, and the degree to which the other person’s
behavior violated social exchange rules.

METHOD

Participants

A total of 179 (79 male, 100 female, Mage = 22.3, SD = 3.8) were recruited. Of the participants,
91 (45 male, 46 female) were graduate students from across the university, and 88 participants
(34 male, 54 female) were undergraduates from the departmental participant pool.

Procedure

Because we did not want to inadvertently lead participants to focus solely on social exchange
rules by specifically asking about them, participants were instructed to “think of a time when
someone did or said something that upset you or made you have negative feelings.” We assumed
that a high proportion of such situations involve violations of social exchange rules.

564 THE JOURNAL OF SOCIAL PSYCHOLOGY

After participants took a moment to recall what happened, they reported the nature of
their relationship with the person and how long ago the event happened. Then, participants
answered three questions on 12-point scales: (a) “After this event happened, how good was your
relationship with this person,” (b) “After this event happened, how close did you feel to the other
person,” and (c) “Overall, how bad were the other person’s actions?”

Participants described the event and their reactions in writing and then rated the degree to
which the other person’s behavior had a negative effect on each of 11 tangible outcomes (1 = not
at all; 5 = extremely): “Your money or financial well-being”; “Your safety, health, or physical
well-being”; “Your performance or outcomes at work or at school”; “Your property or posses-
sions”; “Your ability to have something you wanted”; “The safety, health, or well-being of people
you care about”; “Your relationships with other people”; “How much fun or pleasure you had”;
“A friendship or close relationship”; “Your ability to do something that you wanted to do”; and
“Your membership in a group”. To confirm that these ratings reflected all major categories of
tangible outcomes for people’s well-being, two raters coded pilot data in which 100 participants
described their reactions to undesired events that were caused by other people. No outcomes were
identified that did not fall into one or more of these categories.

Participants also rated on 5-point scales (1 = not at all; 5 = extremely) the degree to which
the other person’s behavior violated 14 social exchange rules, specifically the degree to which
it was unfair, selfish, unethical or immoral, inconsiderate, unreasonable, disrespectful, dishonest,
irresponsible, uncooperative, and irrational; they also rated the degree to which the behavior
reflected a lack of concern for them, reflected lack of reciprocity, took advantage of them, and
showed that the person was undependable. Importantly, ratings of the tangible outcomes and
social exchange rule violations were counterbalanced across participants.

Participants rated how the other person’s behavior made them feel on several emotions that
included three items for anger: angry, irritated, and furious (1 = not at all; 5 = extremely). And
finally, participants answered two questions about their relationship: (a) “How much do you like
the person now?” and (b) “How much do you trust this person not to do something like this
again?” (1 = not at all; 5 = extremely). After completing the questionnaire, participants were
debriefed.

RESULTS

The episodes that participants described spanned a variety of relationships, including roman-
tic partners (28.5%), friends (27.4%), authority figures (11.2%), family members (not including
spouses; 10.6%), coworkers (6.7%), acquaintances (6.7%), and strangers (2.2%). The events they
described occurred between 0 and 12 years ago (M = 14.6 months), and the average rating of the
badness of the person’s actions was 7.5 (SD = 2.9) (“moderately” bad).

Ratings of the tangible consequences were summed to produce an index of tangible outcomes,
and the ratings of exchange rule violations reflected in the other person’s behavior were summed
to create an index of social exchange rule violations. The primary analyses involved hierarchical
multiple regression analyses that tested the effects of tangible consequences and perceived social
exchange rule violations on the outcome variables. To examine the effects fully, two analyses
were conducted for each outcome variable. In the first analysis (labeled Analysis 1 in Table 1),
participant gender (dummy coded) was entered on Step 1 of the regression analysis, followed

LEARY ET AL. 565

TABLE 1
Effects of Tangible Outcomes and Exchange Rule Violations

Analysis 1 Analysis 2
Step
number Predictor �R2 B 95% CI Predictor �R2 B 95% CI

Ratings of person’s actions

1. Gender .00 −.28 −1.13, .56 Gender .00 −.28 −1.13, .56
2. Tangible outcomes .04 .07 .02, .12 Exchange violations .27 .11 .08, .14
3. Tangible

outcomes × Gender
.00 −.03 −.13, .08 Exchange

violations × Gender
.01 .04 −.02, .09

4. Exchange violations .23 .11 .08, .14 Tangible outcomes .00 .01 −.04, .06
5. Exchange

violations × Gender
.01 .04 −.02, .10 Tangible

outcomes × Gender
.00 −.03 −.13, .07

Anger

1. Gender .00 .20 −.68, 1.07 Gender .00 .20 −.68, 1.07
2. Tangible outcomes .10 .11 .06, .16 Exchange violations .44 .15 .12, .17
3. Tangible

outcomes × Gender
.00 .00 −.10, .11 Exchange

violations × Gender
.01 .01 −.04, .06

4. Exchange violations .35 .14 .12, .17 Tangible outcomes .01 .03 −.02, .07
5. Exchange

violations × Gender
.00 .01 −.05, .06 Tangible

outcomes × Gender
.00 .04 −.05, .13

Current liking

1. Gender .01 .24 −.18, .67 Gender .01 .24 −.18, .67
2. Tangible outcomes .00 .00 −.03, .03 Exchange violations .11 −.04 −.05, −.02
3. Tangible

outcomes × Gender
.01 .03 −.02, .08 Exchange

violations × Gender
.01 .02 −.01, .05

4. Exchange violations .12 −.04 −.06, −.02 Tangible outcomes .01 .02 −.01, .05
5. Exchange

violations × Gender
.00 .02 −.02, .05 Tangible

outcomes × Gender
.00 .01 −.04, .05

Trust

1. Gender .00 −.05 −.40, .30 Gender .00 −.05 −.40, .30
2. Tangible outcomes .00 .01 −.02, .03 Exchange violations .03 −.01 −.03, −.001
3. Tangible

outcomes × Gender
.00 −.01 −.05, .04 Exchange

violations × Gender
.03 .03 .01, .05

4. Exchange violations .04 −.02 −.03, −.01 Tangible outcomes .01 .02 −.01, .04
5. Exchange

violations × Gender
.04 .04 .01, .06 Tangible

outcomes × Gender
.01 −.03 −.08, .02

Relationship quality

1. Gender .00 −.18 −1.15, .79 Gender .00 −.18 −1.15, .79
2. Tangible outcomes .01 −.03 −.09, .03 Exchange violations .17 −.10 −.13, −.07
3. Tangible

outcomes × Gender
.00 .02 −.11, .14 Exchange

violations × Gender
.00 .03 −.04, .10

(Continued)

566 THE JOURNAL OF SOCIAL PSYCHOLOGY

TABLE 1
(Continued)

Analysis 1 Analysis 2
Step
number Predictor �R2 B 95% CI Predictor �R2 B 95% CI

4. Exchange violations .17 −.11 −.14, −.07 Tangible outcomes .01 .04 −.02, .09
5. Exchange

violations × Gender
.01 .04 −.03, .11 Tangible

outcomes × Gender
.00 −.04 −.16, .08

Ratings of person’s actions

1. Gender .00 −.28 −1.13, .56 Gender .00 −.28 −1.13, .56
2. Tangible outcomes .04 .07 .02, .12 Exchange violations .27 .11 .08, .14
3. Tangible

outcomes × Gender
.00 −.03 −.13, .08 Exchange

violations × Gender
.01 .04 −.02, .09

4. Exchange violations .23 .11 .08, .14 Tangible outcomes .00 .01 −.04, .06
5. Exchange

violations × Gender
.01 .04 −.02, .10 Tangible

outcomes × Gender
.00 −.03 −.13, .07

Anger

1. Gender .00 .20 −.68, 1.07 Gender .00 .20 −.68, 1.07
2. Tangible outcomes .10 .11 .06, .16 Exchange violations .44 .15 .12, .17
3. Tangible

outcomes × Gender
.00 .00 −.10, .11 Exchange

violations × Gender
.01 .01 −.04, .06

4. Exchange violations .35 .14 .12, .17 Tangible outcomes .01 .03 −.02, .07
5. Exchange

violations × Gender
.00 .01 −.05, .06 Tangible

outcomes × Gender
.00 .04 −.05, .13

Current liking

1. Gender .01 .24 −.18, .67 Gender .01 .24 −.18, .67
2. Tangible outcomes .00 .00 −.03, .03 Exchange violations .11 −.04 −.05, −.02
3. Tangible

outcomes × Gender
.01 .03 −.02, .08 Exchange

violations × Gender
.01 .02 −.01, .05

4. Exchange violations .12 −.04 −.06, −.02 Tangible outcomes .01 .02 −.01, .05
5. Exchange

violations × Gender
.00 .02 −.02, .05 Tangible

outcomes × Gender
.00 .01 −.04, .05

Trust

1. Gender .00 −.05 −.40, .30 Gender .00 −.05 −.40, .30
2. Tangible outcomes .00 .01 −.02, .03 Exchange violations .03 −.01 −.03, −.001
3. Tangible

outcomes × Gender
.00 −.01 −.05, .04 Exchange

violations × Gender
.03 .03 .01, .05

4. Exchange violations .04 −.02 −.03, −.01 Tangible outcomes .01 .02 −.01, .04
5. Exchange

violations × Gender
.04 .04 .01, .06 Tangible

outcomes × Gender
.01 −.03 −.08, .02

Relationship quality

1. Gender .00 −.18 −1.15, .79 Gender .00 −.18 −1.15, .79
2. Tangible outcomes .01 −.03 −.09, .03 Exchange violations .17 −.10 −.13, −.07

(Continued)

LEARY ET AL. 567

TABLE 1
(Continued)

Analysis 1 Analysis 2
Step
number Predictor �R2 B 95% CI Predictor �R2 B 95% CI

3. Tangible
outcomes × Gender

.00 .02 −.11, .14 Exchange
violations × Gender

.00 .03 −.04, .10

4. Exchange violations .17 −.11 −.14, −.07 Tangible outcomes .01 .04 −.02, .09
5. Exchange

violations × Gender
.01 .04 −.03, .11 Tangible

outcomes × Gender
.00 −.04 −.16, .08

by the sum of the tangible consequences (mean centered; Step 2), the product of tangible con-
sequences and gender (Step 3), the sum of the ratings of social exchange rule violations (mean
centered; Step 4), and the product of exchange rule violations and gender (Step 5). This order
of entry was chosen to (a) control for possible gender differences in affect intensity (Diener,
Sandvik, & Larsen, 1985) and (b) partial out the effects of tangible consequences and any possible
interaction of tangible consequences by gender in order to (c) examine the residual contributions
of perceived social exchange rule violations (both by itself and in interactions with gender) on
the strength of participants’ emotional reactions.1

A second analysis (Analysis 2 on Table 1) was conducted to test the residual effects of per-
ceived tangible consequences on emotional reactions after controlling for gender, perceived social
exchange rule violations, and their interaction. Gender was again entered on Step 1, followed by
the sum of social exchange rule violations on Step 2, and the interaction of social exchange rule
violations and gender (Step 3). Then, the sum of the tangible consequences was entered on Step
4, and the interaction of tangible consequences by gender was tested on Step 5.

Ratings of Person’s Actions

Table 1 shows the results of these two analyses for each outcome variable. Gender did not predict
any of the outcomes in Step 1, so we will not mention it further. In Analysis 1, ratings of the
tangible outcomes accounted for 4.4% of the variance in how bad participants rated the other
person’s actions in Step 2, and in Step 3, the interaction of tangible consequences and gender did
not account for an increase in variance (see Table 1 for �R2, B, and 95% CIs for each effect).
In Step 4, social exchange rule violations accounted for 22.5% of the variance in ratings of the
action. The interaction with gender did not account for additional variance. Thus, ratings of social
exchange rule violations accounted for variance in how bad participants rated the person’s actions
beyond ratings of the tangible impact of those actions. In Analysis 2 (in which exchange rule
violations were entered before tangible consequences), social exchange rule violations accounted
for 27.0% of the variance when entered on Step 2, after which tangible consequences no longer
accounted for unique variance in the ratings.

Anger

Participants’ ratings of how angry, irritable, and furious they had felt were summed (α = .76). The
tangible outcomes accounted for 9.6% of the variance in anger on Step 2, but the interaction of

568 THE JOURNAL OF SOCIAL PSYCHOLOGY

tangible consequences and gender in Step 3 did not account for additional variance. In Step 4, the
sum of the social exchange rule violations accounted for 35.3% of the variance in anger, and this
effect was not moderated by gender in Step 5. Thus as expected, ratings of social exchange rule
violations accounted for the intensity of participants’ anger beyond their ratings of the tangible
effects of those events on their well-being. Indeed, violations of social exchange rules accounted
for more than 3 times as much variance as the tangible consequences. Analysis 2 showed that
exchange rule violations accounted for 44.1% of the variance in anger, whereas tangible conse-
quences accounted for no additional variance. These analyses show not only that anger ratings
were much more strongly related to perceived violations of social exchange rules than to tangi-
ble consequences but also that perceived exchange rule violations were associated with unique
variance in anger after tangible consequences were taken into account.

To follow-up on these analyses, we examined the specific social exchange rule violations that
most strongly predicted anger after controlling for ratings of tangible consequences. After con-
trolling for gender and ratings of tangible consequences, all 14 social exchange rule violations
correlated significantly with anger ratings (.19 < r’s < .51). However, when entered stepwise-
fashion into the regression equation, only three rule violations ultimately entered the model:
unreasonable, B = .72, 95% CI [.41, 1.03], �R2 = .23; disrespectful, B = .60, 95% CI [.31,
.90], �R2 = .09; and dishonest, B = .37, 95% CI [.13, .62], �R2 = .03. Although this pattern
may be specific to the kinds of undesired behaviors that our participants reported, it shows that,
not surprisingly, certain perceived rule violations may exert a predominate influence on people’s
reactions.

Relationship Implications

Liking. As seen in Table 1, neither gender, ratings of the tangible outcomes, nor the interac-
tion of gender and tangible consequences accounted for participants’ ratings of how much they
liked the person after the event. However, social exchange rule violations accounted for 11.8% of
the variance in liking. When the predictors were entered in the opposite order, social exchange
rule violations predicted liking ratings, but tangible outcomes had no effect.

Trust. Neither gender, ratings of the tangible outcomes, nor the interaction of gender and
tangible consequences predicted the extent to which participants indicated that they would trust
the person not to do something similar again. Social exchange rule violations accounted for 3.6%
of the variance in trust, and the interaction of exchange rule violations and gender accounted for
additional variance. Decomposition of this interaction revealed that men trusted the other person
less when perceptions of exchange rule violations were high than low, B = −.04, 95% CI [−.06,
−.02], but this effect was not obtained for women, B = .01, 95% CI [−.02, .03]. Also, men were
more trusting than women when exchange violations were low but less trusting when violations
were high.

Relationship quality. Participants’ ratings of how good their relationship was and how close
they felt to the other person correlated .90, so we averaged these items. As seen in Table 1, social
exchange rule violations accounted for 17% of the variance in relationship quality. When the
predictors were entered in the opposite order, no effect of tangible outcomes was obtained.

LEARY ET AL. 569

DISCUSSION

The results showed that the strength of people’s reactions to negative interpersonal events is a
function of both the perceived tangible outcomes of those events and their perceptions that others
have violated rules of social exchange. After controlling for the magnitude of perceived tangible
consequences, perceptions of social exchange rule violations explained variance in anger, how
bad the person’s actions were, how much participants liked and trusted the other person, and
ratings of the quality of the relationship after the event. And, when social exchange rule violations
were entered into the analyses before ratings of tangible consequences, the effects of tangible
consequences on participants’ reactions were minimal. Importantly, perceived social exchange
rule violations accounted for far more unique variance in participants’ reactions than did the
tangible outcomes that participants experienced.

These patterns support the hypothesis that strong emotional reactions to the behavior of other
people are fueled, at least in part, by people’s perceptions that the offending behavior involves
a violation of fundamental rules that guide interdependent interactions and relationships. Even
exceptionally minor transgressions can be viewed as violations of social exchange rules and,
thus, spark a stronger response than would be predicted by the tangible consequences.

Although participants’ reports of social exchange rule violations involved many kinds of
relationships—with friends, romantic partners, family members, authority figures, strangers, and
others—we did not have a large enough sample of each relationship type to examine whether
perceived social exchange rule violations evoke stronger or different kinds of reactions in some
relationships than in others. One might expect that people expect stronger adherence to social
exchange rules in relationships in which interdependence is greater because the costs of rule
violations are higher over time. Yet, if trust and satisfaction are high, people in highly interdepen-
dent relationships may overlook occasional violations (or at least restrain their reactions to them)
because they ultimately trust the other’s intentions and do not wish to damage the relationship
by overreacting. Viewed in this way, we predict that social exchange rule violations evoke the
strongest, most disproportionate reactions in highly interdependent relationships in which trust
(and/or satisfaction) is low. Research is needed to understand the features of relationships that
influence disproportionate reactions.

In many instances, people are motivated, at minimum, to send an unambiguous signal that
they are aware of the violations and will not tolerate such actions in the future. Cohen et al.
(1996) described this deterrence function of strong reactions to signs of disrespect, but the effect
applies to a broad array of situations in which actual or potential exchange partners violate the
rules that guide mutually beneficial social exchanges. From this perspective, expressions of anger,
including overt aggression, in these instances may be efforts to show that one will not tolerate
social exchange rule violations (Felson, 1982; Miller, 2001; Toch, 1992).

In addition, when violations of social exchange rules are particularly egregious, people may be
motivated to punish the perpetrator. Most work on retributive justice has focused on how people
respond to those who commit immoral actions that harm themselves or other people (Carlsmith
& Darley, 2008), but people sometimes engage in retribution for behaviors, that, in themselves,
did not hurt anyone (Felson, 1978). The psychological heuristic underlying such reactions seems
to be that people who cannot be trusted to be good social exchange partners should be put on
notice, if not punished.

570 THE JOURNAL OF SOCIAL PSYCHOLOGY

Our findings are entirely consistent with other research on reactions to unfairness, selfishness,
and disrespect (e.g., Allen & Leary, 2010; Bembenek et al., 2007; Brebels et al., 2008; Cohen
et al., 1996; Colquitt, Conlon, Wesson, Porter, & Ng, 2001; Miller, 2001; Stephenson et al.,
2011) but extend previous thinking. From our perspective, perceived unfairness and disrespect
evoke strong reactions to otherwise trivial events because they are among the most common—and
perhaps the most potentially harmful—violations of social exchange rules. The social exchange
rule violation hypothesis provides a parsimonious, overarching perspective that not only inte-
grates previous findings but also offers insights into the conditions under which people react
more strongly to events than one might expect based on their tangible consequences. Indeed, our
analysis suggested that participants reacted most strongly to violations of rules to be not only
respectful but also reasonable and honest.

Of course, people sometimes react strongly to events for reasons that have little to do with
violations of social exchange rules. Often, the sheer magnitude of tangible consequences can
send people over the edge, and, as noted, displaced aggression and cumulative stress can lower
the threshold at which people respond to events. Furthermore, people are sometimes angered
when events violate their moral convictions, whether or not such events are perceived to be unfair
(Mullen & Skitka, 2006). The social exchange violation hypothesis explains one broad category
of events that evoke emotional reactions, but we are not suggesting that other factors do not
sometimes lead to strong responses to events with no tangible consequences.

When people react strongly to seemingly minor events, the aggrieved party and the perpetrator
often disagree about whether the complainant’s strong reaction is warranted. When a person
becomes angry because a friend picked them up 10 minutes later than promised or a partner leaves
dirty socks on the floor, the perpetrator justifiably wonders why such a small infraction elicited
such a strong reaction. Yet, the disagreement may arise because, whereas the perpetrator rightly
perceives that their behavior had only minor tangible consequences, the aggrieved party is focused
on the violation of a social exchange rule—for example, to be dependable or not behave selfishly.
Even behaviors with no tangible consequences can provide diagnostic information about another
person’s viability as a social exchange partner. In fact, from the aggrieved person’s perspective,
smaller infractions may sometimes be more informative than large ones because less effort is
needed to avoid violating the rule.

One question that remains unanswered is whether people’s reactions to situations in which
they perceive that social exchange rules were violated are focused specifically on the violation or
on the violator. That is, are people’s reactions fundamentally in response to the violation itself or
to the person involved who is viewed as one who breaks important social rules? Psychologically,
this is an important distinction that has implications for understanding and dealing with strong
emotional reactions. Our hunch is that emotional reactions to exchange rule violations are pri-
marily focused on the person rather than the action. The fact that the precipitating event is
often trivial suggests that people are reacting to the perpetrator’s intent, lack of consideration,
or general unsuitability as an exchange partner rather than to the behavior per se. Furthermore,
as Cunningham’s work has shown (Cunningham et al., 2005), in many cases, mild transgressions
initially elicit a very weak reaction, if any at all. Only with repeated exposures do people begin
to react strongly, suggesting that their response is mediated by inferences about the person who
repeatedly does such a thing. So, for example, failing to throw one’s dirty socks in the laun-
dry may evoke little response initially but fuel increasing upset as violations accumulate over

LEARY ET AL. 571

time so that eventually the person explodes over a dirty sock. In addition, work in evolutionary
psychology suggests that cheater detector systems are designed to identify people who take ben-
efits to which they are not entitled rather than isolated acts of cheating per se and that people
classify others as “free-riders” based on inferences about their intentions rather than their actual
contributions to collective outcomes (Delton et al., 2012).

If we are correct that emotional reactions that go beyond the impact of tangible consequences
are based more on violators than on violations, the cognitive inferential processes that lead peo-
ple to draw conclusions about those who violate social exchange rules mediate these reactions in
important ways. Given that the perception of an exchange rule violation underlies the strong reac-
tions, considering attributional processes and biases is informative. Research on attributions about
wrongdoing is particularly relevant. For example, inferences of hostile intent fuel strong reactions
to ambiguous behaviors, and people who are biased to perceive hostile intent are inclined to react
to undesired events with greater anger and aggression (Crick, Grotpeter, & Bigbee, 2002; Dodge,
Price, Bachorowski, & Newman, 1990). In the same way, just as some people are inclined to
infer hostile intent on the part of other people, some individuals may be predisposed to judge
others’ undesired actions as violations of social exchange rules, leading them to respond strongly
to trivial events.

For example, people who score high in psychological entitlement (Campbell, Bonacci,
Shelton, Exline, & Bushman, 2004) may believe not only that they deserve to receive desired
material, financial, and experiential outcomes more than other people but also that they are enti-
tled to be treated particularly well by others. If so, they may be highly sensitive to violations of
social exchange rules. Similarly, people who are high in conscientiousness—who, among other
things, tend to follow socially prescribed norms and rules (Roberts, Jackson, Fayard, Edmonds, &
Meints, 2009)—may be attuned to instances in which other people violate basic norms and rules,
including rules of social exchange. Along the same lines, recent work in organizational behavior
suggests that dispositional trust is related to people’s judgment of and sensitivity to violations of
fairness (Bianchi & Brockner, 2012). Research is needed to examine individual differences that
moderate sensitivity to violations of social exchange rules and the intensity of people’s reactions
to perceived infractions.

Viewed from a functional, social exchange perspective, reactions that are disproportionate
to the tangible implications of an event may or may not be “overreactions.” Not only might
some strong reactions to trivial behaviors be reasonable and functional in alerting people to be
wary of those who violate social exchange rules, but, as noted, strong reactions may show oth-
ers that exchange rule violations will not be tolerated (Cohen et al., 1996). However, when the
violations themselves are inconsequential and do not signify future problems, disproportionate
reactions are neither functional nor reasonable. Much of the time, people react so quickly, and
often automatically, to what other people do that they end up making mountains out of molehills.

572 THE JOURNAL OF SOCIAL PSYCHOLOGY

ACKNOWLEDGEMENT

Data and materials can be located at https://osf.io/nydsz/.

NOTE

1. Although we assumed that any combined effects of tangible outcomes and social exchange rule violations would be
additive, we conducted exploratory analyses to test for the possibility of an interaction. In no case was any hint of an
interaction between tangible outcomes and exchange rule violations on the outcome variables obtained.

AUTHOR NOTES

Mark R. Leary, Kate J. Diebels, Katrina P. Jongman-Sereno, and Xuan Duong Fernandez are affiliated with the
Department of Psychology and Neuroscience, Duke University.

REFERENCES

Allen, A. B., & Leary, M. R. (2010). Reactions to others’ selfish actions in the absence of tangible consequences. Basic
and Applied Social Psychology, 32, 26–34. doi:10.1080/01973530903539861

Barnard, G. W., Vera, H., Vera, M. I., & Newman, G. (1982). Till death do us part: A study of spouse murder. Bulletin of
the American Academy of Psychiatry & the Law, 10, 271–280.

Baron, J. (1993). Heuristics and biases in equity judgments: A utilitarian approach. In B. A. Mellers & J. Baron (Eds.),
Psychological perspectives on justice: Theory and applications (pp. 109–137). New York, NY: Cambridge University
Press.

Baumeister, R. F., Smart, L., & Boden, J. M. (1996). Relation of threatened egotism to violence and aggression: The dark
side of high self-esteem. Psychological Review, 103, 5–33. doi:10.1037/0033-295X.103.1.5

Bembenek, A. F., Beike, D. R., & Schroeder, D. A. (2007). Justice violations, emotional reactions, and justice-seeking
responses. In D. De Cremer (Ed.), Advances in the psychology of justice and emotions (pp. 15–36). Greenwich, CT:
Information Age Publishing.

Bettencourt, B. A., & Miller, N. (1996). Gender differences in aggression as a function of provocation: A meta-analysis.
Psychological Bulletin, 119, 422–447. doi:10.1037/0033-2909.119.3.422

Bianchi, E. C., & Brockner, J. (2012). In the eyes of the beholder? The role of dispositional trust in judgments
of procedural and interactional fairness. Organizational Behavior and Human Decision Processes, 118, 46–59.
doi:10.1016/j.obhdp.2011.12.005

Bies, R. J., & Moag, J. S. (1986). Interactional justice: Communication criteria for fairness. In B. Sheppard (Ed.), Research
on negotiation in organizations (pp. 43–55). Greenwich, CT: JAI.

Blau, P. (1960). A theory of social integration. American Journal of Sociology, 65, 545–556.
doi:10.1086/ajs.1960.65.issue-6

Brebels, L., De Cremer, D., & Sedikides, C. (2008). Retaliation as a response to procedural unfairness: A self-regulatory
approach. Journal of Personality and Social Psychology, 95, 1511–1525. doi:10.1037/a0012821

Brown, J., James, K., & Taylor, A. (2010). Caught in the rejection abuse cycle: Are we really treating perpetrators of
domestic abuse effectively? Journal of Family Therapy, 32, 280–307. doi:10.1111/j.1467-6427.2010.00494.x

Buckley, K. E., Winkel, R. E., & Leary, M. R. (2004). Reactions to acceptance and rejection: Effects of level and sequence
of relational evaluation. Journal of Experimental Social Psychology, 40, 14–28. doi:10.1016/S0022-1031(03)00064-7

Buss, D. M., Abbott, M., Angleitner, A., Asherian, A., Biaggio, A., Blanco-Villasenor, A. . . . Yang, K.-S. (1990).
International preferences in selecting mates: A study of 37 cultures. Journal of Cross-cultural Psychology, 21, 5–47.
doi:10.1177/0022022190211001

LEARY ET AL. 573

Campbell, W. K., Bonacci, A. M., Shelton, J., Exline, J. J., & Bushman, B. J. (2004). Psychological entitlement:
Interpersonal consequences and validation of a self-report measure. Journal of Personality Assessment, 83, 29–45.
doi:10.1207/s15327752jpa8301_04

Carlsmith, M. K., & Darley, J. M. (2008). Psychological aspects of retributive justice. Advances in Experimental Social
Psychology, 40, 284–299. doi:10.1016/S0065-2601(07)00004-4

Cohen, D. (1998). Culture, social organization, and patterns of violence. Journal of Personality and Social Psychology,
75, 408–419. doi:10.1037/0022-3514.75.2.408

Cohen, D., Nisbett, R. E., Bowdle, B. F., & Schwarz, N. (1996). Insult, aggression, and the southern cul-
ture of honor: An ‘experimental ethnography.’ Journal of Personality and Social Psychology, 70, 945–960.
doi:10.1037/0022-3514.70.5.945

Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng, K. Y. (2001). Justice at the millennium: A
meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychology, 86, 425–445.
doi:10.1037/0021-9010.86.3.425

Cosmides, L. (1989). The logic of social exchange: Has natural selection shaped how humans reason? Studies with the
Wason selection task. Cognition, 31, 187–276. doi:10.1016/0010-0277(89)90023-1

Cosmides, L., & Tooby, J. (1992). Cognitive adaptations for social exchange. In J. Barkow, L. Cosmides, & J. Tooby
(Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 163–228). New York, NY:
Oxford University Press.

Cosmides, L., & Tooby, J. (2008). Can a general deontic logic capture the facts of human moral reasoning? How the
mind interprets social exchange rules and detects cheaters. In W. Sinnott-Armstrong (Ed.), Moral psychology: The
evolution of morality: Adaptations and innateness (Vol. 1, pp. 53–119). Cambridge, MA: MIT Press.

Cottrell, C. A., Neuberg, S. L., & Li, N. P. (2007). What do people desire in others? A sociofunctional perspective
on the importance of different valued characteristics. Journal of Personality and Social Psychology, 92, 208–231.
doi:10.1037/0022-3514.92.2.208

Crick, N. R., Grotpeter, J. K., & Bigbee, M. A. (2002). Relationally and physically aggressive children’s intent attribu-
tions and feelings of distress for relational and instrumental peer provocations. Child Development, 73, 1134–1142.
doi:10.1111/cdev.2002.73.issue-4

Cropanzano, R., & Byrne, Z. S. (2000). How workers manage relationships in a complex social world: A multi-foci
approach to procedural and interactional justice. In M. Van Vugt, M. Snyder, T. Tyler, & A. Biels (Eds.), Cooperation
in modern society: Dilemmas and solutions (pp. 142–161). London, UK: Routledge.

Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review. Journal of Management,
31, 874–900. doi:10.1177/0149206305279602

Cunningham, M. R., Shamblen, S. R., Barbee, A. P., & Ault, L. K. (2005). Social allergies in romantic relationships:
Behavioral repetition, emotional sensitization, and dissatisfaction in dating couples. Personal Relationships, 12,
273–295. doi:10.1111/j.1350-4126.2005.00115.x

Delton, A. W., Cosmides, L., Guemo, M., Robertson, T. E., & Tooby, J. (2012). The psychosemantics of free riding:
Dissecting the architecture of a moral concept. Journal of Personality and Social Psychology, 102, 1252–1270.
doi:10.1037/a0027026

Diener, E., Sandvik, E., & Larsen, R. J. (1985). Age and sex effects for emotional intensity. Developmental Psychology,
21, 542–546. doi:10.1037/0012-1649.21.3.542

Dodge, K. A., Price, J. M., Bachorowski, J., & Newman, J. P. (1990). Hostile attributional biases in severely aggressive
adolescents. Journal of Abnormal Psychology, 99, 385–392. doi:10.1037/0021-843X.99.4.385

Eisenberger, R., Lynch, P., Aselage, J., & Rohdieck, S. (2004). Who takes the most revenge? Individual dif-
ferences in negative reciprocity norm endorsement. Personality and Social Psychology Bulletin, 30, 787–799.
doi:10.1177/0146167204264047

Emerson, E. (1976). Social exchange theory. Annual Review of Sociology, 2, 335–362.
Fast, N. J., & Chen, S. (2009). When the boss feels inadequate: Power, incompetence, and aggression. Psychological

Science, 20, 1406–1413. doi:.10.1111/j.1467-9280.2009.02452.x
Felson, R. B. (1978). Aggression as impression management. Social Psychology Quarterly, 41, 205–213.

doi:10.2307/3033557
Felson, R. B. (1982). Impression management and the escalation of aggression and violence. Social Psychology Quarterly,

45, 245–254. doi:10.2307/3033920
Fodor, J. (2000). Why we are so good at catching cheaters. Cognition, 75, 29–32. doi:10.1016/S0010-0277(99)00077-3

574 THE JOURNAL OF SOCIAL PSYCHOLOGY

Folger, R., & Skarlicki, D. P. (1998). A popcorn metaphor for employee aggression. In R. Griffin, A. O’Leary-Kelly, & J.
Collins (Eds.), Dysfunctional behavior in organizations: Violent and deviant behavior. Monographs in organizational
behavior and industrial relations (pp. 43–81). Stanford, CT: JAI.

Frijda, N. (1986). The emotions. Cambridge, UK: Cambridge University Press.
Gonsalkorale, K., & Williams, K. D. (2007). The KKK won’t let me play: Ostracism even by a despised outgroup hurts.

European Journal of Social Psychology, 37, 1176–1186. doi:10.1002/(ISSN)1099-0992
Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25, 161–178.

doi:10.2307/2092623
Hall, J. A. (2011). Sex differences in friendship expectations: A meta-analysis. Journal of Social and Personal

Relationships, 28, 723–747. doi:10.1177/0265407510386192
Homans, G. (1961). Social behavior. New York, NY: Harcourt, Brace & World.
Kernis, M. H., Granneman, B. D., & Barclay, L. C. (1989). Stability and level of self-esteem as predictors of anger arousal

and hostility. Journal of Personality and Social Psychology, 56, 1013–1022. doi:10.1037/0022-3514.56.6.1013
Leary, M. R., & Jongman-Sereno, K. P. (2015). When rejection kills: The central role of low relational value in school

violence. International Journal of Developmental Science, 8, 25–27.
Leary, M. R., Kowalski, R. M., Smith, L., & Phillips, S. (2003). Teasing, rejection, and violence: Case studies of the

school shootings. Aggressive Behavior, 29, 202–214. doi:10.1002/ab.10061
Leary, M. R., Twenge, J. M., & Quinlivan, E. (2006). Interpersonal rejection as a determinant of anger and aggression.

Personality and Social Psychology Review, 10, 111–132. doi:10.1207/s15327957pspr1002_2
Lind, E. A., Kanfer, R., & Earley, P. C. (1990). Voice, control, and procedural justice: Instrumental and non-

instrumental concerns in fairness judgments. Journal of Personality and Social Psychology, 59, 952–959.
doi:10.1037/0022-3514.59.5.952

Lind, E. A., Kulik, C. T., Ambrose, M., & de Vera-Park, M. V. (1993). Individual and corporate dispute resolution: Using
procedural fairness as a decision heuristic. Administrative Science Quarterly, 38, 224–251. doi:10.2307/2393412

Lind, E. A., & Tyler, T. R. (1988). The social psychology of procedural justice. New York, NY: Plenum.
Mann, C. R. (1988). Getting even? Women who kill in domestic encounters. Justice Quarterly, 5, 33–51.

doi:10.1080/07418828800089601
Marcus-Newhall, A., Pedersen, W. C., Carlson, M., & Miller, N. (2000). Displaced aggression is alive and well: A

meta-analytic review. Journal of Personality and Social Psychology, 78, 670–689. doi:10.1037/0022-3514.78.4.670
Miller, D. T. (2001). Disrespect and the experience of injustice. Annual Review of Psychology, 52, 527–553.

doi:10.1146/annurev.psych.52.1.527
Miller, W. I. (1993). Humiliation and other essays on honor, social discomfort, and violence. Ithaca, NY: Cornell

University Press.
Mullen, E., & Skitka, L. J. (2006). Exploring the psychological underpinnings of the moral mandate effect:

Motivated reasoning, group differentiation, or anger? Journal of Personality and Social Psychology, 90, 629–643.
doi:10.1037/0022-3514.90.4.629

Neighbors, C., Vietor, N. A., & Knee, C. R. (2002). A motivational model of driving anger and aggression. Personality
and Social Psychology Bulletin, 28, 324–335. doi:10.1177/0146167202286004

Pedersen, W. C., Gonzales, C., & Miller, N. (2000). The moderating effect of trivial triggering provocation on displaced
aggression. Journal of Personality and Social Psychology, 78, 913–927. doi:10.1037/0022-3514.78.5.913

Roberts, B. W., Jackson, J. J., Fayard, J. V., Edmonds, G., & Meints, J. (2009). Conscientiousness. In M. R. Leary & R.
H. Hoyle (Eds.), Handbook of individual differences in social behavior (pp. 369–381). New York, NY: Guilford.

Robinson, S. L. (1996). Trust and breach of the psychological contract. Administrative Science Quarterly, 41, 574–599.
doi:10.2307/2393868

Robinson, S. L., Kraatz, M. S., & Rousseau, D. M. (1994). Changing obligations and the psychological contract: A
longitudinal study. The Academy of Management Journal, 37, 137–152. doi:10.2307/256773

Rousseau, D. M. (1995). Psychological contracts in organizations: Understanding written and unwritten agreements.
Thousand Oaks, CA: Sage.

Sageman, M. (2008). A strategy for fighting international Islamist terrorists. The ANNALS of the American Academy of
Political and Social Science, 618, 223–231. doi:10.1177/0002716208317051

Scarpa, A., & Raine, A. (2000). Violence associated with anger and impulsivity. In J. C. Borod (Ed.), The neuropsychology
of emotion (pp. 320–339). New York, NY: Oxford University Press.

LEARY ET AL. 575

Scher, S. J. (1997). Measuring the consequences of injustice. Personality and Social Psychology Bulletin, 23, 482–497.
doi:10.1177/0146167297235004

Skitka, L. J., & Crosby, F. (2003). Trends in the social psychological study of justice. Personality and Social Psychology
Review, 7, 282–285. doi:10.1207/pspr.2003.7.issue-4

Stephenson, P. S., Martsolf, D. S., & Draucker, C. B. (2011). Proximal antecedents to violent events in adolescent dating
relationships. Issues in Mental Health Nursing, 32, 670–677. doi:10.3109/01612840.2011.596612

Tannenbaum, D., Uhlmann, E. L., & Diermeier, D. (2011). Moral signals, public outrage, and immaterial harms. Journal
of Experimental Social Psychology, 47, 1249–1254. doi:10.1016/j.jesp.2011.05.010

Taylor, M. S., Tracy, K. B., Renard, M. K., Harrison, J. K., & Carroll, S. J. (1995). Due process in performance appraisal:
A quasi-experiment in procedural justice. Administrative Science Quarterly, 40, 495–523. doi:10.2307/2393795

Toch, H. (1992). Violent men: An inquiry into the psychology of violence. Washington, DC: American Psychological
Association.

Twenge, J., & Campbell, W. K. (2003). “Isn’t it fun to get the respect that we’re going to deserve?” Narcissism, social
rejection, and aggression. Personality and Social Psychology Bulletin, 29, 261–272. doi:10.1177/0146167202239051

Tyler, T. R. (1987). Conditions leading to value-expressive effects in judgments of procedural justice: A test of four
models. Journal of Personality and Social Psychology, 52, 333–344. doi:10.1037/0022-3514.52.2.333

Tyler, T. R., Rasinski, K. A., & Spodick, N. (1985). Influence of voice on satisfaction with leaders: Exploring the meaning
of process control. Journal of Personality and Social Psychology, 48, 72–81. doi:10.1037/0022-3514.48.1.72

Vidmar, N. (2000). Retribution and revenge. In J. Sanders & V. L. Hamilton (Eds.), Handbook of justice research in law
(Vol. 2, pp. 31–63). New York, NY: Kluwer.

Received July 30, 2015
Accepted August 16, 2015

Copyright of Journal of Social Psychology is the property of Taylor & Francis Ltd and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder’s express written permission. However, users may print, download, or email
articles for individual use.

Chadee_cover.indd viiChadee_cover.indd vii 1/14/2011 8:26:11 PM1/14/2011 8:26:11 PM

Chadee_ffirs.indd iiChadee_ffirs.indd ii 12/13/2010 2:12:49 PM12/13/2010 2:12:49 PM

Theories in Social Psychology

Chadee_ffirs.indd iChadee_ffirs.indd i 12/13/2010 2:12:49 PM12/13/2010 2:12:49 PM

Chadee_ffirs.indd iiChadee_ffirs.indd ii 12/13/2010 2:12:49 PM12/13/2010 2:12:49 PM

Theories in Social Psychology

Edited by

Derek Chadee

A John Wiley & Sons, Ltd., Publication

Chadee_ffirs.indd iiiChadee_ffirs.indd iii 12/13/2010 2:12:49 PM12/13/2010 2:12:49 PM

This edition first published 2011

© 2011 Blackwell Publishing Ltd

Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing

program has been merged with Wiley’s global Scientific, Technical, and Medical business to form

Wiley-Blackwell.

Registered Office

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ,

United Kingdom

Editorial Offices

350 Main Street, Malden, MA 02148-5020, USA

9600 Garsington Road, Oxford, OX4 2DQ, UK

The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

For details of our global editorial offices, for customer services, and for information about how to

apply for permission to reuse the copyright material in this book please see our website at

www.wiley.com/wiley-blackwell.

The right of Derek Chadee to be identified as the author of the editorial material in this work has been

asserted in accordance with the UK Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,

or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or

otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior

permission of the publisher.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print

may not be available in electronic books.

Designations used by companies to distinguish their products are often claimed as trademarks. All

brand names and product names used in this book are trade names, service marks, trademarks or

registered trademarks of their respective owners. The publisher is not associated with any product or

vendor mentioned in this book. This publication is designed to provide accurate and authoritative

information in regard to the subject matter covered. It is sold on the understanding that the publisher

is not engaged in rendering professional services. If professional advice or other expert assistance is

required, the services of a competent professional should be sought.

Library of Congress Cataloging-in-Publication Data

Theories in social psychology / edited by Derek Chadee.

p. cm.

Includes bibliographical references and index.

ISBN 978-1-4443-3122-6 (hardcover : alk. paper) – ISBN 978-1-4443-3123-3 (pbk. : alk. paper)

1. Social psychology. I. Chadee, Derek.

HM1033.T434 2011

302.01–dc22

2010041341

A catalogue record for this book is available from the British Library.

Set in 10/13 Minion by SPi Publisher Services, Pondicherry, India

1 2011

Chadee_ffirs.indd ivChadee_ffirs.indd iv 12/13/2010 2:12:49 PM12/13/2010 2:12:49 PM

To my daughter, Rúhíyyih – providing another perspective

Chadee_ffirs.indd vChadee_ffirs.indd v 12/13/2010 2:12:50 PM12/13/2010 2:12:50 PM

Chadee_ffirs.indd viChadee_ffirs.indd vi 12/13/2010 2:12:50 PM12/13/2010 2:12:50 PM

List of Contributors ix
Acknowledgments xiv

Introduction 1

Part I Social Cognition 11

1 Toward Freedom: Reactance Theory Revisited 13
Derek Chadee

2 Inconsistency in Cognition: Cognitive Dissonance 44
Paul R. Nail and Kurt A. Boniecki

3 Attribution Theories: How People Make Sense of Behavior 72
Bertram F. Malle

4 The Elaboration Likelihood Model of Persuasion: Thoughtful and
Non-Thoughtful Social Influence 96

Benjamin C. Wagner and Richard E. Petty

Part II Social Comparison 117

5 Social Comparison: Motives, Standards, and Mechanisms 119
Katja Corcoran, Jan Crusius, and Thomas Mussweiler

6 Relative Deprivation: Understanding the Dynamics of Discontent 140
Jenny Carrillo, Alexandra F. Corning, Tara C. Dennehy,

and Faye J. Crosby

Part III Social Reinforcement 161

7 Evaluating Fairness: Critical Assessment of Equity Theory 163
Denise M. Polk

Contents

Chadee_ftoc.indd viiChadee_ftoc.indd vii 12/13/2010 2:13:28 PM12/13/2010 2:13:28 PM

viii Contents

8 Interdependence in Social Interaction 191
Ann C. Rumble

Part IV Self 209

9 Self-Categorization and Social Identification:
Making Sense of Us and Them 211

Katharina Schmid, Miles Hewstone, and Ananthi Al Ramiah

10 Social Categorization Theories: From Culture to Cognition 232
Richard J. Crisp and Angela T. Maitner

11 Symbolic Interactionism: From Gestalt to Cybernetics 250
Andreas Schneider

12 Impression Management: Influencing Perceptions of Self 280
Meni Koslowsky and Shani Pindek

Author Index 297
Subject Index 301

Chadee_ftoc.indd viiiChadee_ftoc.indd viii 12/13/2010 2:13:28 PM12/13/2010 2:13:28 PM

Contributors

Derek Chadee is Senior Lecturer in the Department of Behavioural Sciences and Director of
the ANSA McAL Psychological Research Centre, University of the West Indies, St. Augustine
Campus. His current research interests include the social psychology of fear of crime and
general fear, antecedents of emotions, copycat behavior, the media, HIV/AIDS and jury
decision-making. In 2004, he was a Fulbright Scholar undertaking research on fear of crime
in two American universities. He has maintained a cross-cultural research agenda.

Ananthi Al Ramiah is a Leverhulme Postdoctoral Research Associate at the Department of
Experimental Psychology, University of Oxford, and her research interests lie broadly in the
area of intergroup conflict and cooperation. One line of her research assesses how social
identity, threat, and status differentials contribute to intergroup conflict. In addition, her
research aims to uncover the role that ethno-religious diversity plays in explaining percep-
tions of intergroup trust and the extent to which these relationships are moderated by
intergroup contact. Another line of work assesses how the intergroup context is perceived
and contributed to by members of majority- and minority-status groups in an effort to find
models of intergroup harmony in diverse settings.

Kurt A. Boniecki is Associate Professor of Psychology at the University of Central Arkansas
and gained his Ph.D. in social psychology from the University of Florida in 1997. His
research primarily examines the causes of prejudice and stereotyping. He also has academic
interests in the teaching of psychology, interpersonal attraction, and the home-field advan-
tage in sports.

Jenny Carrillo is Senior Vice President of External Affairs and Organizational Effectiveness
at Planned Parenthood of Southern New England (PPSNE). Her work at PPSNE is consist-
ent with her doctoral research at Yale University on women’s and immigrant health and the
acculturative process. Prior to this, she worked with adolescents focused on social justice
and education. She is active as a board member in several non-profit organizations that

Chadee_flast.indd ixChadee_flast.indd ix 12/13/2010 2:13:12 PM12/13/2010 2:13:12 PM

x Contributors

center on health, education, and racial disparities. She previously worked as a consultant
with McKinsey & Co. and at Pace University.

Katja Corcoran (née Rüter) currently works as an Assistant Professor at the University of
Cologne, Germany. She studied psychology in Tübingen and Berlin in Germany, and
Amherst, USA. She gained her Ph.D. at the University of Würzburg, Germany, in 2004, and
spent one year at Northwestern University, USA, with a Feodor Lynen fellowship granted by
the Alexander von Humboldt Foundation. Her research focuses on social comparisons,
especially on standard selection processes and aspects of efficiency. She has published in the
Journal of Personality and Social Psychology, the Journal of Experimental Social Psychology,
and Social Cognition.

Alexandra F. Corning is professor in the Department of Psychology, University of Notre Dame.
Her research is concerned with problems that affect groups unevenly. Much of her work is
aimed at identifying and explaining the mechanisms underlying the perception of discrimina-
tion; in particular, how people arrive at the conclusions that discrimination has or has not taken
place when the situation is ambiguous. Taking a social-cognitive approach, her research is
focused on identifying individual-difference as well as situational factors that influence these
perceptions. She also conducts research aimed at identifying the processes that influence young
women’s and girls’ body dissatisfaction and eating problems.

Richard J. Crisp is Professor of Psychology in the Centre for the Study of Group Processes,
University of Kent. He completed his BA in experimental psychology at Oxford University
and his Ph.D. at Cardiff University. He has published widely on the psychology of prejudice,
social categorization, group processes, and intergroup relations. He is a past winner of the
Society for the Psychological Study of Social Issues Louise Kidder Early Career Award and
was the 2006 recipient of the British Psychological Society’s Spearman Medal. He is cur-
rently Associate Editor of the Journal of Experimental Social Psychology and in 2009 he was
elected an Academician of the Academy of Social Sciences.

Faye J. Crosby is Professor in the Psychology Department, University of California, Santa
Cruz. She is a scholar, writer, consultant, and social activist. She received her Ph.D. in 1976.
Previous faculty appointments include Rhode Island College, Yale University, the Kellogg
School of Management, and Smith College. Crosby has authored or coauthored five books
and has edited or co-edited a further ten volumes, and over 150 articles and chapters. Most
of her work concerns sex and race discrimination and focuses on remedies. She is the recip-
ient of numerous awards and is the founder of Nag’s Heart, an organization whose mission
is the replenishment of the feminist spirit.

Jan Crusius is an Assistant Professor in Social Psychology at the University of Cologne,
Germany. After studying psychology in Jena, Germany, and Seville, Spain, he finished his
doctoral degree in Social Psychology at the University in Cologne in 2009. He is interested
in the role of social comparisons in emotion, judgment and decision making, and con-
sumer behavior.

Chadee_flast.indd xChadee_flast.indd x 12/13/2010 2:13:12 PM12/13/2010 2:13:12 PM

Contributors xi

Tara C. Dennehy is a graduate student in the Mind, Brain, & Behavior Program at San
Francisco State University. She received her BA in 2007 with highest honors in psychology
from the University of California, Santa Cruz. Her main research interests lie in the inter-
section between cognitive psychology and social cognition. Specifically, she conducts
research examining the relationship between categorization and psychological essentialism,
and research on the action-related determinants of entry into attentional awareness.

Miles Hewstone is Professor of Social Psychology and Fellow of New College, University of
Oxford. He has published widely on the topics of social cognition and intergroup relations.
He was awarded the British Psychological Society’s Spearman Medal in 1987 and its
President’s Award for Distinguished Contributions to Research in 2001. He was a Fellow at
the Center for Advanced Study in the Behavioral Sciences, Stanford, California 1987–8 and
1999–2000, and is a Fellow of the British Academy.

Meni Koslowsky is Professor at Bar-Ilan University, Ramat Gan, Israel. He received his
Ph.D. in psychology from Columbia University and is presently a Professor in the Psychology
Department of Bar-Ilan University. His main focus of research includes organizational
behavior, the stress–strain relationship, social power, and methodological issues. He has
written more than 100 articles, authored and edited five books, and has presented at nearly
100 conferences.

Angela T. Maitner is currently an Assistant Professor of Psychology at the American
University of Sharjah, United Arab Emirates. She completed her Ph.D. at the University of
California, Santa Barbara. Broadly, her research investigates intergroup relations from a
social cognitive perspective. More specifically she investigates the impact of exposure to
diversity and multiple categorization on person perception and stereotyping, and the regu-
latory role of emotion in intergroup relations.

Bertram F. Malle is Professor in the Department of Cognitive, Linguistic and Psychological
Sciences, Brown University. He received his Ph.D. at Stanford University in 1995 and joined
the University of Oregon Psychology Department. He received the Society of Experimental
Social Psychology Outstanding Dissertation award in 1995 and a National Science Foundation
CAREER award in 1997. His research focuses on social cognition and the folk theory of mind,
exploring such issues as intentionality judgments, mental state inferences, behavior explana-
tions, and moral judgments. Malle is currently writing a book on Social Cognitive Science.

Thomas Mussweiler has been Professor at the University of Cologne since 2004. His research
examines a variety of issues in social cognition, social judgment, and decision making, in
particular how comparison processes shape human judgment, affect, and behavior. He has
published in the Psychological Review, Psychological Science, the Journal of Personality and
Social Psychology, and the Journal of Experimental Psychology: General, among others. His
research has been recognized by awards including the European Science Foundation’s
European Young Investigator Award, the German Science Foundation’s Gottfried-Wilhelm
Leibniz award, and the European Association of Social Psychology’s Jos Jaspars award.

Chadee_flast.indd xiChadee_flast.indd xi 12/13/2010 2:13:12 PM12/13/2010 2:13:12 PM

xii Contributors

Previously Mussweiler taught at the University of Würzburg, Germany, was a postdoctoral
fellow at Northwestern University, USA, and received his Ph.D. from the University of Trier,
Germany.

Paul R. Nail has been Associate Professor of Psychology at the University of Central Arkansas
since 2005 and holds a Ph.D. in experimental psychology from Texas Christian University
(1981). Previously, he was a Professor at Southwestern Oklahoma State University for 25
years. He has published articles in the Psychological Bulletin, the Journal of Personality and
Social Psychology, the Personality and Social Psychology Bulletin, and Basic and Applied Social
Psychology, among others. Besides cognitive dissonance theory, his research interests include
social influence in groups, contemporary theories of racism, political psychology, and indi-
vidual differences in psychological defensiveness. He is currently an Associate Editor of the
interdisciplinary journal Social Influence.

Richard E. Petty is Distinguished University Professor of Psychology at Ohio State University.
He received his BA from the University of Virginia and his Ph.D. from Ohio State. His
research focuses on understanding changes in attitudes and behaviors. He has published
eight books and over 250 articles and chapters. Honors received include the Scientific
Impact Award from the Society of Experimental Social Psychology, Distinguished Scientific
Contribution Awards from the Society for Personality and Social Psychology (SPSP) and
the Society for Consumer Psychology, and service as President of SPSP and the Midwestern
Psychological Association. He is past Editor of the Personality and Social Psychology
Bulletin.

Shani Pindek is a doctoral student in the Psychology Department of Bar-Ilan University,
with particular interests in impression management, organizational citizenship behavior,
and altruism.

Denise M. Polk is Associate Professor in the Department of Communication Studies at
West Chester University of Pennsylvania. Her teaching and research areas focus mainly in
the areas of interpersonal communication, work–family integration, conflict resolution,
and health communication. She has focused mostly on romantic and family relationships.
She holds a BA from Baldwin-Wallace College, a Master’s from Miami University, and a
Ph.D. from Kent State University.

Ann C. Rumble is Assistant Professor of Psychology at Ohio University-Chillicothe. She
received her Ph.D. from Washington State University in 2003, with a specialization in inter-
dependent behavior and social dilemmas. She continued her work with Marilynn Brewer at
Ohio State University as a National Institute of Mental Health postdoctoral fellow, following
which she accepted a position at Ohio University-Chillicothe.

Katharina Schmid is currently a Research Fellow and Lecturer in Psychology at New College,
University of Oxford. She previously held appointments at the Max-Planck-Institute for the
study of ethnic and religious diversity in Germany, and at Royal Holloway University of

Chadee_flast.indd xiiChadee_flast.indd xii 12/13/2010 2:13:12 PM12/13/2010 2:13:12 PM

Contributors xiii

London, and completed her Ph.D. in Social Psychology at Queen’s University Belfast. Her
research interests lie broadly in the areas of social identity and intergroup relations. Her
research focuses in particular on the role of multiple categorization processes, and espe-
cially social identity complexity, in intergroup relations.

Andreas Schneider is Associate Professor of Sociology at Texas Tech University. While
study ing social psychology, sociology, management, and marketing at Mannheim University,
Germany, he visited Indiana University in 1987 and became fascinated with the new emerg-
ing cybernetic symbolic interactionist model of Affect Control Theory. He received his
Ph.D. in sociology from Indiana University, Bloomington, in 1997. His publications in
Social Psychology Quarterly, the Journal of Mathematical Sociology, Quality & Quantity, and
Organization Studies demonstrate his research interests in the methodological development
of symbolic interactionism and the cross-cultural application in the fields of deviance, sex-
uality, and management.

Benjamin C. Wagner is a doctoral candidate in social psychology at Ohio State University,
studying under Richard E. Petty. His research interests include attitudes and persuasion,
embodied cognition, metacognition, moral reasoning, and emotion. In 2007, he earned his
Master’s degree in psychology from Ohio State, having completed his BA two years earlier
at Denison University, Granville, Ohio.

Chadee_flast.indd xiiiChadee_flast.indd xiii 12/13/2010 2:13:12 PM12/13/2010 2:13:12 PM

This book would have been impossible were it not for the assistance of the following persons
and institutions. Thanks to the staff at Wiley-Blackwell, in particular, Andrew McAleer,
Annie Rose, Karen Shield, and Juanita Bullough, for their efficiency and help at every
stage. To all the contributors, I am grateful for your contributions. Specifically I express my
gratitude to Ananthi Al Ramiah, Kurt Boniecki, Jenny Carrillo, Katja Corcoran, Alexandra
Corning, Richard Crisp, Faye Crosby, Jan Crusius, Tara Dennehy, Miles Hewstone, Meni
Koslowsky, Angela Maitner, Bertram Malle, Thomas Mussweiler, Paul Nail, Richard Petty,
Shani Pindek, Denise Polk, Ann Rumble, Katharina Schmid, Andreas Schneider, and
Benjamin Wagner. I am grateful for and acknowledge the comments by the publisher’s
anonymous reviewers, which were instructively helpful in the preparation of the final
manuscript.

I would like to express my gratitude to Professor Ramesh Deosaran for his valuable
mentorship, and to the University of the West Indies for its support, particularly the ANSA
McAL Psychological Research Centre, of whose academic output this publication is a part.
I am grateful to and thank everyone who provided the necessary technical and other
support leading up to final publication, and wish to express my deepest gratitude for the
assistance of anyone whom I may have inadvertently left unacknowledged.

Acknowledgments

Chadee_flast.indd xivChadee_flast.indd xiv 12/13/2010 2:13:12 PM12/13/2010 2:13:12 PM

Gordon Allport (1968) defined social psychology as “an attempt to understand and explain
how the thoughts, feelings and behavior of individuals are influenced by the actual, imag-
ined or implied presence of others.” As limited as definitions are, this definition of social
psychology captured the dynamism, focus, and direction of the discipline. Important to an
understanding of social psychological behavior is taking into consideration not only what
is happening socially to the person but also what is occurring internally and cognitively to
the individual which, in turn, affects social behavior. From its genesis rooted in the work of
William James’s Principles of Psychology to current development of the discipline, there has
always been an emphasis on the individual within the social interaction paradigm.
Theorization, therefore, within the discipline has fallen within this paradigm, which is now
extended to include the neurological functioning of human beings within the social psy-
chological context.

The early works on social psychology by the psychologist William McDougall (1908) and
the sociologist Edward Ross (1908) weighted social behavior on instinctual or social factors,
respectively. Later, Floyd Allport (1924) emphasized a behaviorist stimulus–response para-
digm for the understanding of social psychological behavior. Theories of psychology and
sociology during this early period seem to have been competing to understand a realm that
had neither the theorization nor the research sophistication to claim discovery status. Much
of the work undertaken in social psychology has been done within the discipline of psy-
chology, with sociological social psychology contributions being relatively sparse. Notably,
the discipline of sociology has contributed tremendously to the early development of the
concept and theorization of self, especially via theories of symbolic interactionism, phe-
nomenology, and, later, ethnomethodology. On the other hand, psychological social psy-
chology’s contributions have been crucial to the genesis and development of both the pure
and applied branches of the discipline.

Introduction

Theories in Social Psychology, First Edition. Edited by Derek Chadee.
© 2011 Blackwell Publishing Ltd. Published 2011 by Blackwell Publishing Ltd.

Chadee_intro.indd 1Chadee_intro.indd 1 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

2 Introduction

Social psychology has seen numerous studies testing hypotheses drawn from concepts or
theories. However, less frequent in the literature is the emergence of theories – a renaissance
that is much needed for the development and impetus of the discipline. However, many of
the theories that currently exist within social psychology are as important to the discipline
as they were over forty years ago. A renaissance starts with a reassessment of the efficacy of
current theories.

Theories have the power of insight and understanding, allowing scientists to see phe-
nomena that previously they would have been unable to conceptualize. Albert Einstein
is quoted as saying, “Whether or not you can observe a thing depends on the theory you
use. It is the theory which decides what can be observed.” The assumptions, proposi-
tions, hypotheses, and supporting confirmed “facts” all provide the theory with a power
of vision. The assumptions are givens that the theorist is allowed to utilize in the con-
struction of his theory. They are like the basic tools of a carpenter. These assumptions
are grounded on some philosophy or principle, and are often not the bases on which a
theory is criticized. However, assumptions give a theory direction. On the other hand, a
theory’s power of vision is myopic, limited by the same tools that give the theory its
power. This is a dilemma that the finite scientist must explore in any discipline. However,
the beach-ball approach to the understanding of the world within one’s discipline is a
compromise accepted by social psychologists. That is, it is recognized that a theory is an
academic creation, in this sense, and has limitations. The adoption of a number of theo-
retical positions provides a more comprehensive understanding of the multidimensional
nature of a phenomenon.

A theory can be criticized on a number of grounds, including consistency – how logically
well the theory holds together with its propositions and hypotheses; external – standing up
to criticisms of other theories (theory A vs. theory B); historical – temporal perseverance of
the theory (e.g., does Freud’s psychoanalytical theory or Heider’s attribution theory still
hold today?); applicability – the generalizing of findings from research to social situations;
and methodological – the strengths and weakness of the methodology used in the construc-
tion of the theory, especially if the theory is an empiricist one.

The recent social psychological literature has been lacking a volume systematically dedi-
cated to a range of theories within the discipline. The emphasis of this book, therefore, is on
social psychological theories, with an evaluation of some of the main theories still discussed
and relevant to understanding behavior. The volume is divided into four parts. Part I
presents critical assessments of social cognitive theories – from their genesis to their current
development.

Derek Chadee revisits Brehm’s theory of psychological reactance, identifying the genesis
of this theory in the womb of cognitive dissonance theory. However, the baby grew with
many different characteristics from the mother. Both theories of cognitive dissonance and
psychological reactance are theories of motivational arousal and reduction. The theory of
psychological reactance, however, attempts to explain people’s reactions to perceived or
actual threat to loss of freedom. The theory builds on several assumptions of human behav-
ior, with a major underlying assumption of human persistence in maintaining free behav-
iors and the consequences that arise as a result of threats to importantly defined free
behaviors. The early emphasis of reactance theorizing and research was on psychological

Chadee_intro.indd 2Chadee_intro.indd 2 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

Introduction 3

reactance as being aroused by the situation. However, later studies have given emphasis to
reactance as dispositional – a reactant personality. Numerous instruments have been
designed to measure reactance as a disposition. The merit and demerits of these measures
are discussed.

This first chapter critically assesses the relationship between reactance and dissonance,
proposing reactance as a special case of dissonance though identifying the distinctness of
reactance. Critical to this chapter is the identification of the systematic void in the literature
concerning any discussion on affect in reactance. The last part of the chapter evaluates
the relationship between affect and reactance, proposing a reactance emotion theory.
Chadee concludes by noting that reactance theory is an important theory in the discipline
of social psychology and has contributed to an understanding of reactance behavior to
actual or perceived threats in a wide variety of settings. The theory is as useful today as it
was over forty years ago. However, the theory needs modification.

Paul Nail and Kurt Boniecki critically assess Festinger’s (1957) cognitive dissonance and
related theories, from dissonance theory’s inception and rise to prominence in the 1950s,
1960s, and 1970s, to its near-death by the early 1980s, to its rebirth in the 1990s and 2000s.
The original statement of the theory was stunning in its simplicity – that an unpleasant
psychological state called cognitive dissonance exists whenever one important thought is in
conflict with another. Yet this very simplicity was a major factor that has led to a large
number of academic debates and controversies over the years. Nail and Boniecki describe
how the germ of the theory got started when Festinger read of the events surrounding a
massive earthquake that occurred in India in 1934.

They review Festinger’s major theoretical constructs and how these gave rise, early on,
to a series of counterintuitive predictions that were generally supported by the empirical
evidence. Most famous is the finding of Festinger and Carlsmith (1959) that research par-
ticipants reported a greater liking for boring laboratory tasks if they were paid $1 as
opposed to $20 for performing them. Nail and Boniecki summarize major paradigms
that were created to test dissonance theory, followed by rival theoretical accounts, such as
Bem’s (1967) self-perception theory. The success of these accounts eventually caused inter-
est in the theory to wane, but more recently has led to a flurry of empirical interest in dis-
sonance and related phenomena, reviewed by Nail and Boniecki, which persists to the
present day. One major focus of the chapter are the “self ” theories of dissonance (e.g.,
Aronson, 1968, 2007; Steele, 1988), which hold that dissonance processes have their origins
in the need for self-esteem rather than in a need for logic-like consistency, as originally
conceived by Festinger (1957). Another focus is individual differences, e.g., how the differ-
ent versions of dissonance theory make conflicting predictions for those varying in self-
esteem. The chapter evaluates the theory in terms of its applicability, efficiency, heuristic
value, and originality. Nail and Boniecki close by calling for new research that could
possibly reestablish Festinger’s (1957) version as the single most adequate account of dis-
sonance phenomena.

Bertram Malle examines the history of research on behavior explanations, identifies
missing pieces, and introduces a theoretical model that is meant to account for explana-
tions at the conceptual, psychological, and linguistic levels. Heider (1958) was the first to
examine systematically how people make sense of each others’ behavior. He introduced the

Chadee_intro.indd 3Chadee_intro.indd 3 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

4 Introduction

notion of personal causality – ordinary people’s conception of how purposeful behavior
works. When a behavior obeys personal causality, it is seen as caused by the agent’s inten-
tion, whereas such an intention is absent in behavior that obeys impersonal causality. Thus,
Heider captured what was later called intentionality – a core distinction in people’s under-
standing of human behavior.

Subsequent attribution research turned to different directions. Jones and Davis
(1965) shifted from considerations of intention inferences toward considerations of
personality and attitude inferences. Kelley (1967) set aside Heider’s distinction between
personal and impersonal causality and instead focused on a distinction between person
(internal) and situation (external) causes. This internal–external dimension, however,
applies only to people’s explanations of unintentional events, not to their explanations
of intentional action.

Decades passed before Heider’s original concern with intentionality and people’s
inferences of motives and reasons became a topic of research again. Committed to these
concerns, the second part of the chapter introduces the folk-conceptual theory of behav-
ior explanation (Malle, 1999, 2004). It locates explanations in the network of folk con-
cepts people use to make sense of human behavior and specifies the psychological
processes and linguistic manifestations of explanations. For example, people offer very
different kinds of explanations for intentional and unintentional behavior. Unintentional
behavior is explained by causes, which can be classified in a variety of ways, including an
internal–external dimension. Intentional behavior, by contrast, is more complex. People
offer either reason explanations (referring to the beliefs and desires in light of which the
agent formed an intention to act) or causal history of reason explanations (referring to
factors that led to those reasons in the first place – upbringing, personality, unconscious
mental states, etc.). People’s choice between these two explanation modes reflects both
cognitive and motivational processes and is sensitive to the explainer’s role (actors vs.
observers), the type of agent (group vs. individuals), and the explainer’s impression-
management goals.

Thus, the folk-conceptual theory tries to carve out the concepts and processes that mat-
ter when people construct and respond to explanations; and these distinctions reveal a rich,
sophisticated system of folk-behavior explanations. Malle’s chapter is a dynamic contribu-
tion to the evolution of the field of attribution.

Benjamin Wagner and Richard Petty examine the Elaboration Likelihood Model (ELM),
which is a general theory of persuasion that is also applicable to social judgment (Petty &
Cacioppo, 1986). The ELM holds that people’s attitudes can be modified in both relatively
effortful (central route) and non-effortful (peripheral route) ways. Persuasion via the cen-
tral route depends on the strength and cogency of the arguments contained in the message
and is determined by the number, valence, and confidence people have in their thoughts to
the advocacy. Persuasion via the peripheral route is determined largely by reliance on sim-
ple cues and heuristics that are not necessarily central to the merits of the advocacy (e.g.,
being in a good mood). The extent of elaboration is the chief determinant of the route to
persuasion, and several factors have been shown to influence the extent of thinking. Broadly
speaking, these factors relate to motivation (e.g., personal relevance of the topic) and ability
(e.g., knowledge about the topic) to think about the advocacy. With greater motivation and

Chadee_intro.indd 4Chadee_intro.indd 4 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

Introduction 5

ability comes an increased likelihood that a person will extensively elaborate a persuasive
message and be less reliant on issue-irrelevant cues. Importantly, attitudes formed under
the central route are generally more durable and impactful than are attitudes formed under
the peripheral route.

An important component of the ELM is the idea that any variable can serve multiple
roles in persuasion. Specifically, persuasion variables such as source expertise or momen-
tary emotions can operate in five roles: (a) as simple cues to persuasion (e.g., “experts are
usually right”), (b) as arguments that are relevant to a particular conclusion (e.g., “an
expert’s endorsement speaks to the product’s quality”), (c) by biasing the thoughts that a
person generates (i.e., expert sources lead to more favorable thoughts concerning the
message than non-expert sources), (d) by affecting the amount of thinking a person does
(i.e., expert sources enhance thinking about the arguments since they seem more wor-
thy), and (e) by affecting whether people use their thoughts in response to the message
in forming their attitudes toward the topic (i.e., “if my thoughts were provoked by an
expert, they must be valid”).

The ELM has not only integrated research in persuasion but has also been applied in
diverse areas of application such as consumer attitudes, health promotion, and legal
domains. The model continues to generate interesting and important research findings,
both at the basic and applied levels.

Social comparison has evolved and been modified since the 1950s. The social compari-
son process is a process that is present in numerous theories. Relative deprivation assumes
the presence of social comparison. Both social comparison and relative deprivation are
discussed in Part II.

Katja Corcoran, Jan Crusius, and Thomas Mussweiler provide a comprehensive over-
view of social comparison theory. Social comparisons – comparisons between the self
and others – are a fundamental psychological mechanism influencing people’s judg-
ments, experiences, and behavior. In this chapter, they review the social psychological
theorizing and research on social comparison. They start by summarizing the basic
tenets of Leon Festinger’s seminal social comparison theory and continue by addressing
three questions that are central to Festinger’s ideas and to the research that followed his
initial work. The first question is: Why do people engage in social comparisons? While
social comparison is mostly understood as a process which is engaged to fulfill funda-
mental needs like self-evaluation, self-enhancement, and self-improvement, the chapter
discusses logical reasons for social comparisons and considers the efficiency advantage
of comparative information-processing. The second question is: To whom do people
compare themselves? The chapter explores how motivational concerns influence the
selection of comparison standards and how routine standards can provide an efficient
means to fulfill the need to self-evaluate. The third question is: How do social compari-
sons influence the self? The diverse factors that lead to assimilation or contrast of the self
as a consequence of social comparison are reviewed. Furthermore, these factors are dis-
cussed in light of the Selective Accessibility Model, which explains them by the changes
of accessible self-knowledge during social comparisons. Finally, Corcoran et al. address
the role of social comparisons in health psychology and the impact of idealized media
images on self-evaluation as applied examples of social comparison research. Their

Chadee_intro.indd 5Chadee_intro.indd 5 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

6 Introduction

review shows that comparative processes are marked by striking complexity and multi-
facetedness and that consequences of social comparisons span all core areas of human
psychological functioning. They argue that future research could benefit from a perspec-
tive that integrates cognitive, motivational, and affective determinants and consequences
of social comparisons.

Underlying the theory of relative deprivation is the social comparison process. Relative
deprivation (RD) theory helps social scientists predict who will become dissatisfied under
what conditions. RD theory is a theory of perceived social inequity which helps to explain
why some people with paltry resources experience contentment while some others with
abundant access to a wealth of resources are dissatisfied. Jenny Carrillo, Alexandra Corning,
Tara Dennehy, and Faye Crosby review the history, development, and utility of RD. After
detailing the initial writings about the concept of RD, they trace the development of various
models of RD articulated from the 1960s to the 1980s and assess the more contemporary
work on RD, focusing on the distinction between personal and group-based feelings of
relative deprivation and on a validated means of measuring the construct that was pre-
sented in 2000. Directions for the future application of RD theory to social problems are
considered.

Part III assesses two theories that start with the behavioral premise that human beings
are hedonistic. The theories also assume that actions are governed by a reinforcement–
punishment structure which extends into interpersonal interaction. The power dynamics
of interpersonal interaction is absent from equity theory but elaborated in interdependency
theory.

Denise Polk’s chapter provides a useful understanding of equity theory. The theory
stems from principles of reinforcement and basic principles of economics. The basic
premise of the theory is that people evaluate their relationships in terms of inputs and
outcomes. The principle of distributive justice is core to equity theory (Deutsch, 1985).
Imbalances in input–outcome ratios result in inequity. Two types of inequity can occur.
People may be underbenefited, or they may be overbenefited. However, according to
equity theory, people are driven to restore equity once they perceive inequity. When peo-
ple experience inequity, they may attempt to restore actual equity or psychological equity.
Polk posits that because no magic formula for equity exists, relational partners must
determine equity for themselves. Equity is a key consideration in relationships because
people’s perceptions about equity shape people’s feelings, decisions, and actions toward
their relational partners (Adams, 1965; Sprecher & Schwartz, 1994), so equity theory is
appropriate to help explain the development, maintenance, and dissolution of relation-
ships. Polk explores research which has tested equity for links with many variables
including personality, emotions, gender, the distribution of domestic duties, and rela-
tional quality.

The chapter by Ann Rumble explores interdependence theory, as developed by John
Thibaut and Harold Kelley. Thibaut and Kelley employed outcome matrices in order to
understand an actor’s available behavior choices and outcomes. The given matrix repre-
sents the choices and outcomes that are available to the actors in a specific situation, and
through the transformation process develops into the effective matrix (Thibaut & Kelley,
1959). By examining the components of the given and effective matrix, we will be

Chadee_intro.indd 6Chadee_intro.indd 6 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

Introduction 7

able to explain human behavior in a number of interdependent situations, including
close relationships.

The final part of the book is devoted to theories related to self and identity and covers
social identity, social categorization, symbolic interactionism, and impression manage-
ment. Katharina Schmid, Miles Hewstone, and Ananthi Al Ramiah’s chapter provides a
general overview of social psychological theory on social identity, including social iden-
tity theory (Tajfel, 1978; Tajfel & Turner, 1979) and self-categorization theory (Turner,
Hogg, Oakes, Reicher, & Wetherell, 1987). However, this chapter takes a somewhat more
encompassing approach to portraying what is commonly referred to as the social identity
perspective (Abrams & Hogg, in press). The authors thus not only focus on social iden-
tity theory and self-categorization theory but also pay particular attention to defining the
concept of social identity, and consider the consequences of social identity phenomena
for intergroup relations, above and beyond the predictions of social identity theory,
taking into consideration the extent to which multiple categorization processes help
explain intergroup relations.

The chapter is organized into four sections, the first two of which conceptually define
self-categorization and social identification, whereas the last two examine the extent to
which self-categorization and social identification are involved in, and help explain,
intergroup relations. Section one of this chapter provides a brief definition of self-
categorization and describes the theoretical foundations of self-categorization theory.
The authors describe in detail what is meant by self-categorization, and how, why, and
when individuals make use of social categories. In Section two, they define social iden-
tification as a psychological process associated with group membership and explain that
identification is a multidimensional phenomenon. Section three presents a detailed
description of social identity theory, with reference to the findings that emerged from
the minimal group paradigm. Schmid, Hewstone, and Al Ramiah address, with refer-
ence to the predictions of social identity theory, the extent to which self-categorization
and social identification are related to intergroup attitudes and behavior. Finally, Section
four is devoted to an overview of multiple categorization and its consequences for inter-
group relations, showing how more complex consideration of others and oneself in
terms of multiple group memberships is associated with tolerance and improved inter-
group relations.

Richard Crisp and Angela Maitner’s chapter complements the previous chapter. They
argue that in contemporary society the traditional boundaries that have previously defined
social group memberships are being steadily eroded and replaced with more complex con-
ceptualizations of identity. Crisp and Maitner review classic and contemporary theories of
social categorization in the context of this increasing social and cultural diversity. They
argue that broad-ranging and pervasive changes to the categorical structure of society have
fundamental implications for how individuals perceive, represent, and understand their
social environments. They review existing social cognitive, self-categorization, and situ-
ated cognition accounts, arguing that an increasing focus on the context-specific nature of
social categorization reflects the more fluid and fluctuating nature of identity in contem-
porary society. They conceptualize a diversity-driven social categorization theory, arguing
that the functional nature of human cognition implies that exposure to diversity must

Chadee_intro.indd 7Chadee_intro.indd 7 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

8 Introduction

change how individuals psychologically engage with their social worlds. The authors con-
clude that to understand fully the evolving nature of social categorization psychologists
should seek to incorporate a broader multidisciplinary analysis of the changing nature of
culture and society.

Andreas Schneider’s chapter is a comprehensive overview of symbolic interactionism
(SI). Today the framework of symbolic interactionism has been delineated into many theo-
retical approaches using qualitative and quantitative methods of investigation that are
applied in numerous areas of research. SI has evolved a long way from early philosophies of
North American pragmatism to the computer simulation of human interaction. Describing
this path, this chapter overcomes dichotomies such as the Chicago school versus the Iowa
school, or quantitative versus qualitative, that have been used in the past to pigeonhole
one or the other line of research. Instead it portrays the development from the grandfathers
and their philosophical backgrounds to explain the different agendas addressed by the
founding fathers of the Chicago school. This historical context is necessary to understand
contemporary contrasting schools of thought and their roots in social psychology, as well
as sociology. This chapter then shows how ideas and methodologies of these qualitative and
quantitative approaches are integrated into the most recent development of cybernetic
control models in SI. Finally it is demonstrated how SI is applied in the fields of deviance,
sexuality, children, gender, emotions, organization/management, cross-cultural compari-
son, and ethnomethodology/conversation analysis. Descriptions of these applications are
supported by interviews of key researchers in the respective fields.

Meni Koslowsky and Shani Pindek’s chapter on impression management (IM) is a
refreshing contribution to the literature. They note that IM is an activity which takes place
in many, if not most, interactions between people. In this chapter they start by exploring the
different definitions of the construct, from a narrow view of IM as a set of manipulative
behaviors, performed mainly in order to present them in a positive light, to a more expan-
sive definition which assumes that all people unconsciously manage their impressions in
ways that assist in achieving goals both at the individual and group level. This expansive
view of IM allows us to deal with its association to constructs such as the self-concept, indi-
viduals’ social identities, and other social phenomena.

When applying or measuring IM, behaviors are usually considered as belonging to one
of several distinct subcategories. These categories include verbal/nonverbal behaviors,
defensive/promotional, positive/negative, and several other related taxonomies. The chap-
ter also explores specific antecedents and outcomes of IM. Prominent among the former
are gender, self-monitoring, and self-regulation, as well as demographic and personality
variables. In addition, research on situational antecedents and on IM outcomes has been
conducted in applied settings such as human resource management. Since the work envi-
ronment supplies the individual with many incentives as well as opportunities to benefit
from impressions that are well managed, this area has been the focus of much of the research
in recent years.

Finally, they identify several areas for future researchers to consider so as to better explain
the phenomenon.

Chadee_intro.indd 8Chadee_intro.indd 8 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

Introduction 9

References

Abrams, D., & Hogg, M. (in press). Social identity and self-

categorization. In J. F. Dovidio, M. Hewstone, P. Glick, &

V. M. Esses (Eds.), Handbook of prejudice, stereotyping

and discrimination. London: Sage.

Adams, J. S. (1965). Inequity in social exchange. In

L. Berkowitz (Ed.), Advances in experimental social psy-

chology (Vol. 2, pp. 267–300). New York: Academic

Press.

Allport, F. H. (1924). Social psychology. New York: Houghton

Mifflin.

Allport, G. (1968). The historical background of modern

social psychology. In G. Lindzey and E. Aronson (Eds.),

The handbook of social psychology (Vol. 1). Reading, MA:

Addison-Wesley.

Aronson, E. (1968). Dissonance theory: Progress and prob-

lems. In R. P. Abelson, E. Aronson, W. J. McGuire, T. M.

Newcomb, M. J. Rosenberg, & P. H. Tannenbaum (Eds.),

Theories of cognitive consistency: A source book (pp. 5–27).

Skokie, IL: Rand McNally.

Aronson, E. (2007). The evolution of cognitive dissonance

theory. In A. R. Pratkanis (Ed.), The science of social influ-

ence: Advances and future progress (pp. 115–135).

New York: Psychology Press.

Bem, D. J. (1967). Self-perception: An alternative interpreta-

tion of cognitive dissonance phenomena. Psychological

Review, 74, 183–200.

Deutsch, M. (1985). Distributive justice: A social-psycholog-

ical perspective. New Haven, CT: Yale University Press.

Festinger, L. (1957). A theory of cognitive dissonance. Stanford,

CA: Stanford University Press.

Festinger, L., & Carlsmith, J. M. (1959). Cognitive conse-

quences of forced compliance. Journal of Abnormal and

Social Psychology, 58, 203–210.

Heider, F. (1958). The psychology of interpersonal relations.

New York: Wiley.

Jones, E. E., & Davis, K. E. (1965). From acts to dispositions:

The attribution process in person perception. In

L. Berkowitz (Ed.), Advances in experimental social psy-

chology (Vol. 2, pp. 219–266). New York: Academic Press.

Kelley, H. H. (1967). Attribution theory in social psychol-

ogy. In D. Levine (Ed.), Nebraska symposium on

motivation (Vol. 15, pp. 192–238). Lincoln: University of

Nebraska Press.

Malle, B. F. (1999). How people explain behavior: A new

theoretical framework. Personality and Social Psychology

Review, 3, 21–43.

Malle, B. F. (2004). How the mind explains behavior: Folk

explanations, meaning, and social interaction. Cambridge,

MA: MIT Press.

McDougall, W. (1908). Introduction to social psychology.

London: Methuen.

Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration

Likelihood Model of persuasion. In L. Berkowitz (Ed.),

Advances in experimental social psychology (Vol. 19,

pp. 123–205). New York: Academic Press.

Ross, E. A. (1908). Social psychology. New York: Macmillan.

Sprecher, S., & Schwartz, P. (1994). Equity and balance in

the exchange of contributions in close relationships. In

M. L. Lerner & G. Mikula (Eds.), Entitlement and the

affectional bond: Justice in close relationship (pp. 11–41).

New York: Plenum.

Steele, C. M. (1988). The psychology of self-affirmation:

Sustaining the integrity of the self. In L. Berkowitz (Ed.),

Advances in experimental social psychology (Vol. 21,

pp. 261–302). San Diego, CA: Academic Press.

Tajfel, H. (1978). Interindividual behaviour and intergroup

behaviour. In H. Tajfel (Ed.), Differentiation between

social groups: Studies in the social psychology of intergroup

relations (pp. 28–60). London: Academic Press.

Tajfel, H., & Turner, J. C. (1979). An integrative theory of

intergroup conflict. In W. G. Austin & S. Worchel (Eds.),

The psychology of intergroup relations (pp. 33–48).

Monterey, CA: Brooks/Cole.

Thibaut, J. W., & Kelley, H. H. (1959). The social psychology

of groups. New Brunswick, NJ: Transaction Books.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., &

Wetherell, M. S. (1987). Rediscovering the social group:

A self-categorization theory. Cambridge, MA: Blackwell.

Chadee_intro.indd 9Chadee_intro.indd 9 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

Chadee_intro.indd 10Chadee_intro.indd 10 12/13/2010 2:13:46 PM12/13/2010 2:13:46 PM

Part I

Social Cognition

Chadee_p01.indd 11Chadee_p01.indd 11 12/13/2010 2:14:12 PM12/13/2010 2:14:12 PM

Chadee_p01.indd 12Chadee_p01.indd 12 12/13/2010 2:14:12 PM12/13/2010 2:14:12 PM

Theories in Social Psychology, First Edition. Edited by Derek Chadee.
© 2011 Blackwell Publishing Ltd. Published 2011 by Blackwell Publishing Ltd.

Reactance theory was developed by Jack W. Brehm and articulated in his Theory of
Psychological Reactance (1966). Leon Festinger and Stanley Schachter were the editors of the
monograph series which included Brehm’s seminal work. In fact, Brehm’s Ph.D. supervisor
was Leon Festinger and his Ph.D. dissertation tested the free-choice dissonance paradigm
which later appeared in 1956 in the Journal of Abnormal and Social Psychology as the first
empirical research on cognitive dissonance. Later, Brehm systematically wrote on cognitive
dissonance (e.g., Brehm, 1962; Brehm & Cohen, 1962).

The genesis of the theory of psychological reactance appears to have taken place in the womb
of cognitive dissonance theory. However, the baby grew with many different characteristics
from the mother. Both theories of cognitive dissonance and psychological reactance are theories
of motivational arousal and reduction. In fact, Miron and Brehm (2006, p. 9) recognize that

Festinger had constructed a theory (cognitive dissonance) that assumed an inner motivational

process rather than assuming that all influences between stimuli and behavior were simple and

direct. It was in this context that Brehm and Cohen (1962), both in the Yale Attitude Change

Program at the time, carried out an extensive program of research on persuasion, largely based

on dissonance theory, but with some attention to special cases of resistance to social influence.

After [Brehm] … became more interested in the occurrence of resistance to social influence,

and that interest eventuated in the formulation of reactance theory.

Assumptions

The theory of psychological reactance attempts to explain people’s reactions to a perceived
or actual threat to loss of their freedom(s). The concept of freedom is defined “as a belief
that one can engage in a particular behavior” (Brehm & Brehm, 1981, p. 35). Brehm
(1966, p. 9) defines “psychological reactance as a motivational state directed toward the

1

Toward Freedom: Reactance
Theory Revisited

Derek Chadee

Chadee_c01.indd 13Chadee_c01.indd 13 12/13/2010 2:16:50 PM12/13/2010 2:16:50 PM

14 Derek Chadee

reestablishment of the free behaviors which have been eliminated or threatened with
elimination.” The theory makes no assumption about psychological reactance being
aroused to acquire a freedom but refers solely to reinstating a threatened or eliminated
freedom (Brehm & Brehm, 1981). Psychological reactance, therefore, arouses an individual
and motivates toward restoration of freedom. As Miron and Brehm (2006) point out, reac-
tance is not a proactive force but is reactive to actual or perceived threats toward freedoms.

The theory makes a number of assumptions about human behavior. Brehm (1966, p. 1)
notes the following about human behavior: First, freedom of behavior is pervasive. Second,
it is an important aspect of human interaction and living. Third, human beings are goal-
oriented and incessantly reflect on themselves and the external environment in assessing
what, how, and when they will undertake particular behaviors. He states: “They consider
their wants and needs, the dangers and benefits available in their surroundings, and the
ways in which they can accomplish various ends” (1966, p. 1). Many times people undertake
behaviors when they are not fully aware of the rationale that drives the behaviors. At other
times behaviors are performed while they are fully aware of constraints and lack of freedom
in the performance of the behaviors. As a fourth assumption, however, Brehm assumes that
most of the time human beings feel relatively free to participate in a range of behaviors.
Fifth, the theory assumes that people have a set of behaviors to engage in. These behaviors
have been engaged in in the past, are engaged in in the present, and will be engaged in in the
future. He refers to these behaviors as the individual’s “free behaviors” and identifies a
number of criteria to define a behavior as a free behavior. These include the behaviors being
practically achievable, the individual having the physical and psychological capacity to
engage in these behaviors, and individuals also being fully cognizant that they can partici-
pate in these behaviors. Such awareness of engagement emerges from social norms, cus-
toms, legislations, and other kinds of formal agreement and informal interaction. Worchel
(2004) elaborates on how particular actions over time are perceived as free behavior. For
reactance to occur an individual must believe in the possession (perceived or actual) of a
freedom (Miron & Brehm, 2006). A threat to this freedom creates the arousal – reactance.

Threat

A threat emerges from any power or force that attempts to reduce or eliminate the expres-
sion of a specific freedom (Brehm & Brehm, 1981). Threats can be explicit or implicit (i.e.,
subtle by implication). Brehm and Brehm (1981) describe threats to freedom as either
external or internal. Internal threats arise out of choices and decisions we make from alter-
natives, accepting and rejecting among options. On the other hand, external threats have
two dimensions – impersonal versus personal and social versus nonsocial. Impersonal
threats are not easily perceived, unlike personal threats in which motives and intentions
of threats can be identified by the individual and have implications for perception of future
threats. Specifically, personal threats will carry greater implications for future threats from
the same source because of possible future interaction. Impersonal threats, however, create
less reactance arousal, since they are not directly focused at the individual, and therefore
no implications for future threats. However, the term personal, as Brehm and Brehm (1981)

Chadee_c01.indd 14Chadee_c01.indd 14 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 15

indicate not only refers to interaction with known others but can apply to infrequent inter-
action with an entity/person.

Impersonal threats can also be derived from persons or institutions that we interact with
on a personal level, for example, working for a large corporation or governmental organiza-
tion. An impersonal threat from such an organization is not directed toward a specific
individual, but any threat to freedom would have consequences to individuals in the organiza-
tion. The depth of future interaction with a threatening source is an important prediction of
psychological reactance. Brehm and Brehm (1981) postulate that anticipation of future inter-
action with a threatening source is likely to increase psychological reactance as compared to
anticipation that there will be minimal or no future interaction with a threatened source.

Another dimension of threat which Brehm and Brehm (1981) identify is the social versus
nonsocial dimension. However, they also hold the view that nonsocial exists in degrees,
since humans are social and reactance emerges out of social interaction, and there will be
social interpretations of nonsocial dimensions. The source of threat can emanate from
either level. Studies that consider the social dimension expose subjects to a socially threat-
ening situation (e.g., one individual directly threatens the freedom of another). The non-
social studies created threats via nonsocial interaction (e.g., barriers to toys, removal of
choice alternatives).

Magnitude of Reactance

Psychological reactance is influenced by both threat and freedom (Wright, Greenberg, &
Brehm, 2004). The theory postulates that there are a number of factors which influence the
magnitude of psychological reactance. These include the strength of the threat to one’s free-
dom; the presence of freedom which emerges from the interruption of free behaviors or
barriers preventing their expression; the importance of freedom in the realization of needs;
and the extent to which the needs are core to the individual’s existence. There is a direct
relationship between reactance and the importance of the threatened. For example, being
forced to choose between two valueless alternatives will not create as high a psychological
reactance as being forced to choose between attractive alternatives. The relative importance
of the freedom, and not only the importance, is a determining factor of reactance. For exam-
ple, if choice alternatives are of equal cost (e.g., the choice alternative of dining at similar
kinds of restaurants), removal of one alternative or the other will lead to relatively the same
degree of reactance. However, in the same example, if the choice alternative included a high-
end computer and this alternative was denied, the magnitude of psychological reactance
would be relatively much higher because of the relatively higher attractiveness of the com-
puter. The proportion of freedom threatened also impacts on the magnitude of reactance
and arises from the number of free behaviors that are at risk in proportion to the number of
free behaviors. Threats that have implications for future threats to free behavior also arouse
reactance (Brehm, 1966; Brehm & Brehm, 1981; Miron & Brehm, 2006; Wicklund, 1974).

Threats may be defined or arbitrary. Threats that are defined and specific have limits to
the threats and are justified (e.g., the denial of entry to any vehicles other than taxis into a
particular roadway). This threat has a physical limit and a rationale which allows for

Chadee_c01.indd 15Chadee_c01.indd 15 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

16 Derek Chadee

reduction of traffic congestion. Arbitrary threats are ill-defined (e.g., an office imposing
restrictions on casual conversations during working hours). This kind of threat carries high
ambiguity and leads to extrapolations and speculations about different kinds of future
prohibitions and restrictions. The magnitude of reactance will, therefore, be higher for
arbitrary threats than defined threats.

Brehm and Brehm (1981) discuss the aggregated impact of a number of sub-threshold
threats from a source that, when collectively combined, is potent enough to be considered
a threat to one’s freedom. Each sub-threat on its own does not have the efficacy to constitute
a threat. Threats reach an optimum point. Brehm and Brehm (1981, p. 58) write about the
threshold of freedom elimination which exists when the “perceived magnitude of threat
overshadows the perceived importance of freedom,” and about the optimum point of a
threat: “the point at which a threat turns into an elimination of freedom is the point at
which the magnitude of reactance is greatest. Additional force (and perceived threat)
beyond that point has no effect on the magnitude of reactance.”

Reactance is moderated by the importance of the freedom. In other words, the threat in
itself does not produce the psychological reactance. For example, threats to freedoms of low
importance, even with high-level threats, would create low levels of arousal and non-
expression of the freedom. However, freedoms of high importance create high reactance
and an arousal to express the threatened freedom.

The magnitude of the threat is influenced by the number of freedoms threatened by one
or more factors. Time is a moderator here. Reasonable time lapses between one threat and
another prime the person to a second threat, and reactance is increased as compared to two
or more threats being experienced simultaneously. Primed persons are more likely to
express greater psychological reactance (Brehm & Brehm, 1981; Miron & Brehm, 2006).

Wicklund (1974) postulates the hydraulic principle or summation effect of reactance.
The principle states that when two freedoms are threatened, the reactance arousal of the first
threat (in which there are restrictions in the expression of reactance) is channeled into
the reactance arousal of the second threat to freedom. Brehm and Brehm (1981) later
complemented the principle by identifying conditions under which the principle would best be
applied. They identified three conditions: threats must be different, and at least two present,
with some relationship between the threats or freedoms; a time lapse must exist between threats;
and there is no reduction of reactance for the first threat – the reactance is still present.

Some social factors that contribute to the magnitude of psychological reactance include
monetary and verbal inducements, the attractiveness of the communicator, instructions,
modeling behavior, role play, and the exertion of pressure for compliance (Brehm &
Brehm, 1981).

Effects of Reactance

Reactance as a motivational force pushes the individual to reduce levels of arousal. Display
of reactance decreases arousal and leads to restoration of free behavior either directly, via
exhibiting the threatened behavior, or indirectly, via focusing on the source (i.e., discred-
iting or attacking). Reactance is displayed on two levels – behavioral and nonbehavioral

Chadee_c01.indd 16Chadee_c01.indd 16 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 17

(subjective) – and a range of reduction options exist on these levels (Brehm, 1966; Brehm &
Brehm, 1981; Miron & Brehm, 2006).

The individual adopts a number of behaviors to reestablish a sense of freedom, includ-
ing direct behavioral reassertion of freedom; developing a greater liking for the behavior
that was threatened and displaying the threatened behavior; indirect reassertion of free-
dom by adoption of extreme (e.g., costly, taboo, or dangerous) behaviors that imply power
to display eliminated behaviors (boomerang effect); and aggressive behavior directed
toward the entity threatening the freedom. There are situations in which outcomes are
uncontrollable (a situation of learned helplessness) and freedoms are impossible to restore.
In these situations an individual may relinquish a freedom so as to reduce or eliminate
reactance (Brehm & Wortman, 1975). In situations of exposure to threats by powerful
communicators reactance is reduced when there is possibility of future interaction, nega-
tive consequences to others, and negative consequences to oneself (Miron & Brehm, 2006).
In group situations, individuals may comply with the group’s alternatives rather than
display reactance for the eliminated alternative by the group. Consequences for future
interaction within the group are considered. In both situations, the cost and benefit analysis
will determine the reduction of reactance strategy (Miron & Brehm, 2006).

Reactance effects exist on a continuum from physiological arousal with no observable
signs to hostility. Other reactance reduction strategies include the following:

1. Derogate and discredit threatening sources. This strategy is adopted especially if the
source does not have legitimate power to threaten the freedom.

2. Display aggression and hostility toward the threatening source. Hostility is a subjective
experience of reactance and should be distinguished from instrumental hostility.

3. Decrease the proportion of freedoms that were threatened.
4. Compensate by increasing other freedoms.
5. Seek opportunities to counter threatening messages/sources.
6. Increase the number of available freedoms.
7. Reduce compliance behavior.
8. Position oneself to increase future freedom behavior.
9. Increase self-direction to one’s own behavior in achieving one’s goal of reducing

reactance.
10. Change cognition (e.g., reevaluate the threat and define it as not posing a threat

to freedom).
11. Change behavior (e.g., engage in behaviors that would compensate for the threat).
12. Remove the threat (e.g., leave the relationship).
13. Reconcile oneself to the loss of freedom.
14. Reduce reactance by implication. Reactance by implication occurs when we observe

someone’s freedom being threatened. Reactance can also be reduced vicariously in the
same way by observing the restoration of the other’s freedom.

15. Seek out others who can assist in the restoration of our freedom.
16. Increase perceived available behavior.
17. Subjectively, imaginatively, seek out others who can assist in the restoration of our free

behavior.

Chadee_c01.indd 17Chadee_c01.indd 17 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

18 Derek Chadee

18. Deny the presence of a threat.
19. Persevere by focusing on maintaining and protecting non-threatened free behaviors.
20. Adopt impression-management strategies.

Research on Reactance

Research on psychological reactance has been immense over the years, with findings that
have questioned conventional wisdoms. Research on reactance has been useful in critically
developing the dynamism of the theory to offer explanations for a wide range of phenom-
ena. This section describes some of the important reactance research undertaken over the
years and focuses on situational factors in the arousal of reactance.

Social interaction

Andreoli, Worchel, and Folger’s (1974) research examines arousal by implication
(i.e., observing another’s freedom being threatened – indirect) and not only by the presence
or intended presence of a future threatener. Earlier work by Sensenig and Brehm (1968)
found that the magnitude of reactance was increased for persons who expected future
interaction with a previously direct threatener. The 1974 study asked persons to give a
ranking of five desired topics for discussion. Subjects were randomly assigned to one of
three conditions and then placed in a room and, while waiting, heard a discussion in which
a person’s (B) freedom-of-topic choice was threatened by another person (A) (threat con-
dition). The subject’s first-ranked topic choice was chosen by A without consultation with
B. The threatened and restored condition by the experimenter (restored condition) involved
the subject’s first-ranked topic choice being chosen by A without consultation with B, but
the experimenter intervened to ensure B’s participation in the topic selection. The third
condition was a non-threat (neutral condition), where there was a preference for the
subject’s first topic choice, but consultation with B was implied. Following this exposure,
subjects rated the five topics.

The major findings of the study suggest that threats to the freedom of others (indirect threat
to the observer) provoked reactance in the observer as though it was a direct threat and in spite
of whether there would be future interaction with the threatener. As a result of the threat to
their freedom to choose their first-ranked topic in the threat condition, subjects were more
desirous of their second- and third-ranked topics than the first. The study also found that reac-
tance emerging from implication can also be restored by implication, as in the restored condi-
tion with the experimental intervention. Unlike previous studies (Worchel, 1974; Worchel &
Andreoli, 1974), derogation of the threatener was absent, with no significant difference in
the attractiveness rating of the threatener among the three conditions. In the previous studies
there were direct interactions with the threatener. However, in this study there is an indirect
interaction.

Baer, Hinkle, Smith, and Fenton’s (1980) study assesses the extent to which the expression
of autonomy moderated by self-presentation can reduce psychological reactance. Though
their research tested a similar hypothesis to Brehm and Mann’s (1975) study, which did not

Chadee_c01.indd 18Chadee_c01.indd 18 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 19

find differences in psychological reactance in public and private expressions of autonomy,
they used a different experimental design than the 1975 study. Instead of the within-subject
design utilized by Brehm and Mann, Baer et al. randomly assigned subjects to each of
the experimental conditions to ensure that subjects were not primed in their response in the
public or private attitude conditions. The study utilized a 3 (prior exercise of freedom: public
exercise, private exercise, no exercise) × 2 (threat: mild or severe) × 2 (manner of postcom-
munication attitude disclosure: public or private) factorial design. The study hypothesized
that the prior exercise of public expression of freedom condition would more likely reduce
reactance than private and nonprior conditions since it acted as a mechanism for the pro-
jecting of autonomy. The study also found that postcommunication attitude change occurred
in an opposite direction to the threatening communication, and only in the the projection
of autonomy condition, that is, the public postcommunication condition.

The findings of Baer et al. (1980) suggest that private attitudes, although maintaining
autonomy, do not allow for the projection of autonomy and are unable to reduce psycho-
logical reactance. The major findings of the study support the self-presentation model of
projected autonomy in the reduction of psychological reactance. This held true whether the
expression of the autonomy was prior or subsequent to the threat. The inability of project-
ing autonomy motivated people to engage in postcommunication reactance attenuation.

Social exchange theory suggests that norms of reciprocity underlie social interaction and
people are motivated to return rewards and favors of others (Homans, 1974). Counter to
the social exchange proposition on norms of reciprocity, Brehm and Cole (1966) argue that
favors and rewards may provoke non-reciprocation. They posit that the favored person will
experience reactance because there is a pressure, as a result of the norms of
reciprocity (see Gouldner, 1960), to return the favor which is interpreted as a threat to one’s
freedom. The intensity of reactance actually experienced will result from the degree to
which one will want to be freed from the pressures of reciprocation, that is, returning the
favor. The authors (p. 420) note:

If a person thought he was free to engage in Behaviors X, Y, and Z, and Behavior X were then

somehow made impossible, he would experience reactance and would be motivated to recover

his freedom to engage in X. Conversely, if he were “forced” to engage in X, his consequent

reactance would lead him to avoid X and attempt Y or Z.

The reactance reduction strategy may involve the elimination of any sense of obligation
toward the favorer and show little or no gratitude to the favorer.

Reactance theory also questions the hypothesis on the inverse relationship between group
attractiveness and disagreement with others in groups – that is, as group attractiveness
increases, group disagreement within the group decreases. Brehm and Mann’s (1975) study
hypothesizes that threats to freedoms of low importance encourage a direct relationship
between group attractiveness and movement toward the group position. However, there is
an inverse relationship between group attractiveness and gravitation to group position in
conditions of high importance of freedom. The findings showed that a boomerang effect
(reactance resistance) occurred privately and pushed behaviors toward public display of
this effect in the high importance of freedom condition with pressure for conformity.
However, modifications of this condition (high importance of freedom condition with

Chadee_c01.indd 19Chadee_c01.indd 19 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

20 Derek Chadee

pressure for change), with removal of the pressure but keeping the high importance of
freedom, created a sleeper effect (change occurring but not immediately), with private
opinion modifying in a positive direction toward the group.

The forbidden fruit

Messages are sources of information. Reactance theory emphasizes the sensitivity of the
message in persuasion. For example, in one of the many studies undertaken in this area,
Bushman (1998) assesses the relationship between reactance and warning labels, that is, to
what extent were warning labels viewed by consumers as a constraint to their choice of
freely using a product – did labels create the effect of “forbidden fruit” or “tainted fruit”?
The forbidden fruit theory underlies psychological reactance in that restrictions on the
acquisition of something act as a constraint on our freedom, inducing reactance. Proscribed
behaviors become more attractive and reduce reactance. Authoritative warning labels on
products may create a boomerang effect, making the product more attractive to the con-
sumer. Bushman posits that the tainted fruit theory reduces attractiveness to the product
because consumers conceptualize warning labels as signals of possible harm. However,
citing Bushman and Stack (1996) and Snyder and Blood (1992), he notes that research is
more supportive of attraction than repulsion by warning labels. Utilizing information
labels, warning labels (information labels from an authoritative source), and no label (no
warning/information label) on a fatty product, Bushman found that though the warning
label and no label conditions were more attractive, people were less likely to taste the prod-
uct in the information and warning label conditions. Bushman (2006) also found that
warning labels on violent television programs across five age groups (ranging from 9 to 21
years and over) were more likely to attract persons in these groups to the violent program
than information labels and no label. Psychological reactance was utilized to explain the
attraction to the “forbidden fruit.”

Gender (nature–nurture)

Is psychological reactance only experienced and displayed by those who have an under-
standing of their sense of freedom? As Brehm and Weinraub (1977) put it, must there be
complex cognitive ability for psychological reactance to be displayed, in that an under-
standing of sense of freedom is needed for the arousal of psychological reactance? Their
study utilized children (boys and girls) at the age of 2 and randomly assigned them to one
of three physical barrier conditions: (1) a large physical barrier with identical objects behind
the barrier; (2) a large barrier with dissimilar objects; and (3) a small barrier with dissimilar
objects. Boys were more attracted to objects in the second condition, while girls were more
attracted to objects in the third condition. The boys were more sensitized to situations of
threats to their freedom to freely obtain objects created by the barriers. Once the desired
object was obtained there was a decrease in psychological reactance and factors other than
reactance became influential, including the intrinsic attraction of toys and attention span.
Girls were more likely to approach objects that were less difficult to obtain. This study dem-
onstrated that there are motivational processes that may transcend cognitive ability in the

Chadee_c01.indd 20Chadee_c01.indd 20 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 21

arousal of psychological reactance. Possible explanations proposed in the study for the
gender differences in the findings include a greater emphasis on visual cues by boys and
verbal cues by girls (Maccoby & Jacklin, 1974), the higher accommodation of restrictions by
girls than boys, and the evasion of obstacles by boys (Goldberg & Lewis, 1969).

Social Influence

Reactance is also a function of pressure to comply (social influence), which can vary from
subtle to outright attempts to achieve compliance. A consequence of social influence is its
eroding effect on an individual’s effort to exert influence in a social setting. For example,
Brehm and Sensenig (1966) argue that social influence that attempts usurpation of choice,
especially when we are aware of our freedom to make the particular choice, will arouse
reactance and lead to rejecting the source of the social influence. Their study also showed
that the evaluation of continuing perception of threat was a function of the magnitude of
reactance and inclination toward the denunciation of the social influence.

How does reactance or compliant behavior influence regret? Crawford, McConnell, Lewis,
and Sherman (2002) assess the relationship among reactance, compliance, regret, and per-
suasive attempts at changing behavior. The study found that individuals, in assessing future
anticipated regret, were more likely to estimate higher regret for any loss if reactant behavior
rather than compliance was likely to be displayed to a persuasive attempt. However, in actual-
ity, in an attempt to reduce anticipated regret, individuals were more likely to engage in
complying behavior to a request. In fact, individuals displaying anticipated regret and com-
pliant behavior, as against persons displaying reactant behavior, actually experienced higher
regret. The authors argue that compliant behavior is associated with dependency and attri-
bution of blame is externalized, as compared to reactance behavior, which involves an inde-
pendent choice in which the acting person assumes responsibility for their behavioral
choice.

Silvia (2005) assesses the relationship between interpersonal interaction, similarity, and
psychological reactance. Similarity between interacting parties reduces resistance and
enhances compliance in that similarity tends to increase attraction and liking, which, in
turn, leads to the threatening behavior of liked persons being interpreted as less coercive.
The findings of this study suggest that reactance is more likely to be aroused when a message
that is threatening emanates from a dissimilar other. However, when the communicator of a
threatening message was similar, persuasion occurred as though the message was not threat-
ening (a normal message). Similarity as a positive interaction factor intervened in influenc-
ing compliance and reducing resistance which the threatening message aroused. The findings
also suggest that even in a threatening situation the magnitude of reactance can be reduced
by positive social influence. In this study, the researchers created similarity by creating iden-
tical first name, date of birth, and congruent values of communicator-participants.

Resistance can be used to reestablish one’s sense of freedom via a freedom-affirmation
intervention. Heilman and Toffler (1976) studied the reaffirming of freedom, assessing the
relationship among message (threat or promise), option (no choice or choice option), and
interpersonal interaction (interpersonal or non-interpersonal). The research found that

Chadee_c01.indd 21Chadee_c01.indd 21 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

22 Derek Chadee

compliance was higher in threatening situations when a choice was available, and both
threats and promises were equally effective in obtaining compliance in a choice situation.
An important factor in the interpretation of a threatening situation is the degree of inter-
personal interaction. Overall, the greater the interpersonal interaction, the greater the
compliance. However, an important finding of the study was the fact that though liking
(interpersonal concern) was not the only variable that induced compliance, it moderated
responses to freedom reduction (reactance).

Table 1.1 summarizes research that focuses on psychological reactance as a state condi-
tion (i.e., influenced by the situation).

Reactance as a Trait

Is psychological reactance strongly associated with a disposition or, as most of the earlier
research assumed, is it a state condition? The relationship between reactance and personality
traits was discussed by Dowd, Wallbrown, Sanders, and Yesenosky (1994). They identified a
number of factors that may contribute to obstacles in therapeutic sessions including reactance.
A psychological reactant personality profile was identified using the California Psychological
Inventory – Revised (CPI-R), the Therapeutic Reactance Scale (TRS), and the Questionnaire
Measuring Psychological Reactance (QMPR). Their study found that psychologically reactant
individuals were less concerned with impressing others or adhering to social norms and regu-
lations, were “somewhat careless about fulfilling duties and obligations,” were inclined to
express strong emotions and feelings, and were preoccupied with future possible problems.
The researchers also found that women who were classified as reactant displayed more capa-
bility in decision making and were more “self-assured, sociable, and action-oriented.” The
psychological profile developed in this study highlighted certain similarities between reactant
persons and individuals with psychopathic and narcissistic personality disorders.

Another study (Huck, 1998) found that paranoid, antisocial, and sadistic disorders were
highly correlated with reactance, and that persons with histrionic, avoidant, and/or depend-
ent disorders displayed low reactance behavior. According to Seibel and Dowd (2001), there
is an optimum normal level of reactance. Once this threshold is crossed, excessive reactance
behavior loses utility and begins to have a negative impact on the individual. The authors
found that persons with personality disorders (e.g., obsessive-compulsive) characterized by
autonomy or mistrust exhibit high reactance.

Dowd and Wallbrown (1993) tested motivational components of psychological reactance
utilizing the QMPR, the TRS, and a Personal Research Form (PRF). Their findings suggest
that persons high in reactance are aggressive, dominant, defensive, autonomous, quick to
feel offended, and low in social desirability. In sum, they are deemed loners who are deficit
in strong relations with others. Positive characteristics associated with high-reactant indi-
viduals were high levels of confidence in their beliefs and goal orientation.

Persons higher on external locus of control were more likely to be reactant (Pepper, 1996).
These findings contradict previous studies (Mallon, 1992; Morgan, 1986 as cited in Pepper,
1996). However, an earlier study found no significant difference in levels of reactance between
internally and externally oriented persons on locus of control (Cherulnik & Citrin, 1974).

Chadee_c01.indd 22Chadee_c01.indd 22 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 23

Table 1.1 Research on situational reactance.

Author Comments

Brehm, J. W. (1966) Psychological reactance theory formally presented in this monograph.

Brehm, J. W., &

Cole, A. H. (1966)

Explored the effects of a favor on reactance arousal. This study found that the importance

placed on the behavioral freedom influences the arousal and magnitude of reactance.

Brehm, J. W., &

Sensenig, J. (1966)

Examined social influence as a function of implied or attempted threat to freedom. Results

from this study showed that when a person’s freedom of choice is usurped, the individual will

experience psychological reactance and thus reject the influence.

Hammock, T., &

Brehm, J. W. (1966)

Explored the attractiveness of alternatives when freedom is reduced or eliminated. The two

experiments yielded the results that a choice alternative which is eliminated by another

person will tend to become more attractive, and a choice alternative forced by another

person will tend to become less attractive.

Kornberg, A., Linder,

D., & Cooper, J.

(1970)

Proposed how psychological reactance can provide a new perspective on understanding

political behavior. Results supported the idea that reactance theory can be applied to political

behavior.

Linder, D. E., &

Crane, K. A. (1970)

Analyzed the converging attractiveness of two alternatives as the time for a final decision

approaches. This study found that the importance of the decision and the initial attractiveness

of the alternative combined to arouse reactance within a short period of time.

Wicklund, R. A.

(1970)

Examined cognitive dissonance theory and reactance theory as theoretical explanations for

regret when decision freedom is threatened. The results indicated that under conditions

designed specifically to discriminate between the two theories, the prediction from reactance

theory was supported while the prediction from dissonance theory was not.

Worchel, S., &

Brehm, J. W. (1970)

Examined the effect of strong threats to a person’s freedom in adopting the attitudinal position

and the effect of responses to threat as a function of initial agreement or disagreement with

the position of the communicator. Findings detected that subjects who received freedom

threatening communications in agreement with their own position tended to move away from

the advocated position, while subjects in all other conditions tended to move toward the

advocated position.

Linder, D. E.,

Wortman, C. B., &

Brehm, J. W. (1971)

Time prior to decision making was manipulated, with results supporting the hypothesis that

the shorter the decision time the more the attractiveness of alternatives converges.

Worchel, S., &

Brehm, J. W. (1971)

Explored direct and implied restoration of freedom. Results from both experiments supported

the hypothesis that restoration of freedom reduces the increase in desirability of the alternative

which results from a threat to freedom.

Berkowitz, L. (1973) The author reviewed studies that displayed individuals’ unwillingness to aid others and

utilized reactance theory.

Andreoli, V. A.,

Worchel, S., &

Folger, R. (1974)

Identified conditions necessary for the arousal of reactance by implication. This

study supported the idea that reactance can be aroused by implication (by observing another

person’s freedom being threatened) and restored by implication (by observing another person’s

freedom restored).

(cont’d)

Chadee_c01.indd 23Chadee_c01.indd 23 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

24 Derek Chadee

Table 1.1 (cont’d)

Author Comments

Wicklund, R. A.

(1974)

Reviewed studies undertaken on reactance.

Brehm, J. W., &

Mann, M. (1975)

Examined the effect of group conformity pressure on public and private opinions

with regard to reactance. Results displayed that if importance of freedom is relatively

great to the individual, both private and public compliance decrease, as group attraction

increases.

Heilman, M. E., &

Toffler, B. L. (1976)

Investigated the conditions under which the negative consequences of social influence

attempts can be mitigated by freedom-affirming interventions. The results showed that in

social situations individuals’ concerns about their freedom were interpersonally

motivated.

Miller, R. L. (1976) Assessed the impact of varied intensities of exposure to a persuasive message on attitude

change and psychological reactance. This study showed that mere exposure enhances the

evaluation of stimuli whereas overexposure dampens positive attitudinal effects.

Snyder, M., &

Wicklund, R. A.

(1976)

Examined the effect of exercising prior freedom and reactance effects. The two experiments of

this study showed that when individuals exercise their freedom prior to it being threatened,

they do so to prevent the onset of reactance.

Brehm, S. S., &

Weinraub, M. (1977)

Applied the theory of reactance to the attractiveness of goal objects when impeded by barriers.

This study found that 2-year-old boys were more reactant than 2-year-old girls and this was

due either to perception of cues or to culture.

Carver, C. S. (1977) Investigated the role of self-awareness in the perception of threat and resultant reactance.

Found that an individual will resist persuasion once a communication is interpreted as

coercive, and this can happen when the individual has some degree of self-awareness.

Stillman, P. G. (1977) A review of Freedom and Reactance by Robert A. Wicklund.

Baer, R., Hinkle, S.,

Smith, K., &

Fenton, M. (1980)

Examined the extent that individuals can project autonomy before or after a threat to

freedom. This study supported the self-presentational view of reactance, in that reactance

effects were confined to participants’ public attitudes, whereas their private attitudes were

unchanged. When prior exercise of freedom was public, the reactance process for

participants was hindered.

Clee, M. A., &

Wicklund, R. A.

(1980)

A review of psychological reactance and its broad applicability to consumer behavior.

Brehm, S. S., &

Brehm, J. W. (1981)

Described the theory and research of psychological reactance.

Wright, R. A., &

Brehm, S. S. (1982)

Critically reviewed reactance and impression management.

Brockner, J.,

Gardner, M.,

Bierman, J.,

Mahon, T., Thomas,

B., & Weiss, W. (1983)

Explored the influential properties of self-esteem and self-consciousness in interaction with

the Wortman–Brehm model of reactance and learned helplessness. Low-self-esteem

individuals exhibited more reactance effects when they had high self-consciousness, and

with extended failure they were more likely to show helplessness than individuals with high

self-esteem.

Chadee_c01.indd 24Chadee_c01.indd 24 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 25

Table 1.1 (cont’d)

Author Comments

Seltzer, L. F. (1983) Explored the dynamics of paradoxical intervention by stressing external or situational

pressures and the link to psychological reactance. Results suggested that the final choice was

influenced by reactance, although its primary determinants were the student’s self-efficacy and

opportunistic ideals.

Mikulincer, M.

(1988)

Examined the effects of the internal–external attributional style on affective and performance

reactions following different amounts of failure. Results indicated that following exposure to

four unsolvable problems, internal attributors exhibited stronger feelings of incompetence and

a decrease in performance compared with external attributors.

Nail, P. R., Van

Leeuwen, M. D., &

Powell, A. B. (1996)

Tested the effectance vs. self-presentational view of reactance. The study claims support the

idea that interpersonal processes can affect the expression of reactance and suggests that in

some cases concern for self-presentation may be a necessary condition for measurable

reactance effects to occur.

Wicklund, R. A. (1997) Reviewed the theory of psychological reactance.

Bushman, B. J. (1998) Investigated the effect of informational warning labels on consumables. Results were consistent

with the prediction, although not significant, that warning labels may have considerable

influence on behavior in situations in which there are clear low-cost behavioral alternatives

that are reasonably satisfactory.

Burger, J. M. (1999) A review of foot-in-the-door manipulation and psychological reactance and other

psychological processes.

Beutler, L. E.,

Moleiro, C., &

Talebi, H. (2002)

Review of resistance and reactance in psychotherapy.

Crawford, M. T.,

McConnell, A. R.,

Lewis, A. C., &

Sherman, S. J. (2002)

Assessed whether anticipated regret impacted on one’s decision to react or comply. The results

suggest that anticipated regret cannot account for reactance effects. Participants were more

likely to perceive greater anticipated regret associated with reactance versus compliance, and

thus complied.

Silvia, P. J. (2005) Explored psychological reactance with focus on individuals and their similarity to persons/

communicators who are threatening their freedom. Findings indicated that dissimilarity with

the communicator invoked reactance whereas similarity to the communicator increased liking

and consequently compliance.

Bushman, B. J. (2006) Investigated the effect of informational warning labels on attraction to violence in television

viewers of different ages. Across age groups, viewers consider warning labels to be a restriction

on their freedom to watch what they want.

Miron, A. M., &

Brehm, J. W. (2006)

Review of psychological reactance from 1966 to 2006.

Silvia, P. J. (2006) Explored whether threats to freedom can cause disagreement and examined the implications of

reactance-based sleeper effects. This study showed that disagreement based on cognitive

responses (threat at the start) was more stable over time; however, disagreement based on

motivation to restore freedom (threat at the end) was unstable over time and persons agreed

with the communicator.

Chadee_c01.indd 25Chadee_c01.indd 25 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

26 Derek Chadee

Findings from Pepper (1996) suggest that type A personalities are more reactant,
oppositional, perceive more threats to freedom, and show increased attraction to the
eliminated choice and non-compliance to clinical treatment. Individuals with high reac-
tance were more likely to make self-attributions, having a stronger belief that they can
manage problems for themselves. In actuality they were unable to perform equally or bet-
ter than persons who had sought assistance (Rhodewalt & Marcroft, 1988 as cited in
Pepper, 1996). Carver (1977), exploring self-awareness and reactance, suggested that
greater self-awareness led to increased feelings of coercion by reactant persons. Further
studies (Brockner et al., 1983; Carver & Sheier, 1981 as cited in Pepper, 1996) categorized
self-awareness into private self-consciousness and public self-consciousness. Private
self-consciousness encompasses the subjective awareness of an individual’s feelings and
perceptions, whereas public self-consciousness is an awareness of oneself as a social
object. Private self-consciousness was associated with high reactance, whereas public self-
consciousness acted as a suppressant of reactance.

Pepper (1996) found no significant relationship between gender and reactance, identify-
ing gender as a situational variable better explained by cultural factors. These findings have
also been supported by Dowd, Trutt, and Watkins (1992) and Carli (1989).

Using Erikson’s stage theory, Pepper (1996) argues that inherent in many of the stages is
the potential for psychological reactance. Reactance was linked with both the successful and
unsuccessful completion of stages, and he identifies autonomy, mistrust, intimacy, and iso-
lation as being associated with psychological reactance.

Family history was also found to be a predictor of psychological reactance. Persons who lived
in a high-conflict family environment or a family that emphasized autonomy, achievement,
and moral values were more likely to show psychological reactance (Buboltz, Johnson, & Woller,
2003). These persons may be more aware of perceived or actual threats to their freedom.

The literature, with few exceptions, paints a negative portrait of the personality profile of
a psychologically reactant person, who is seen as antisocial, low in social desirability, inca-
pable of strong relations with peers, isolated, independent, aggressive, not easily trusting,
dominant, and worried about an uncertain future. Similarly, there is a negative association
between reactance and histrionic personality styles.

Table 1.2 summarizes research that focuses on psychological reactance as a disposition.

Measuring Reactance

From its genesis, psychological reactance was conceptualized as arising from situational
factors, and research and measurements therefore were consistent with psychological reac-
tance as a state phenomenon. In attempting to study reactance, researchers created experi-
mental freedom-threat situations to arouse psychological reactance. In many of the studies
discussed, reactance was inferred by comparing responses across conditions experimentally
created (e.g., reactance vs. non-reactance), whether those measures were on attitudes, eval-
uation of message, or rating of communicator (Wright et al., 2004).

Sharon Brehm (1976) wrote on the implications of reactance theory to that area of
clinical psychology concerned with changing behavior and non-successful treatments.

Chadee_c01.indd 26Chadee_c01.indd 26 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 27

Table 1.2 Research on dispositional psychological reactance.

Author Comments

Cherulnik, P. D., &

Citrin, M. M. (1974)

Explored individual differences and psychological reactance with relevance to locus of control.

Found both external and internal locus of control participants exhibited the same levels of

reactance.

Tucker, R. K., &

Byers, P. Y. (1987)

Assessed the factorial validity of Merz’s Psychological Reactance Scale. Findings were

inconsistent with Merz’s, concluding that the instrument is psychometrically unacceptable.

Hong, S-M., &

Ostini, R. (1989)

Evaluated the questionnaire for the measurement of psychological reactance. Identified a

four-factor structure that was inconsistent with findings of Merz (1983) and Tucker & Byers

(1987), concluding the scale psychometrically unstable.

Hong, S-M., & Page,

S. (1989)

Developed Hong’s Psychological Reactance Scale. Identified a four-factor structure to measure

reactance that was reliable.

Joubert, C. E. (1990) Used Hong’s Psychological Reactance Scale to assess the relationship among self-esteem,

reactance, and personality variables. Men scored significantly higher than women on

psychological reactance measures. Happiness ratings correlated negatively with psychological

reactance. Women’s self-esteem scores were negatively correlated with psychological reactance.

Dowd, E. T., Milne,

C. R., &

Wise, S. L. (1991)

Developed the Therapeutic Reactance Scale (TRS) to measure trait psychological reactance.

Factorial analysis identified two sub-scales: verbal reactance and behavioral reactance

Hong, S-M. (1992) Assessed the validity of Hong’s Psychological Reactance Scale. Findings supported the factorial

stability and reliability of the scale.

Mallon, K. F. (1992) Reviewed the Questionnaire for Measuring Psychological Reactance (QMPR), Hong’s Psychological

Reactance Scale, and TRS. Developed the Proneness Reactance Inventory to measure dispositional

reactance.

Dowd, E. T., &

Wallbrown, F. (1993)

Determined the motivational personality characteristics associated with psychological

reactance. Findings identified a personality pattern of the psychologically reactant person as

defensive, aggressive, dominant, autonomous, and nonaffiliative.

Dowd, E. T.,

Wallbrown, F.,

Sanders, D., &

Yesenosky, J. M.

(1994)

Further explored personality characteristics of a psychologically reactant individual. Identified

that psychologically reactant women were more decisive and self-assertive than non-reactant

women. Reactant individuals tend to worry more about future problems and have weak social

relations.

Hong, S-M.,

Giannakopoulos, E.,

Laing, D., &

Williams, N. A.

(1994)

Used Hong’s Psychological Reactance Scale to explore gender and age effects on dispositional

reactance. Found that younger persons displayed more reactance than older participants, and no

difference between genders was observed.

Hong, S-M., &

Faedda, S. (1996)

Refined Hong’s Psychological Reactance Scale to 11 items that showed greater factorial stability

than the original 14-item scale.

Pepper, H. F. (1996) Explored the psychosocial precursors of psychological reactance with emphasis on Erikson’s

developmental theory. The following factors predicted psychological reactance: autonomy,

trust, intimacy, and isolation.

(cont’d)

Chadee_c01.indd 27Chadee_c01.indd 27 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

28 Derek Chadee

Table 1.2 (cont’d)

Author Comments

Hellman, C. M., &

McMillin, W. L.

(1997)

Examined the relationship between psychological reactance and self-esteem using Hong’s

Psychological Reactance Scale. These results suggest that the scale should be used with caution, since

combining its four factors in an additive format may suppress its potential to measure reactance.

Huck, N. O. (1998) Assessed psychological reactance and its relations to personality through the utilization of

Millon’s personality theory. Found seven personality disorders that are more likely than others

to evoke reactance.

Seibel, C. A., &

Dowd, E. T. (1999)

Analyzed the relationship between the client’s psychological reactance and specific

compliance behaviors and general improvements in an actual therapy situation. Identified

that reactant clients would engage in Boundary Augmentation and Boundary Reducing

behaviors, as opposed to non-reactant clients.

Johnson, P., &

Buboltz, W. C., Jr.

(2000)

Explored the link between Bowen’s (1978) concept of differentiation of self and psychological

reactance. Detected three measures of differentiation that predicted psychological reactance

(intergenerational individuation, peer intimacy, and peer individuation).

Donnell, A. J.,

Thomas, A., &

Buboltz, W. C., Jr.

(2001)

Assessed the factor structure and internal consistency of the Questionnaire for Measuring

Psychological Reactance. Concluded that the QMPR was an unreliable measure of the

dimensions of psychological reactance.

Seibel, C. A., &

Dowd, E. T. (2001)

Further explored the psychological profile of a psychologically reactant person as well as

delved into corrective techniques for developmental issues that arise due to separation and

individuation. Identified that individuals have an optimum level of reactance that balances

issues of engulfment and separation. Personality disorders based on trust and autonomy were

predictors of reactance.

Baumeister, R. F.,

Catanese, K. R., &

Wallace, H. M. (2002)

Explored a narcissistic reactance theory approach to sexual coercion and rape. Concluded that

theories of reactance and narcissism combined are a good tool of analysis when exploring

cases of rape.

Buboltz, W. C. Jr.,

Thomas, A., &

Donnell, A. J. (2002)

Analyzed and assessed the factors and consistencies of the Therapeutic Reactance Scale (TRS).

Results showed that reactance is seemingly a multidimensional construct with four relatively

independent structures underlying psychological reactance. The use of a one-dimensional

score may not give an accurate picture of the reactance level/potential of an individual.

Buboltz, W. C., Jr.,

Johnson, P., &

Woller, K. M. P.

(2003)

Assessed the relationship between variables concerned with origins of family and psychological

reactance. Findings displayed that five family dimensions (cohesion, conflict, moral–religious

emphasis, independence, and achievement orientation) can encourage psychological reactance.

Dillard, J. P., &

Shen, L. (2005)

Applied reactance theory to persuasive health communication. Showed that the tendency for

reactance proneness interacting for some topics and not others is additive for one topic and

multiplicative for another, thereby suggesting that some topics are especially potent with regard to

their ability to elicit state reactance from individuals high in the propensity to experience reactance.

Seeman, E. A.,

Buboltz, W. C., Jr.,

Thomas, A., Soper,

B., & Wilkinson, L.

(2005)

Further elaborated the profile of a psychologically reactant person by implementing the

five-factor model of personality as measured by the NEO PI-R. Results suggested that highly

reactant individuals appear very independent and somewhat suspicious; are likely to be skeptical

of others’ intentions; are competitive, intolerant, distrustful, secretive, and detached; and they put

on a good social face, but are actually uncomfortable in social situations.

Chadee_c01.indd 28Chadee_c01.indd 28 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

Toward Freedom: Reactance Theory Revisited 29

Table 1.2 (cont’d)

Author Comments

Shen, L., &

Dillard, J. P. (2005)

Assessed the psychometric properties of the 11-item version of Hong’s Psychological

Reactance Scale. They concluded that a single score on the scale is theoretically and empirically

justifiable.

Jonason, P. K., &

Knowles, H. M.

(2006)

Assessed the one-dimensional measure of Hong’s Psychological Reactance Scale. They

concluded that the scale is only moderately stable and reliable, although its one-dimensional

characteristic appears to be the most consistent and reliable construct.

However, she referred to reactance theory as a motivational state that is aroused from
situational factors. The construct of reactance was later modified to accommodate to a
clinician’s environment, with reactance being conceptualized as a personality trait.
A major concern of the clinician was the interaction of this trait (reactant personality)
with treatment (Beutler, Mohr, Grawe, Engle, & MacDonald, 1991; Shoham-Salomon &
Hannah, 1991; Snow, 1991). The genesis toward a psychometric scale emerged from
attempts to measure reactance as a disposition (Buboltz, Thomas, & Donnell, 2002; Hong &
Page, 1989; Merz, 1983). Reactance is most likely to occur in the clinical setting to persons
with volitional/free problem behaviors (e.g., drinking, smoking) as compared to uncon-
trollable behaviors (e.g., panic, obsessive-compulsive) (Shoham, Trost, & Rohrbaugh,
2004). Psychological reactance as a disposition has been measured by a number of scales
and techniques.

Merz’s (1983) Questionnaire Measuring Psychological Reactance (QMPR) was the first-
ever scale designed to capture a measure of psychological reactance and consisted of 26
items utilizing a four-point Likert scale ranging from “not at all appropriate” to “extremely
appropriate.” The original scale attempted to obtain measures on resistance, defiance,
boom erang effect and the inclination to do the opposite to what others expect of you
(Tucker & Byers, 1987). The scale was later modified by Tucker and Byers (1987) to an
18-item reactance scale utilizing a five-point Likert range from 1 = “not at all appropriate”
to 5 = “extremely appropriate.” Two major constructs in the scale were behavioral freedom
(encompassing measures of reactions to praise, subservient behavior and expectations, and
advice of others) and freedom of choice (encompassing measures of freedom and decision
making). The QMPR has been found to have convergence validity with the Personality
Research Form (PRF), suggesting that psychologically reactant individuals are more likely
to be defensive, easily offended, aggressive, autonomous, and impulsive (Dowd & Wallbrown,
1993). These characteristics are consistent with the QMPR’s reactant individual who is seen
as dominant, individualistic, with low social ties, and who creates unfavorable impressions
on others (Dowd, 1993).

Hong and Ostini (1989), utilizing a four-point Likert format of the translated QMPR, found
factorial instability and low correlations among factors. Hong and Page (1989) argue that Tucker
and Byers’s QMPR was considered to be psychometrically unstable. Other criticisms came from
Donnell, Thomas, and Buboltz (2001), who also questioned Tucker and Byers’s modification of
the original scale, noting that the instability of the scale may have also been a result of both the
translation of the original scale from German and the adoption of a five-point scale format.

Chadee_c01.indd 29Chadee_c01.indd 29 12/13/2010 2:16:51 PM12/13/2010 2:16:51 PM

30 Derek Chadee

The stability of the QMPR translated by E. T. Dowd was further tested by Donnell et al.
(2001), utilizing a six-point Likert-type scale ranging from 1 = “does not apply at all” to 6 =
“always applies.” The results of a three-factor scale structure were found to be inconsistent
with both Tucker and Byers (1987) and Hong and Ostini (1989). Though reactance appeared
to be a multidimensional construct, the QMPR was not accepted as a stable measure of
reactance (Donnell et al., 2001).

Hong and Page’s (1989) 14-item psychological reactance scale attempted to create a
factorial stable measure of psychological reactance and emerged as a critique of the QMPR.
Hong’s Psychological Reactance Scale (HPRS) utilizes a five-point Likert scale (1 = “disa-
gree completely” to 5 = “agree completely”) and has four distinctive factors: freedom of
choice, conformity reactance, behavioral freedom, reactance to advice and recommen-
dation. Hong (1992), a replication of Hong and Page (1989), supported the factorial struc-
ture of the HPRS. The HPRS was later revised (Hong & Faedda, 2006) to an 11-item scale.
Three items of the original scale were considered to be vague and were removed.

Shen and Dillard (2005) argue that a single score on the HPRS to measure trait psycho-
logical reactance is theoretically and empirically justifiable. However, Jonason and Knowles
(2006) and Jonason (2007) indicate that while the HPRS tends toward unidimensionality,
it is a moderately stable and a reliable measure of reactance and improvement of the scale
is needed.

Another method utilized in the measuring of reactance is the Trait Reactance Scale (TRS).
This scale uses a 28-item self-report questionnaire in attempting to measure reactance as a
trait. The TRS has recorded convergence validity with personality characteristics of domi-
nance, locus of control (internal), and aggression (Dowd & Wallbrown, 1993; Shoham et al.,
2004). A major problem identified with this measurement of reactance is the lack of con-
struct validity. Shoham et al. (2004) argue that what has been conceptually identified as a
reactant disposition may not necessarily be a disposition or trait but rather an oppositional
style consistent with particular behavioral styles.

In another attempt to measure reactance, Beutler et al. (1991) utilized two indices
from the Minnesota Multiphasic Personality Inventory (MMPI) in developing what they
refer to as a reactant score. The two scales that they adopted were Taylor’s Manifest
Anxiety Scale and the Edwards Social Desirability Scale. The authors utilized an arithme-
tic summation of the scores from both scales to derive a reactance score. Critics of Beutler
et al. (1991) noted that there is no rationale articulated as to why high anxiety and social
desirability should correlate with reactance (Lukin, Dowd, Plake, & Kraft, 1985; Shoham
et al., 2004). In fact, Spielberg’s State-Trait Anxiety Inventory and the TRS have no sta-
tistical relationship. Shoham et al. (2004) reported that Dowd et al. (1994) showed a
weak correlation between the TRS and social interaction. A question that can be asked is:
Are Beutler et al. (1991) and the TRS measuring the same construct? Baker, Sullivan, &
Marszalek (2003) found that there was no correlation between these two reactance
measures.

Shoham, Bootzin, Rohrbaugh, and Urry (1996) utilize a naturalistic method to measure
expressed reactance. This method involves the rating of participants’ content-filtered voice
in reply to a question that provokes reactance. High-reactant persons’ voice tones sound
inhibited, spiteful, and active.

Chadee_c01.indd 30Chadee_c01.indd 30 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Toward Freedom: Reactance Theory Revisited 31

Cognitive Dissonance and Reactance

To what extent is psychological reactance the same as cognitive dissonance? Do we always
experience cognitive dissonance when we experience reactance? There are a number of
commonalities between psychological reactance and cognitive dissonance. Both theories
are motivational arousal theories emphasizing cognitive evaluation and a gravitation toward
psychological consistency. Arousal is a noxious state (i.e., dissonance or psychological reac-
tance) and a drive is buildup-directed to achieve a state of equilibrium (consonance or
removal of threat to freedom). The level of importance of the cognition or threatened free-
dom determines the degree of cognitive dissonance or psychological reactance, respectively.
A number of strategies are used to obtain equilibrium, and these methods vary from cogni-
tive to behavioral to emotional changes. The reevaluation of importance, changing or
adding new cognitions, adoption of new behavior, or changing existing behaviors are some
of the common strategies of arousal reduction adopted by both theories. Neither theory is
set up to be scientifically falsified, and therefore they cannot be considered scientific theo-
ries. Also, both theories are parsimonious.

However, there are some major differences. Although both theories state that arousal is
unpleasant, in the case of reactance the unpleasantness is always associated with negative
emotions, unlike dissonance, in which both positive and negative emotions can be experi-
enced. Sources of the arousal are different: reactance involves a threat associated with a
freedom, and therefore there is a “victim” and “victimizer” interaction, an externally social
threatening factor, which is not the typical cognitive dissonance scenario. Therefore, there
is a narrower scope of the application of reactance theory as compared to cognitive disso-
nance. The difference in parsimoniousness of the theories is one of degree.

Wicklund (1974) was one of the first authors to attempt to address the issue of overlaps
between the theories of dissonance and reactance. He argues that as far as decision making is
concerned, cognitive dissonance occurs at the postdecisional stage but reactance at the pre-
decisional stage. This sequential effect is a distinguishing difference. That is, prior to the
reactant behavior is the motivational drive of psychological reactance. After the behavior, the
decision, is the postdecisional dissonance and dissonance reduction to justify the committed
action. The force compliance studies of dissonance (e.g., Festinger & Carlsmith, 1959) and
the barrier studies of reactance research provide a conceptually similar situation. However,
Wicklund (1974) advocates that the difference between these groups of studies lies in the
pre-decisional making without a behavioral commitment in reactance and the postdeci-
sional dissonance as a result of behavioral commitment. However, dissonance is not only
experienced in postdecisional situations. The threat to remove a freedom or the actual
removal of a freedom is inherently a dissonant phenomenon. In fact, Wicklund (1974, p. 57)
acknowledges this:

There is a parallel between the two theories when simple statements of preference are made,

independent of whether or not those statements constitute absolute commitments. Dissonance

theory allows that the statement will produce dissonance and regret (convergence) while reac-

tance theory indicates that the preference statement will threaten freedom and thereby result in

convergence.

Chadee_c01.indd 31Chadee_c01.indd 31 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

32 Derek Chadee

Therefore, if there is an overlap between reactance and dissonance the investigation of
the neurosocial psychological processes has to be evaluated. However, it should be remem-
bered that reactance theory, from its genesis, had conceptual similarities and was being
developed as a special case of cognitive dissonance. The literature has still to address a
precise demarcation between the two theories.

Application of Reactance

The application of the theory of psychological reactance has been diverse, from its use in
the clinical setting to the understanding of social problems, consumer behavior, and power
relationships. Specifically, the theory of psychological reactance has been applied within
psychology and other fields and has provided useful insights in areas such as persuasion
and resistance to persuasion, attitude change, pro-social behavior, group attractiveness,
family interaction, effects of warning labels, littering, promotional influence, manipulative
advertisements, product availability, government regulations as threats to freedom, internet
and website interaction, health communication, clinical interaction/counseling and reac-
tant personality, family therapy, political behavior, jury behavior, learned helplessness, and
even the understanding of behavior in dispute resolution.

Some studies on littering (e.g., Brasted, Mann, & Geller, 1979) utilized a reactance theory
model to understand the relationship between antilittering advertisements and compli-
ance. Messages that were strongly stated and suggested threats to freedom aroused psycho-
logical reactance and attempts to restore freedom. Consumers are influenced by promotional
campaigns but the hard-sell advertisements have less of an impact than the soft-sell (low-
threat) ones (Regan & Brehm, 1972).

A multitude of studies were undertaken in the area of reactance and attitude change and
persuasion, including Worchel and Brehm (1970) on freedom-threatening communication
and attitudinal position; Miller (1976) on persuasive messages and attitude change;
Carver (1977) on self-awareness, coercion, and persuasion; and Baer et al. (1980) on self-
presentation and public attitudes. Studies by Bushman (1998, 2006) and Clee and Wicklund
(1980) are just a few of the many studies that emphasize the impact of message insensitivity–
sensitivity on persuasion and compliance.

Baumeister, Catanese, and Wallace (2002) assess the relationship among narcissism, reac-
tance, and sexual concern, arguing that a man with high sexual expectancy of sexual inter-
action would experience psychological reactance when there is refusal of sexual contact by
a woman. Narcissistic inclination may mediate reactance, motivating the individual to
attempt sexual coercion and rape.

Family socialization impacts on freedom sensitivity, which in turn makes the individual
more responsive to freedom threats. In particular, persons with an achievement orienta-
tion, high independence and moral–religious values were positively associated with psycho-
logical reactance (Buboltz et al., 2003).

The application of reactance in understanding virtual interaction on the internet was
undertaken by Daily (2004). Restrictions and blocks to navigation on the internet create psy-
chological reactance and lead to negative emotions and avoidance behavior toward the blocked

Chadee_c01.indd 32Chadee_c01.indd 32 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Toward Freedom: Reactance Theory Revisited 33

websites. Another study assessed forced exposure via pop-up advertisements and psychological
reactance, attempting to understand factors that contribute to negative reactions to informa-
tion and particularly advertisements via the internet (Edwards, Li, & Lee, 2002).

Criticisms of Reactance

The theory of psychological reactance has been criticized on a number of grounds. First, the
theory is difficult to falsify. The emphasis of research has been on situational reactance, with
experimental conditions set up to create threats to freedom. Inconsistent findings are easily
explained by reference to the situation rather than an exploration of the theory to identify
theoretical problems. The lack of explicitly stated propositions and hypotheses provides
further insulation of the theory from the systematic assaults of science. However, other
theories such as Freud’s psychoanalytical theory and Festinger’s cognitive dissonance the-
ory have also been criticized on these grounds.

Second, the concept of reactance is a scientifically vague concept. There is tremendous
latitude left to the interpretation of exactly what behavior is reactance or non-reactance.
Premised on the concept of reactance, the magnitude of reactance is discussed. The concept
of magnitude assumes some kind of scale or degree distinction between one level of reac-
tance and another. Exactly how to evaluate the magnitude of reactance is not stated in the
theory. Therefore, the concept of magnitude of reactance has two ambiguous terms which
have implications for scientific measurement of the concept. Reactance cannot be physi-
cally observed but is inferred from measures of indicators of reactance which vary from
study to study. Third, the theory is parsimonious and can, therefore, explain a wide range of
behavior in a variety of social settings. A theory with such versatility loses its efficacy in
explaining and creates a diminishing level of confidence.

Fourth, reactance theory is not designed to assess the dynamism of social interaction and
exchanges in the movement toward reduction of the psychological tension and achieve-
ment of equilibrium. Freedom can only be threatened by a powerful other. The exchanges
and power dynamics in reasserting one’s freedom are not examined or articulated in the
theory. The magnitude of reactance, the persistence of reduction, the kinds of reduction
strategies, and the acceptance or removal of threat to freedom can all be better compre-
hended if an analysis of the power dynamics complements the theory.

Fifth, the psychological processes involved in internalizing these freedoms are not
articulated by the theory (Wicklund, 1997). This is an assumption, a given, that the theory
builds upon. An elaboration of the dynamics involved in the development of free behaviors
will provide a profound understanding of the processes involved in reactance. The cultural
contexts and their contribution to one’s sense of freedom need elaboration via cross- cultural
research.

Sixth, is reactance a behavioral attempt to demonstrate and project an impression of
autonomy and power to others in the interaction situation? In other words, is there a dis-
tinction between impression management and reactance? Heilman and Toffler (1976), for
example, argue that reduction of reactance was a direct function of balance of power, or the
impression of it, in interaction. Baer et al.’s (1980) findings support the view that there is a

Chadee_c01.indd 33Chadee_c01.indd 33 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

34 Derek Chadee

greater concern with demonstrating the possession of a freedom than concern about the
removal of the freedom. Wright and Brehm (1982, p. 616), in being critical of the impression-
management explanation of reactance, wrote:

In summary, then, the impression management explanation of reactance, in contrast to reactance

theory, predicts no effects when (a) threats to freedom are nonsocial or social threats are uninten-

tional, (b) subjects’ responses are private (nobody who has witnessed the threat will see them, or

subjects’ responses cannot be traced back to them individually), and (c) freedom restoration

occurs without the agent or an observer being aware of it. We believe that it is doubtful that reac-

tance effects can best be understood as a manifestation of impression management.

Seventh, reactance theory places great emphasis on the cognitive but very little emphasis is
given to the affect. The next section proposes the systematic introduction of the affect into
the theory of psychological reactance.

Toward a Reactance Emotion Theory

Philip Cowan, in his preface to Piaget’s Intelligence and Affectivity (1981, p. xi), wrote:

Affect is likened to the gasoline that activates the car, while the engine (cognition) provides

structure for the energy and direction of the car’s motion. Affect as “energetics” can combine

with cognitive structural schemes to focus the individual’s interest on a specific thing or idea.

Because it influences an individual choice of whether to exert intellectual effort, affect serves as

a regulator of action. Because it influences the choice of specific goals, affect also plays a role in

determining values (internal interest projected outwards so that things and people appear to

have a certain worth). By regulating action and determining values, affect influences our ten-

dency to approach or avoid situations.

An understanding of human emotions in influencing the magnitude of psychological reac-
tance is crucial to the development of reactance theory, which, over the last forty years, has
comfortably maintained itself as a purely cognitive theory. Even though reactance theory
can be classified within an appraisal theory of emotions, the literature has consistently
focused on a cognitive model in the application of the theory. The degree of psychological
reactance is moderated by the particular emotion(s) aroused as a result of the cognitive
evaluation of a particular threat to freedom. The degree of reactance, therefore, is a func-
tion of emotional arousal, which, in turn, is influenced by the particular emotion provoked.
Reactance is unpleasant, and therefore will provoke negative emotions (primary and
secondary) such as frustration, shame, fear, anger, sadness, guilt, contempt, and disgust
(Smith & Ellsworth, 1985, 1987). Some important contributions to the understanding of
emotions provide a strong rationale for a reactance emotion theory.

An emotion energizes and prepares the individual to respond and has at least four different
aspects – feelings, actions, physiological arousal, and motivational programs (Rosenzweig,
Breedlove, & Watson, 2005). As Nolen-Hoeksema, Fredrickson, Loftus, and Wagenaar
(2009) recognize (also see Lazarus, 1991; Rosenberg, 1998), an emotion involves
cognitive appraisal, subjective experience, thought–action tendencies, internal bodily

Chadee_c01.indd 34Chadee_c01.indd 34 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Toward Freedom: Reactance Theory Revisited 35

changes, and facial expression. Plutchik (1994) identifies eight basic emotions. These
emotions are arranged in pairs of opposites: joy/sadness, affection/disgust, anger/fear,
and expectation/surprise. However, Smith and Ellsworth (1985) argued earlier that com-
paring emotions in such pairs is somewhat problematic since, for example, joy is a posi-
tive emotion and sadness a negative one, but both anger and fear are negative
emotions – these pairs do not exist in comparable continuums from positive to negative,
but are typologies. Smith and Ellsworth (1985) utilized six dimensions in the description
of 15 emotions (primary and secondary emotions). These dimensions were: desirability
of the situation (pleasant or unpleasant); anticipated effort in the situation (low effort or
high effort); situational certainty (certain or uncertain); attentional activity to the
situation (low attention to high attention); degree of control over the situation (self or
others); and control attributed to the situation (situational control and human control).
Therefore, emotions (e.g., anger and frustration) would vary depending on the dimen-
sions that are involved in determining the particular emotion. A major conclusion drawn
from their study was (p. 831):

Our results show that people’s emotions are intimately related to their cognitive appraisals of

their circumstances. The subjects rated past emotional experiences along six appraisal dimen-

sions, and different patterns of appraisals were strongly associated with different emotions.

Zajonc (1980) puts another light on the expression of emotions. He argues that there are
emotions that are grounded on cognitive appraisal (postcognitive emotions) and emotions
that emerge before cognitions (precognitive emotions). On postcognitive emotions Zajonc
(1980, p. 151) writes:

An affective reaction, such as liking, disliking, preference, evaluation, or the experience of

pleasure or displeasure, is based on a prior cognitive process in which a variety of content dis-

criminations are made and features are identified, examined for their value, and weighted for

their contributions. Once this analytical task has been completed, a computation of the com-

ponents can generate an overall affective judgement.

Bargh and Apsley (2001), writing on Zajonc’s argument against this temporal order of cog-
nition and emotion, state:

He (Zajonc) asserted that the notion that people go through life in a rational manner, objec-

tively weighing the pros and cons of various alternatives prior to taking a position as suggested

by extant information-processing models, might be little more than wishful thinking. Instead,

he argued that the affective qualities of stimuli such as good/bad, like/dislike, or approach/

avoid might be processed extremely quickly and efficiently and, consequently, could be among

the very first reactions of an organism to its environment. Within an affective primary frame-

work, individuals gather information about various alternatives, not to make more informa-

tion choices, but to corroborate their initial preferences and desire.

The intensity of emotions differs. The sympathetic branch of the autonomic nervous system
is crucial in creating the necessary physiological arousal for negative emotions involving

Chadee_c01.indd 35Chadee_c01.indd 35 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

36 Derek Chadee

brain activity in both the hypothalamus and the amygdala. The debate on autonomic
arousal of emotions has consistently appeared in the literature, from the James-Lange the-
ory to Canon-Bard’s critique, and from advocacy on similarities of autonomic response to
empirical testing of physiological arousal of different emotions (Funkenstein, 1955).
However, this debate was systematically addressed by Levenson, Ekman, and Friesen (1990),
who found distinctive autonomic response among emotions. Their study also shows that
the highest emotional intensity was displayed for anger, followed by fear and sadness –
negative emotions that are associated with reactance.

Research by Keltner, Ellsworth, and Edwards (1993) is supportive of the view that affects
influence the way we make judgments and attributions of future events and responsibility.
Siemer’s (2001) study supports the dispositional mood model’s view that our moods influ-
ence our generalized appraisal inclinations. Findings by Lerner and Keltner (2001) have
also shown that our cognitive evaluation of risk is influenced by the specific emotional
state, in that fear states were more likely to incite pessimistic risk evaluation than anger,
which was more likely to create optimistic risk assessment. These effects were moderated by
certainty and control (see also Johnson & Tversky, 1983).

The importance of appraisal to emotions (primary, secondary, reappraisal) was articu-
lated by Lazarus (1982) and later elaborated by Smith and Lazarus (1993), who catego-
rized six appraisal components related to primary and secondary appraisals. These
categories falling within the primary and secondary appraisals include motivational con-
gruency (evaluation of goals); motivational relevance (commitment); accountability
(assignment of responsibility of blame and praise); problem-coping potentiality (resolv-
ability); emotional-focused potentiality (emotional management of situation); and future
expectancy (changeability of situation). The emotions aroused would be determined by
the set of components that are present. Smith and Kirby (2001), in elaborating on Smith
and Lazarus’s (1993) theory, note that the appraisal processes take place correspondingly
and involve an automatic associative process, a conscious reasoning of events, and a
continuous appraisal- detector process that assess the information that comes from the
associative and reasoning processes. The emotion that is experienced is a result of these
dynamics.

Berkowitz and Harmon-Jones (2004) posit that anger emerges as blame and is assigned
as a negative event to a causal agent external to the individual. Unlike other negative emo-
tions, anger is associated not with an avoidance tendency but with an approach motivation
inclination. Berkowitz’s (1990, 1993, 1999, 2003) cognitive–neoassociationistic (CNA)
model of anger emergence postulates that both avoidance and approach tendencies can be
generated by an aversive situation and not just one or the other. Extending this argument to
reactance theory, a removal of a freedom can create both fear (flight) and anger (fight),
depending on the circumstances. One emotion may overpower the other, which will deter-
mine the behavior.

Harmon-Jones and Allen (1998) found that anger was associated with more left anterior
cortical activity than right. This brain activity was explained more by approach motivation
associated with dispositional anger. Generally, negative emotions and avoidance motivation
are associated with the right anterior cortical area of the brain, while positive emotions and

Chadee_c01.indd 36Chadee_c01.indd 36 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Toward Freedom: Reactance Theory Revisited 37

approach motivation are associated with the left anterior cortical area (Harmon-Jones &
Allen, 1998). The authors suggest that high dispositional anger is associated with higher
approach motivation and lesser withdrawal motivation. Later, Harmon-Jones and Sigelman’s
(2001) research findings confirmed that state-induced anger was also associated with left
prefrontal cortical activity. Further, left frontal activity created by anger does not restrain
amygdala action but is “co-varying with it” (Harmon-Jones, 2004). The relationship between
left frontal cortical activity and resolution of inconsistency via approach motivation is also
supported by Harmon-Jones (2004). Reactance will evoke particular kinds of emotion, and
emotions evoked will determine whether there is left or right frontal brain activity. This fact
has implications for the reduction of reactance, especially in volatile and high-risk
situations.

Mention of physiological arousal and emotions are scattered across the many works indi-
rectly related to reactance theory. A few reactance studies have used the emotion dimension
to assist in explanations of behavior. For example, earlier research had indicated that reac-
tance was associated with somatic tension and increased activation of the sympathetic
nervous system, as reflected in increases in both epinephrine and norepinephrine (Price,
Barell, & Barell, 1985). Brehm and Wortman’s (1975) model implies emotional intensity as
a factor that increases striving to overcome failure and decreases in motivation as a function
of being unable to control and determine outcome. However, the concept of emotional
intensity was never integrated or elaborated in the theory.

The literature is practically void of systematic research on reactance and emotions.
Neither is there an integration of findings on affect into the theory or any kind of theory
modification. Brehm (2004) notes, with respect to cognitive dissonance, that more consid-
eration is needed of the role of affect, to which little emphasis has been given. Miron and
Brehm (2006, p. 8) acknowledged the importance of understanding the processes in arousal
when they stated:

Obviously, more work is needed to chart the physiology of reactance. An investigation of this

sort would help explore important questions such as what happens to reactance arousal when

one cannot restore an eliminated freedom or whether observers experience reactance arousal

when they witness threats to other people’s freedoms.

Figure 1.1 describes the process involved in the emergence of psychological reactance but
taking into consideration the affect state. A threat to a free behavior is evaluated, provoking
a negative emotion – the more important the threat, the greater the intensity of the emo-
tion. Cognitive appraisal, together with the concomitant emotion, leads to a motivational
or withdrawal approach. Continuing evaluation is undertaken and, should threat persist,
reactance behavior continues. However, as discussed, reactance is moderated by a number
of factors.

The subjective experiences of emotions influence: the processing of information,
decision-making processes, our attention, reaction time, and the way we intend to behave
or actually behave (e.g., Clore, Gasper, & Garvin, 2001; Derryberry & Tucker, 1994). Recent
research undertaken by Kostic, Chadee, and Nedeljkovic (in press) has shown that the

Chadee_c01.indd 37Chadee_c01.indd 37 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

38 Derek Chadee

subjective experiences of emotions vary across cultures with respect to intensity. So an
unpleasant situation of uncertainty needing high attention and effort which is controlled
by others will definitely lead to frustration and anger, two emotions with the highest levels
of intensity compared to other emotions. Therefore, threat to freedom that provokes these
emotions will create a higher magnitude of reactance, other things remaining equal, than
other emotions. However, the magnitude of psychological reactance may vary across cul-
tures, and this may not necessarily be a result of situational factors but a consequence of the
subjective experience of emotions in these cultures. Clearly, research is also needed to test a
reactance emotion theory.

Conclusion

Reactance theory is an important theory in the discipline of social psychology and has
contributed to an understanding of behavior in response to actual or perceived threats in a
wide variety of settings. The theory is as useful today as it was over forty years ago. However,
the theory needs modification. The systematic introduction into the theory of the affect
state is crucial, with an understanding of the relationship between emotions and motiva-
tion. This much-needed structural modification will definitely give new insights into previ-
ous research. A reactance emotion theory will have heuristic utility in the understanding
and reduction of arousal states, including situations that are volatile and interactions that
are potentially aggressive. As the theory evolves so too will our understanding of human
behavior.

Figure 1.1 Reactance emotion model.

COGNITIVE
EVALUATION

UNIMPORTANT

NEGATIVE
EMOTION

IMPORTANT

THREAT TO
FREEDOM

PRIMARY EMOTIONS:
disgust, sadness, fear, anger

PHYSIOLOGICAL
AROUSAL

SECONDARY EMOTION:
shame, guilt, hate

MOTIVATIONAL
APPROACH

Frontal lobe activity

PSYCHOLOGICAL
REACTANCE

WITHDRAWAL
APPROACH

Reactance
reducing
strategies

Number of threats
Number of freedoms

Decrease in threat

Persistence of threat

REACTANCE
REDUCTION

Chadee_c01.indd 38Chadee_c01.indd 38 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Toward Freedom: Reactance Theory Revisited 39

Andreoli, V. A., Worchel, S., & Folger, R. (1974). Implied

threat to behavioural freedom. Journal of Personality and

Social Psychology, 30(6), 765–771.

Baer, R., Hinkle, S., Smith, K., & Fenton, M. (1980).

Reactance as a function of actual versus projected auton-

omy. Journal of Personality and Social Psychology, 38(3),

416–422.

Baker, K., Sullivan, H., & Marszalek, J. (2003). Therapeutic

reactance in a depressed client sample: A comparison of

two measures. Assessment, 10(2), 135–142.

Bargh, J. A., & Apsley, D. K. (2001). Introduction. In J. A.

Bargh & D. K. Apsley (Eds.), Unravelling the complexities

of social life (pp. 3–10). Washington, DC: American

Psychological Association.

Baumeister, R. F., Catanese, K. R., & Wallace, H. M. (2002).

Conquest by force: A narcissitic reactance theory of rape

and sexual coercion. Review of General Psychology, 6(1),

92–135.

Berkowitz, L. (1973). Reactance and the unwillingness to

help others. Psychology Bulletin, 79(5), 310–317.

Berkowitz, L. (1990). On the formation and regulation of

anger and aggression: A cognitive–neoassociationistic

analysis. American Psychologist, 45, 494–503.

Berkowitz, L. (1993). Aggression: Its causes, consequences,

and control. New York: McGraw-Hill.

Berkowitz, L. (1999). Anger. In T. Dalgleish & M. Power

(Eds.), Handbook of cognition and emotion (pp. 411– 428).

Chichester and New York: Wiley.

Berkowitz, L. (2003). Affect, aggression, and antisocial

behavior. In R. J. Davidson, K. R. Scherer, & H. H.

Goldsmith (Eds.), Handbook of affective sciences

(pp. 804–823). New York: Oxford University Press.

Berkowitz, L., & Harmon-Jones, E. (2004). Toward an

understanding of the determinants of anger. Emotion,

4(2), 107–130.

Beutler, L. E., Mohr, D. C., Grawe, K., Engle, D. and

MacDonald, R. (1991), Looking for differential treat-

ment effects: Cross-cultural predictors of differential

psychotherapy efficacy. Journal of Psychotherapy

Integration, 1, 121–141.

Beutler, L. E., Moleiro, C., & Talebi, H. (2002). Resistance in

psychotherapy: What conclusions are supported by

research. JCLP/ In Session: Psychotherapy in Practice,

58(2), 201–217.

Bowen, M. (1978). Family therapy in clinical practice.

New York: Jason Aronson.

Brasted, W., Mann, M., & Geller, E. S. (1979). Behavioral

interventions for litter control: A critical review. Cornell

Journal of Social Relations, 14, 75–90.

Brehm, J. W. (1962). Motivational effects of cognitive disso-

nance. In M. R. Jones (Ed.), Nebraska symposium on moti-

vation (pp. 51–81). Lincoln: University of Nebraska Press.

Brehm, J. W. (1966). A theory of psychological reactance.

London Academic Press.

Brehm, J. W. (2004). Observations from Mount Oread. In

R. A. Wright, J. Greenberg, & S. S. Brehm (Eds.),

Motivational analyses of social behavior: Building on Jack

Brehm’s contributions to psychology (pp. 247–256).

Mahwah, NJ: Erlbaum.

Brehm, J. W., & Cohen, A. R. (1962). Explorations in cogni-

tive dissonance. New York: Wiley.

Brehm, J. W., & Cole, A. H. (1966). Effect of the favour

which reduces freedom. Journal of Personality and Social

Psychology, 3(4), 420–426.

Brehm, J. W., & Mann, M. (1975). Effect of importance of

freedom and attraction to group members on influence

produced by group pressure. Journal of Personality and

Social Psychology, 31(5), 816–824.

Brehm, J. W., & Sensenig, J. (1966). Social influence as a func-

tion of attempted and implied ursurpation of choice.

Journal of Personality and Social Psychology, 4(6), 703–707.

Brehm, J. W., & Wortman, C. B. (1975). Responses to

uncontrollable outcomes: An integration of reactance

theory and the learned helplessness model. In L. Berkowitz

(Ed.), Advances in experimental social psychology (Vol. 8,

pp. 277–336). New York: Academic Press.

Brehm, S. S. (1976). The application of social psychology to

clinical practice. New York: Halsted Press.

Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance:

A theory of freedom and control. New York: Academic

Press.

Brehm, S. S., & Weinraub, M. (1977). Physical barriers and

psychological reactance: 2-year-olds’ responses to threats

to freedom. Journal of Personality and Social Psychology,

35(11), 830–836.

Brockner, J., Gardner, M., Bierman, J., Mahon, T., Thomas,

B., & Weiss, W. (1983). The roles of self esteem and self

conciousness in the Wortman–Brehm model of reactance

References

Chadee_c01.indd 39Chadee_c01.indd 39 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

40 Derek Chadee

and learned helplessness. Journal of Personality and Social

Psychology, 45(1), 199–209.

Buboltz, W. C., Jr., Johnson, P., & Woller, K. M. P. (2003).

Psychological reactance in college students: Family-of-

origin predictors. Journal of Counseling and Development,

81, 311–317.

Buboltz, W. C., Jr., Thomas, A., & Donnell, A. J. (2002).

Evaluating the factor structure and internal consistency

reliability of the Therapeutic Reactance Scale. Journal of

Counseling and Development, 80, 120–125.

Burger, J. M. (1999). The foot-in-the-door compliance

procedure: A multiple-process analysis and review.

Personality and Social Psychology Review, 3(4), 303–325.

Bushman, B. J. (1998). Effects of warning and information

labels on consumption of full-fat, reduced-fat and no-

fat products. Journal of Applied Psychology, 83(1),

97–101.

Bushman, B. J. (2006). Effects of warning and informa-

tion labels on attraction to television violence in view-

ers of different ages. Journal of Applied Social

Psychology, 36(9), 2073–2078.

Bushman, B. J., & Stack, A. D. (1996). Forbidden fruit ver-

sus tainted fruit: Effects of warning labels on attraction

to television violence. Journal of Experimental Psychology:

Applied, 2, 207–226.

Carli, L. L. (1989). Gender differences in interaction style

and influence. Journal of Personality and Social Psychology,

56(4), 565–576.

Carver, C. S. (1977). Self-awareness, perception of threat,

and the expression of reactance through attitude change.

Journal of Personality, 4, 501–512.

Carver, C. S., & Sheier, M. F. (1981). Self-consciousness

and reactance. Journal of Research in Personality, 15,

16–29.

Cherulnik, P. D., & Citrin, M. M. (1974). Individual difference

in psychological reactance: The interaction between locus

of control and mode of elimination of freedom. Journal of

Personality and Social Psychology, 29(3), 398–404.

Clee, M. A., & Wicklund, R. A. (1980). Consumer behav-

iour and psychological reactance. Journal of Consumer

Research, 6, 389–405.

Clore, G. L., Gasper, K., & Garvin, E. (2001). Affect as infor-

mation. In J. P. Forgas (Ed.), Handbook of affect and social

cognition (pp. 121–144). Mahwah, NJ: Erlbaum.

Crawford, M. T., McConnell, A. R., Lewis, A. C., & Sherman,

S. J. (2002). Reactance, compliance and anticipated regret.

Journal of Experimental Psychology, 38, 56–63.

Daily, L. (2004). Navigational web atmospherics: Explaining

the influence of restrictive navigation cues. Journal of

Business Research, 57, 795–803.

Derryberry, D., & Tucker, D. M. (1994). Motivating the

focus of attention. In P. M. Neidenthal & S. Kitayama

(Eds.), The Heart’s eye: Emotional influences in perception

and attention (pp. 176–196). San Diego: Academic Press.

Dillard, J. P., & Shen, L. (2005). On the nature of reactance

and its role in persuasive health communication.

Communication Mono graphs, 72(2), 144–168.

Donnell, A. J., Thomas, A., & Buboltz, W. C., Jr. (2001).

Psychological reactance: Factor structure and internal

consistency of the Questionnaire for the Measurement

of Psychological Reactance. Journal of Social Psychology,

141(5), 679–687.

Dowd, E. T. (1993). Motivational and personality correlates

of psychological reactance and implications for cogni-

tive therapy. Psicologia Conductual, 1(1), 131–140.

Dowd, E. T., Milne, C. R., & Wise, S. L. (1991). The

Therapeutic Reactance Scale: A measure of psychologi-

cal reactance. Journal of Counseling and Development, 69,

541–545.

Dowd, E. T., Trutt, S. D., & Watkins, C. E. (1992) Inter-

pretation style and reactance in counsellor’s social influ-

ence. Psychological Reports, 70, 247–254.

Dowd, E. T., & Wallbrown, F. (1993). Motivational compo-

nents of client reactance. Journal of Counseling and

Development, 71, 533–538.

Dowd, E. T., Wallbrown, F., Sanders, D., & Yesenosky, J. M.

(1994). Psychological reactance and its relationship to

normal personality variables. Cognitive Therapy and

Research, 18(6), 601–612.

Edwards, S. M., Li, H., & Lee, J. (2002). Forced exposure

and psychological reactance: Antecedents and conse-

quences of the perceived intrusiveness of pop-up ads.

Journal of Advertising, 31(3), 83–95.

Festinger, L., & Carlsmith, J. M. (1959). Cognitive conse-

quences of forced compliance. Journal of Abnormal and

Social Psychology, 58, 203–210.

Funkenstein, D. (1955). The psychology of fear and anger.

Scientific American, 192, 74–80.

Goldberg, S., & Lewis, M. (1969). Play behavior in the year-

old infant: Early sex differences. Child Development, 40,

21–32.

Gouldner, A. W. (1960). The norm of reciprocity: A prelimi-

nary statement. American Sociological Review, 25,

161–178.

Chadee_c01.indd 40Chadee_c01.indd 40 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Toward Freedom: Reactance Theory Revisited 41

Hammock, T., & Brehm, J. W. (1966). The attractiveness of

choice alternatives when freedom to choose is eliminated

by a social agent. Journal of Personality, 34, 546–554.

Harmon-Jones, E. (2004). Contributions from research on

anger and cognitive dissonance to understanding the

motivational functions of asymmetrical frontal brain

activity. Biological Psychology, 67, 51–76.

Harmon-Jones, E., & Allen, J. J. B. (1998). Anger and frontal

brain activity: EEG asymmetry consistent with approach

motivation despite negative affect valence. Journal of

Personality and Social Psychology, 74, 1310–1316.

Harmon-Jones, E., & Sigelman, J. (2001). State anger and

prefrontal brain activity: Evidence that insult-related

relative left prefrontal activation is associated with expe-

rienced anger and aggression. Journal of Personality and

Social Psychology, 80, 797–803.

Heilman, M. E., & Toffler, B. L. (1976). Reacting to reac-

tance: An interpersonal interpretation of the need for

freedom. Journal of Experimental Social Psychology, 12(6),

519–529.

Hellman, C. M., & McMillin, W. L. (1997). The relationship

between psychological reactance and self-esteem. Journal

of Social Psychology, 137(1), 135–138.

Homans, G. C. (1974). Social behavior: Its elementary forms.

New York: Harcourt Brace Jovanovich.

Hong, S-M. (1992). Hong’s Psychological Reactance Scale:

A further factor analytic validation. Psychological Reports,

70, 512–514.

Hong, S-M., & Faedda, S. (2006). Refinement of Hong’s

Psychological Reactance Scale. Educational and

Psychological Measurement, 56, 173–180.

Hong, S-M., & Ostini, R. (1989). Further evaluation of

Merz’s Psychological Reactance Scale. Psychological

Reports, 64, 707–710.

Hong, S-M., & Page, S. (1989). A psychological reactance

scale: Development, factor structure and reliability.

Psychological Reports, 64, 1323–1326.

Hong, S.-M., Giannakopoulos, E., Laing, D., & Williams,

N. A. (1994). Psychological reactance: Effects of age and

gender. Journal of Social Psychology, 134, 223–228.

Huck, N. O. (1998). Psychological reactance and Millon’s

personality theory: An integrative comparison. Unpubli-

shed doctoral dissertation, Kent State University,

Kent, OH.

Johnson., E. J., & Tversky, A. (1983). Affect, generalisation,

and the perception of risk. Journal of Personality and

Social Psychology, 45(1), 20–31.

Johnson, P., & Buboltz, W. C., Jr. (2000). Differentiation of

self and psychological reactance. Contemporary Family

Therapy, 22(1), 91–102.

Jonason, P. K. (2007). Further test of validity for the one-

dimensional Hong’s Psychological Reactance Scale.

Psychological Reports, 101, 871–874.

Jonason, P. K., & Knowles, H. M. (2006). A unidimensional

measure of Hong’s Psychological Reactance Scale.

Psychological Reports, 98, 569–579.

Joubert, C. E. (1990). Relationship among self- esteem, psycho-

logical reactance and other personality variables.

Psychological Reports, 66, 1147–1151.

Keltner, D., Ellsworth, P. C., & Edwards, K. (1993). Beyond

simple pessimism: Effects of sadness and anger on social

perception. Journal of Personality and Social Psychology,

64(5), 740–752.

Kornberg, A., Linder, D., & Cooper, J. (1970). Understanding

political behaviour: The relevance of reactance theory.

Midwest Journal of Political Science, 14, 131–138.

Kostic, A., Chadee, D., & Nedeljkovic, J. (in press). Subjective

evaluation of emotional experience in two cultures:

Serbia and Trinidad. In D. Chadee & A. Kostic (Eds.),

Social Psychological Dynamics. Mona, Jamaica: University

of the West Indies Press.

Lazarus, R. S. (1982). Thoughts on the relations between

emotion and cognition. American Psychologist, 37,

1019–1024.

Lazarus, R. S. (1991). Emotion and adaptation. New York:

Oxford University Press.

Lerner, J. S., & Keltner, D. (2001). Fear, anger, and risk. Journal

of Personality and Social Psychology, 81(1), 146–159.

Levenson, R. W., Ekman, P., & Friesen, W. V. (1990). Voluntary

facial action generates emotion-specific autonomic nerv-

ous system activity. Psychophysiology, 27, 363–384.

Linder, D. E., & Crane, K. A. (1970). Reactance theory anal-

ysis of predecisional cognitive processes. Journal of

Personality and Social Psychology, 15(3), 258–264.

Linder, D. E., Wortman, C. B., & Brehm, J. W. (1971).

Temporal changes in predecision preferences among

choice alternatives. Journal of Personality and Social

Psychology, 19(3), 282–284.

Lukin, M., Dowd, E. T., Plake, B. S., & Kraft, R. G. (1985).

Comparing computerised versus traditional psychologi-

cal assessment in a counselling setting. Computers in

Human Behavior, 1, 49–58.

Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of

sex differences. Stanford, CA: Stanford University Press.

Chadee_c01.indd 41Chadee_c01.indd 41 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

42 Derek Chadee

Mallon, K. F. (1992). A scale for assessing psychological reac-

tance proneness: Reliability and validity. Unpublished doc-

toral dissertation. University of Nebraska, Lincoln, NE.

Merz, J. (1983). Fragebogen zur Messung der psycholo-

gischen Reaktanz [Questionnaire for measuring psycho-

logical reactance]. Diagnostica, 29, 75–82.

Mikulincer, M. (1988). Reactance and helplessness follow-

ing exposure to unsolvable problems: The effects of

attributional style. Journal of Personality and Social

Psychology, 54(4), 679–686.

Miller, R. L. (1976). Mere exposure, psychological reactance

and attitude change. Public Opinion Quarterly, 40,

229–233.

Miron, A. M., & Brehm. J. W. (2006). Reactance theory: 40

years later. Zietzchrift für Sozialpsychologie, 37(1), 9–18.

Morgan, R. D. (1986). Individual differences in the occur-

rence of psychological reactance and therapeutic outcome.

Unpublished doctoral dissertation. University of

Nebraska, Lincoln, NE.

Nail, P. R., Van Leeuwen, M. D., & Powell, A. B. (1996). The

effectance versus self-presentational view of reactance:

Are important ratings influenced by anticipated surveil-

lance. Journal of Social Behaviour and Personality, 11(3),

573–584.

Nolen-Hoeksema, S., Fredrickson, B. L., Loftus, G. R., &

Wagenaar, W. A. (2009). Atkinson & Hilgard’s Introduction

to Psychology (15th ed.). Andover, England: Cengage.

Pepper, H. F. (1996). An investigation of the psychosocial

precursors of psychological reactance. Unpublished doc-

toral dissertation. Kent State University, Kent, OH.

Piaget, J. (1981). Intelligence and affectivity: Their relation-

ship during child development. Palo Alto, CA: Annual

Review.

Plutchik, R. (1994). The psychology and biology of emotion.

New York: HarperCollins.

Price, D. D., Barell, J. E., & Barell, J. J. (1985). A quantitative

experiential analysis of human emotions. Motivation

and Emotion, 9, 19–38.

Regan, J. W., & Brehm, J. W. (1972). Compliance in buying

as a function of inducements that threaten freedom. In

L. Bickman & T. Henchy (Eds.), Beyond the laboratory:

Field research in social psychology (pp. 269–274). New

York: McGraw-Hill.

Rhodewalt, F., & Marcroft, M. (1988). Type A behavior and

diabetic control: Implication of psychological reactance

for health outcomes. Journal of Applied Social Psychology,

18 139–159.

Rosenberg, E. L. (1998). Levels of analysis and the organi zation

of affect. Review of General Psychology, 2(3), 247–270.

Rosenzweig, M. R., Breedlove, S. M., & Watson, N. V. (2005).

Biological psychology: An introduction to behavioral and

cognitive neuroscience (4th ed.). Sunderland, MA: Sinaur

Associates.

Seeman, E. A., Buboltz, W. C., Jr., Thomas, A., Soper, B., &

Wilkinson, L. (2005). Normal personality variables and

their relationship to psychological reactance. Individual

Differences in Research, 3(2), 88–98.

Seibel, C. A., & Dowd, E. T. (1999). Reactance and thera-

peutic noncompliance. Cognitive Therapy and Research,

23(4), 373–379.

Seibel, C. A., & Dowd, E. T. (2001). Personality characteris-

tics associated with psychological reactance. Journal of

Clinical Psychology, 57(7), 963–969.

Seltzer, L. F. (1983). Influencing the “shape” of resistance: An

experimental exploration of paradoxical directives and psy-

chological reactance. Basic and Applied Social Psychology,

4(1), 47–71.

Sensenig, J., & Brehm, J. W. (1968). Attitude change from

an implied threat to attitudinal freedom. Journal of

Personality and Social Psychology, 8(4.1), 324–330.

Shen, L., & Dillard, J. P. (2005). Psychometric properties of

Hong’s Psychological Reactance Scale. Journal of Personality

Assessment, 85(1), 74–81.

Shoham, V., Bootzin, R. R, Rohrbaugh, M. J., & Urry, H. (1996).

Paradoxical versus relaxation treatment for insomnia:

The moderating role of reactance. Sleep Research, 24a, 365.

Shoham, V., Trost, S. E., & Rohrbaugh, M. A. (2004). From

state to trait and back again: Reactance theory goes clini-

cal. In R. A. Wright, J. Greenberg, & S. S. Brehm,

Motivational analysis of social behaviour: Building on Jack

Brehm’s contributions to psychology (pp. 167–186).

Mahwah, NJ: Erlbaum.

Shoham-Salomon, V., & Hannah, M. T. (1991). Client–

treatment interaction in the study of differential change

processes. Journal of Consulting and Clinical Psychology, 59,

217–225.

Siemer, M. (2001). Mood-specific effects on appraisal and

emotion judgements. Cognition and Emotion, 15(4),

453–485.

Silvia, P. J. (2005). Deflecting reactance: The role of similar-

ity in increasing compliance and reducing resistance.

Basic and Applied Psychology, 27(3), 277–284.

Silvia, P. J. (2006). A skeptical look at dispositional reactance.

Personality and Individual Differences, 40, 1291–1297.

Chadee_c01.indd 42Chadee_c01.indd 42 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Toward Freedom: Reactance Theory Revisited 43

Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive

appraisal in emotion. Journal of Personality and Social

Psychology, 48(4), 813–838.

Smith, C. A., & Ellsworth, P. C. (1987). Patterns of appraisal

and emotion related to taking an exam. Journal of

Personality and Social Psychology, 52(3), 475–488.

Smith, C. A., & Kirby, L. D. (2001). Toward delivering on

the promise of appraisal theory. In K. R. Scherer, A.

Schorr, & T. Johnstone (Eds.), Appraisal processes in emo-

tion: Theory, methods, research (pp. 121–138). New York:

Oxford University Press.

Smith, C. A., & Lazarus, R. S. (1993). Appraisal compo-

nents, core relational themes, and the emotions.

Cognition and Emotion, 7, 233–269.

Snow, R. E. (1991). Aptitude–treatment interaction as a

framework for research on individual differences in psy-

chotherapy. Journal of Consulting and Clinical Psychology,

59, 205–216.

Snyder, L. B., & Blood, D. J. (1992). Caution: Alcohol adver-

tising and the Surgeon General’s alcohol warnings may

have adverse effects on young adults. Journal of Applied

Communication Research, 20, 37–53.

Snyder, M., &. Wicklund, R. A. (1976). Prior exercise of free-

dom and reactance. Journal of Experimental Social

Psychology, 12(2),120–130.

Stillman, P. G. (1977). Review of the book freedom and

reactance. American Political Science Review, 7(2), 673–674.

Tucker, R., &. Byers, P. Y. (1987). Factorial validity of Merz’s

Psychological Reactance Scale. Psychological Reports, 61,

811–815.

Wicklund, R. A. (1970). Prechoice preference reversal as a

result of threat to decision freedom. Journal of Personality

and Social Psychology, 14(1), 8–17.

Wicklund, R. A. (1974). Freedom and reactance. Potomac,

MD: Erlbaum.

Wicklund, R. A. (1975). Objective self-awareness. Advances

in Experimental Social Psychology, 8, 233–275.

Wicklund, R. A. (1997). The theory of psychological reac-

tance: A reliably revolutionary perspective. Contemporary

Psychology, 42(8), 679–681.

Worchel, S. (1974). The effects of three types of arbitrary

thwarting on the instigation to aggression. Journal of

Personality, 42, 300–318.

Worchel, S. (2004). The diamond in the stone: Exploring the

place of free behavior in studies of human rights and cul-

ture. In S. Wright, J. Greenberg, & S. Brehm (Eds.),

Motivational analyses of social behavior: Building on Jack

Brehm’s contributions to psychology (pp. 107–128). Mahwah,

NJ: Erlbaum.

Worchel, S., & Andreoli, V. A. (1974). Attribution of causal-

ity as a means of restoring behavioral freedom. Journal of

Personality and Social Psychology, 29, 237–245.

Worchel, S., &. Brehm, J. W. (1970). Effect of threats to atti-

tudinal freedom as a function of agreement with the

communicator. Journal of Personality and Social

Psychology, 14(1), 18–22.

Worchel, S., & Brehm, J. W. (1971). Direct and implied

social restoration of freedom. Journal of Personality and

Social Psychology, 18(3), 294–304.

Worchel, S., & Folger, R. (1974). Implied threat to behav-

ioural freedom. Journal of Personality and Social

Psychology, 30(6), 765–771.

Wright, R. A., & Brehm, S. S. (1982). Reactance as impres-

sion management: A critical review. Journal of Personality

and Social Psychology, 42(4), 608–618.

Wright, R. A., Greenberg, J., & Brehm, S. S. (Eds.). (2004).

Motivational analyses of social behavior: Building on Jack

Brehm’s contributions to psychology. Mahwah, NJ: Erlbaum.

Zajonc, B. (1980). Feeling and thinking: Preferences need

no inferences. American Psychologist, 35, 151–175.

Chadee_c01.indd 43Chadee_c01.indd 43 12/13/2010 2:16:52 PM12/13/2010 2:16:52 PM

Festinger’s (1957) theory of cognitive dissonance is arguably one of the most well-known,
studied, powerful, applicable, debated, controversial, and resilient theories in the field of social
psychology, if not psychology as a whole. Unlike more modern theories that are not yet time-
tested, or older theories that have faded into obscurity, cognitive dissonance theory has a long,
colorful, and still unfolding history (see Aronson, 2007; Bendersky & Curhan, 2009; Cooper,
2007; Hart et al., 2009; Nail & MacDonald, 2007; Tavris & Aronson, 2007). Its influence on
basic and applied psychology as well as popular culture has been almost unprecedented. In
the first 15 years after its publication, the theory generated over a thousand experiments
(Thibodeau & Aronson, 1992), most of which supported the theory. In preparing to write this
chapter, we conducted a computer search of the professional psychological literature with the
term dissonance as a key word or subject heading. It yielded more than 1,800 hits.

From the outset, one of the most appealing aspects of the theory has been its ability to
make correct, often counterintuitive predictions. In so doing, it frequently explains, inte-
grates, and offers compelling insights into otherwise mystifying observations of human
behavior, for example: (a) why people report liking a boring task more if they are paid less
for doing it (Festinger & Carlsmith, 1959); (b) why members of various groups such as the
Marines, military academies, college fraternities and sororities, and athletic teams report
high levels of commitment to these groups despite the fact that they had to go through
“Hell on Earth” to become members (Aronson & Mills, 1959; Gleick, 1997; Styron, 1977);
(c) why millions of Americans and seemingly countless authors remain fascinated by, and
construct fantastic conspiracy theories to explain, the Kennedy assassination, this despite
overwhelming scientific evidence that Lee Harvey Oswald acted alone (Bugliosi, 2007;
Manchester, 1967); (d) why religious cults increase the strength of their beliefs after their
leader’s prophecies or group’s beliefs have been unequivocally disconfirmed or discredited
(Batson, 1975; Festinger, Riecken, & Schachter, 1956; Stafford, 2009); (e) why people often

2

Inconsistency in Cognition:
Cognitive Dissonance

Paul R. Nail and Kurt A. Boniecki

Theories in Social Psychology, First Edition. Edited by Derek Chadee.
© 2011 Blackwell Publishing Ltd. Published 2011 by Blackwell Publishing Ltd.

Chadee_c02.indd 44Chadee_c02.indd 44 12/13/2010 2:17:34 PM12/13/2010 2:17:34 PM

Inconsistency in Cognition: Cognitive Dissonance 45

blame innocent victims, whether the victims of fate (Klugel & Smith, 1986) or the targets of
one’s own wrongdoing (Davis & Jones, 1960; Glass, 1964), etc. (Aronson, 1999, p. 108).

Another strong feature of dissonance theory is that its predictions have frequently pre-
vailed when compared to the predictions of competing theories (e.g., Bushman, 2002;
Gerard & Mathewson, 1966; Kahn, 1966). For example, the catharsis hypothesis from
orthodox psychoanalytic theory holds that venting anger through aggression will often help
diminish its lingering, unwanted effects. Thus, expressing aggression should help dissipate
anger, predisposing actors to see their victims more favorably. Almost all of the empirical
evidence, however, is diametrically opposed to this prediction. In line with dissonance the-
ory and (e) above, most evidence indicates that once people behave aggressively toward
others, they have a strong tendency to justify their behavior by derogating their victims.
“When people vent their feelings aggressively they often feel worse, pump up their blood
pressure, and make themselves even angrier” (Tavris & Aronson, 2007, p. 26).

In this chapter, we focus on some of the highlights from the story of cognitive disso-
nance, from its origin, through its evolution, to its decline in the 1980s, to its resurgence
over the last 20 years or so (e.g., Aronson, 1999; Cooper, 2007; Harmon-Jones & Mills, 1999;
Hart et al., 2009; Higgins, 1987; Nail, MacDonald, & Levy, 2000; Steele, 1988). Over the
years, the meaning of the term cognitive dissonance has changed in various and important
ways, often depending on the theorist and the context. Yet, in the concluding section of this
chapter, we suggest a line of research that might restore the original definition of cognitive
dissonance as proposed by Festinger (1957).

Prophecies of Doom

The story begins with a tragedy. On January 15, 1934, a magnitude 8.1 earthquake shook
the Himalayan region of northern India and southern Nepal (Dunn, Auden, Gosh, & Roy,
1939/1981). Over the coming days and months, news reports of the death toll climbed to
over 10,000. While residents of the most devastated areas grappled with rescue and clean-up
operations, extraordinary levels of fear gripped the outlying populace less affected by the
earthquake. Rumors spread that even greater disasters were about to hit these nearby vil-
lages (Prasad, 1950). Although these villagers had only experienced minor tremors, many
believed that floods, cyclones, and further earthquakes were imminent.

Nearly two decades later and a half a world away, a group of social scientists headed by
Leon Festinger was searching the existing literature on the propagation of rumors when
they came across the events following that 1934 earthquake. Festinger and his group were
puzzled, not by how the rumors spread, but rather, why people created them in the first
place. Why in the aftermath of such a natural disaster would people create such frightening,
largely unfounded beliefs? After thinking about this question for some time, Festinger even-
tually hit upon a tentative solution. Perhaps, he reasoned, the rumors of further impending
destruction were being spread not in spite of the fact that they caused increased fear and
anxiety, but rather, because of the fact that fear and anxiety already existed. The rumors
were not created to provoke fear, but to justify it (Festinger, 1957). Thus, the rumors did not
cause the fear; rather, the fear caused the rumors. This germ of an idea – that people might

Chadee_c02.indd 45Chadee_c02.indd 45 12/13/2010 2:17:35 PM12/13/2010 2:17:35 PM

46 Paul R. Nail and Kurt A. Boniecki

create or change their beliefs to justify their current affective state – was the seed that even-
tually grew into cognitive dissonance theory.

The essence of the theory can be summarized as follows: When people become aware
that they simultaneously hold two inconsistent cognitions, they experience an unpleasant
state of psychological tension, which Festinger (1957) called cognitive dissonance. A cogni-
tion can be any bit of knowledge people may hold about the world, their more immediate
environment, or themselves, including their attitudes, beliefs, behaviors, and affective states.
The Asian villagers experienced an involuntary rush of anxiety as a result of the nearby
earthquake and devastation that befell their neighbors. Who would not feel such trepida-
tion after such a near-miss? Yet, having only felt minor tremors, there was no rational expla-
nation for the very high level of fear the villagers were experiencing. In dissonance theory
terms, they knew they felt fearful (first cognition), but they also knew, logically, that there
was not much to fear (second, inconsistent cognition). Because dissonance is aversive and
unpleasant, Festinger proposed that people are motivated to reduce it. One obvious way is
to change one of the inconsistent cognitions. Thus, the Asian villagers could eliminate the
dissonance by creating a belief and convincing themselves that there was indeed something
to fear: Impending doom was about to strike.

What happens when the doom does not occur, however? Imagine that you believe with
tremendous conviction that a great disaster is about to occur on a particular date, such as
the end of the world. Further, imagine that you state this belief publicly and go about mak-
ing numerous preparations for the impending disaster. When that disaster does not come
to pass, Festinger’s theory predicts that you would experience a great deal of cognitive dis-
sonance. However, in this case, reducing the dissonance by changing one of the inconsistent
cognitions may not work. You could not change your knowledge of the evidence in front of
you, namely that the disaster did not occur. You might admit that there was never any
imminent danger after all, but then you would only be exchanging one inconsistency for
another. If there was no danger, why did you say there was? Why did you go about preparing
for the disaster? Are you a fool? Are you crazy? Surely not. Again, you would experience
cognitive dissonance. Anticipating such circumstances, Festinger (1957) offered alternative
methods of dissonance reduction; for example, you could rationalize by adding a third
cognition, one that helps explain away and diminish the inconsistency inherent in the origi-
nal, conflicting cognitions. For instance, you might come to believe that your conviction,
faith, and prayers had, in fact, saved the world.

Festinger had the opportunity to test this very prediction even as he was still working on
the book that would become the first formal statement of cognitive dissonance theory
(Festinger, 1957). He did so by infiltrating a doomsday cult and documenting the members’
behaviors preceding and following the prophesized cataclysm. In the classic, When Prophecy
Fails, Festinger and his colleagues, Henry Riecken and Stanley Schachter (1956), detailed
how they became professed members for a few weeks of a small religious cult whose mem-
bers believed that much of North America, and eventually most of the world, would be
destroyed by a great flood. The flood was to come just before dawn on December 21, 1954,
the consequence of the rising of the lost continents of Mu and Atlantis from the Atlantic
Ocean. Fifteen members of the cult, including Festinger, Riecken, and Schachter, gathered
on the evening of December 20 at the home of the group’s leader, a Mrs. Marian Keech, to

Chadee_c02.indd 46Chadee_c02.indd 46 12/13/2010 2:17:35 PM12/13/2010 2:17:35 PM

Inconsistency in Cognition: Cognitive Dissonance 47

await the disaster. The core members of the group believed that they alone would be saved
from the cataclysm. Mrs. Keech, in communication with a group of extraterrestrials known
as the Guardians through the means of automatic writing, learned that she and her group,
the true believers, would be whisked away to safety just after midnight in the early morning
of December 21 by a flying saucer sent by the Guardians.

Midnight came and went, however, and no spacemen arrived. Word soon came from the
Guardians to be patient, but the group was understandably shaken and becoming increas-
ingly distraught and anxious; the cataclysm was less than seven hours away, and their proph-
esized rescue was apparently not going to happen. At 4:45 a.m., however, Mrs. Keech excitedly
summoned everyone to reassemble in the living room for a major announcement. She had
just received yet another message from the Guardians, but this message was unprecedented
compared to all previous communications. It would change everything. As per the Guardians’
instructions, it was labeled “The Christmas Message to the People of Earth.” Herein the
Guardians proclaimed that “the force of Good and light” emanating from Mrs. Keech and
her followers this very night had moved God to take the unusual step of supernaturally
intervening in the Earth’s geology. The flood was not going to happen! The group’s faith had
averted the disaster! Thus, the Festinger et al. (1956) prediction – that the group would need
to add cognitions or beliefs to help rationalize, explain away, and diminish the inconsistency
inherent in the failed prophecy – was supported (see also Stafford, 2009).

Festinger et al. (1956) further predicted that such rationalizations would require social
support to become sustained and be sufficiently effective. Indeed, the Guardians instructed
the group to inform the newspapers of the miracle. Festinger et al. report that whereas in the
days leading up to December 21 the group had lived in relative secrecy, generally shunning
outsiders, nonbelievers, and the occasional reporter, now the group vigorously sought pub-
licity. If the rest of the world could be convinced of the “truth,” the group’s beliefs and the
rationalization that held them together would gain further legitimacy. The group’s residual
dissonance could be further reduced. In reporting the findings of this study, Festinger et al.
changed the names of the cult leader and her members to protect their identities.

Major Theoretical Concepts

Festinger et al. (1956) provided the first outline of cognitive dissonance theory. In a book
published the very next year, however, Festinger (1957) formalized, expanded, and elabo-
rated the theory in elegant detail. The basic idea that dissonance is caused by cognitive
inconsistency is accurate but inadequate if the theory is to make correct predictions in spe-
cific circumstances. According to Festinger (1957), “elements” of cognition could be conso-
nant, dissonant, or irrelevant to each other. Two elements are consonant if a person perceives
that one element follows logically from another. If you are a Democrat and you vote for a
Democratic candidate, these elements would be consonant. Two elements are dissonant if
the perceived opposite of one element is expected to follow from another element. For
example, being a Democrat and preferring a Republican candidate would be dissonant.
In Festinger’s words, “x and y are dissonant if not-x follows from y” (p. 13). Elements are not
necessarily dissonant just because one element does not follow from another, however; the

Chadee_c02.indd 47Chadee_c02.indd 47 12/13/2010 2:17:35 PM12/13/2010 2:17:35 PM

48 Paul R. Nail and Kurt A. Boniecki

elements may be simply irrelevant. For example, being a Democrat and purchasing a Ford
Mustang would not induce dissonance because the elements have nothing to do with one
another. However, if the Democratic Party began vigorously promoting increases in fuel-
efficiency standards, then being a Democrat and purchasing a gas-guzzling sports car may
become very dissonant indeed. Thus, the addition of further knowledge may cause two
previously irrelevant elements to become relevant to each other, and thereby either conso-
nant or dissonant.

Festinger (1957) noted that any specific element of cognition is, in all likelihood, conso-
nant and dissonant with many other elements. Thus, the total magnitude of dissonance
experienced at any moment is more than the discrepancy between any two elements. Rather,
the total dissonance is dependent on the proportion of all relevant elements that are
dissonant to a given element. It can be expressed in the following formula:

*
D

D
D C

=
+

where D* equals the total magnitude of dissonance, D equals the sum of all elements
dissonant with the element in question, and C equals the sum of all elements consonant
with the same element. For example, if you are a Democrat but decide to vote for a
Republican candidate, you should experience cognitive dissonance. The overall dissonance
could be reduced, however, by adding the consonant cognition that this particular
Republican candidate supports more typically Democratic issues than this year’s Democratic
candidate. Mathematically, adding any consonant cognition to the denominator must nec-
essarily reduce the quotient, thereby reducing the overall magnitude of dissonance.

Festinger (1957) further proposed that the magnitude of dissonance would be affected by
the importance of the elements involved in any particular consonant or dissonant relation – the
more important the dissonant elements, the greater the magnitude; the more important the
consonant elements, the lower the magnitude. Once more, Festinger’s primary thesis was
that cognitive dissonance is psychologically aversive and unpleasant. As a result, a person
experiencing it will feel pressure to reduce or eliminate it entirely. Festinger conceptualized
dissonance as similar to other physical and psychological drives, such as hunger and frus-
tration. Just as someone who is hungry will engage in behaviors to reduce the hunger, a
person experiencing dissonance will engage in behaviors to reduce the dissonance. The
degree of pressure to reduce dissonance will be in direct proportion to the magnitude of the
dissonance felt. The lower the magnitude, the less pressure the person feels. As summarized
above, three factors – the dissonant elements, the consonant elements, and the importance
of the elements – can be manipulated to reduce the magnitude, and thereby the pressure.

Perhaps the most direct method of reducing cognitive dissonance is to make the disso-
nant elements consonant by changing one of the inconsistent elements. If a dissonant
element is based on one’s own behavior, changing the behavior may suffice. For instance, a
smoker, knowing that smoking is bad for one’s health, can just stop smoking. If a dissonant
element is a belief or attitude about the environment, then a person can reduce or eliminate
dissonance by changing that belief or attitude. Thus, a smoker could decide that smoking
does not pose so much of a health hazard after all. Which dissonant element is changed

Chadee_c02.indd 48Chadee_c02.indd 48 12/13/2010 2:17:35 PM12/13/2010 2:17:35 PM

Inconsistency in Cognition: Cognitive Dissonance 49

depends on each element’s resistance to change. Some dissonant behaviors may be seen as
irrevocable, and therefore behavioral change would be impossible. Consider a Democrat
who voted for a Republican candidate; this person cannot return to the voting booth and
vote again. Thus, he or she may abandon affiliation with the Democratic Party and self-
identify now as an independent to reduce the dissonance. Other dissonant behaviors may
be so reinforced by normative social pressure or so addictive (like smoking) that stopping
the behavior is more difficult than changing one’s discrepant beliefs or attitudes. However,
changing one’s belief or attitude is not always easier. People’s cognitions are constrained by
the “reality” they perceive. Convincing oneself that cigarettes are not hazardous to one’s
health may be daunting when each package is emblazoned with the Surgeon General’s
warning and one is inundated by numerous anti-smoking print and television ads.

What happens if the dissonant elements are so resistant to change that a person experi-
encing dissonance is unable to eliminate the discrepancies? Although completely alleviat-
ing cognitive dissonance would require eliminating all of the dissonant elements (so that
the sum of the dissonances would equal zero), again, one could sufficiently reduce disso-
nance by increasing the number of consonant elements. Because the overall magnitude of
cognitive dissonance is dependent on the proportion of dissonant elements, adding conso-
nant elements reduces that proportion and thereby reduces the dissonance. For example,
smokers, if unable to quit or change their beliefs about smoking, may add the consonant
cognition that smoking suppresses their appetite and therefore helps them maintain a
healthy weight. Alternatively, smoking helps them relax and thus reduces unhealthy stress.
The more credible rationalizations a person creates, the less pressure one feels to reduce the
dissonance. Ultimately, the magnitude of dissonance becomes small enough that it is bear-
able and motivates no further action besides avoidance of situations and information that
may add dissonant elements. Of course, adding any consonant element is not always
possible; like changing dissonant elements, rationalizations are constrained by a person’s
subjective reality. Although a rationalization may appear irrational, unbelievable, or non-
credible to others (e.g., Mrs. Keech’s claim that she and her group had saved the world), as
long as it does not contradict that person’s “reality” or, frequently more importantly,
that person’s shared “social reality,” it may effectively reduce dissonance (Stafford, 2009).
The fact that groups often create their own reality points to the important role of yet
another possible, often supplementary method of dissonance reduction – social support
(Festinger et al., 1956).

If people are unable either to change the dissonant elements or to add consonant ele-
ments, aided or not by social support, a possible means of dissonance reduction that
remains open is the importance of the elements. Specifically, people can reduce dissonance
by decreasing the importance of the dissonant elements through cognitive reframing or
distortion – a process Simon, Greenberg, and Brehm (1995) have referred to as trivializa-
tion. For example, people who engage in unhealthy behaviors – smoking, binge drinking,
overeating, steroid use, tanning, etc. – may embrace a “devil-may-care” attitude about their
health. Have you heard, or maybe used, some of the following statements? “Everything that
is fun is bad for you.” “Who wants to live forever?” “If (add one’s vice here) does not kill me,
something else will.” None of these remarks are designed to eliminate the dissonant
elements; people who make such statements still acknowledge the potential hazards of their

Chadee_c02.indd 49Chadee_c02.indd 49 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

50 Paul R. Nail and Kurt A. Boniecki

actions, but they do not plan to quit. Rather, they are minimizing the impact of those dis-
sonant elements by trivializing the goal of staying healthy and living for a long time. By
lowering the importance of the dissonant elements, the overall magnitude of dissonance
decreases. Some of the quotes above may also reduce dissonance by increasing the relative
importance of consonant elements. Thus, when smokers say, “Life is too short not to enjoy
a good cigarette,” they are simultaneously trivializing their values in a healthy life while
strengthening their values in having a pleasurable life. Once more, the social support of
others for any means of dissonance reduction can add to its effectiveness. Many smokers
agreeing that “Life is too short not to enjoy a good cigarette” is more effective than one
smoker saying the same thing either to oneself or in a group of skeptical nonsmokers.

A key prediction of the theory concerns dissonance that is aroused in research settings
when one is induced to engage in behavior that is inconsistent with one’s attitudes or beliefs
(viz., counter-attitudinal behavior). In this situation, the degree of dissonance will be
inversely proportional to the degree of perceived external pressure on the person to engage in
the counter-attitudinal behavior. External pressure could be defined operationally, for exam-
ple, by the degree of financial incentive one is offered to perform the inconsistent behavior,
or the amount of social pressure one is under from the experimenter to behave counter-
attitudinally. Whichever way external pressure is operationalized, if there is too much of it,
there is too much justification for the behavior, and the person should experience little or no
dissonance. Again, the D* = D/D + C formula is informative. Strong external pressures can
be seen as adding consonant cognitions to the dissonance equation, thereby justifying one’s
counter-attitudinal behavior and reducing the overall magnitude of dissonance.

Cognitive dissonance theory was firmly grounded in the developing zeitgeist of social
psychology that began with Kurt Lewin (1943). Indeed, Festinger and Lewin were colleagues
at the Massachusetts Institute of Technology (MIT) in the early 1940s, before Festinger
moved to the University of Minnesota and later to Stanford University. Lewin had immi-
grated to the United States from Germany shortly after Hitler came to power in 1933, bring-
ing with him a Gestalt perspective of social behavior (Blass, 2004). German Gestalt
psychologists had demonstrated that perception went beyond the environmental stimuli
that people experienced through their sensory systems; rather, the Gestaltists convincingly
demonstrated that mental processes organize sensory information into a meaningful whole
that is more than the sensory representations of the environment. Lewin expanded on this
general idea, suggesting that behavior is not only a reaction to environmental stimuli, but
also dependent on a person’s subjective construal of these stimuli. Like Lewin, Festinger
(1957) emphasized this interaction between the person and the environment. The various
factors that contribute to cognitive dissonance – dissonant elements, consonant elements,
social support, and importance – are dependent on individuals’ goals, social groups, cul-
ture, and expectations that influence the “reality” they perceive. Thus, given the same two
relevant cognitive elements, one person may perceive them to be dissonant whereas another
perceives them as consonant. Alternatively, one may perceive the elements to by very impor-
tant whereas another perceives them as trivial.

While Lewin, Festinger, and others (e.g., Heider, 1958) followed this Gestalt-based
approach, other social psychologists such as Carl Hovland, Irving Janis, Harold Kelley, and
John Thibaut were applying learning principles from behaviorism to understanding

Chadee_c02.indd 50Chadee_c02.indd 50 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

Inconsistency in Cognition: Cognitive Dissonance 51

persuasion and attitude change (Hovland, Janis, & Kelley, 1953) and interpersonal relations
(Thibaut & Kelley, 1959). In contrast to the Gestalt approach, behaviorism generally ignored
mental processes and focused primarily on the environmental contingencies that influ-
enced observable behavior. Thus, learning theorists argued that people’s attitudes and
behaviors were the result of reinforcement and punishment; we like those things that ben-
efit us and do not like those things that harm us. During these formative years in the history
of social psychology, Gestalt theorists and behaviorists battled for the soul of the field, and
cognitive dissonance theory became the front line of that battle.

Early Research Paradigms and Tests of Cognitive
Dissonance Theory

Dissonance researchers have created a number of laboratory paradigms over the years
whereby dissonance can be aroused and investigated under controlled conditions. The most
important of these paradigms include induced compliance (e.g., Scher & Cooper, 1989; orig-
inally labeled forced compliance, Festinger & Carlsmith, 1959), free choice (Brehm, 1956),
justification of effort (Aronson & Mills, 1959), forbidden toy (Aronson & Carlsmith, 1963),
blame the victim (Glass, 1964), and hypocrisy (Aronson, Fried, & Stone, 1991). To illustrate,
we focus on the first two of these.

The induced compliance paradigm

In the first laboratory study designed by Festinger to test his new theory, Festinger and
Carlsmith (1959) tackled one of the more provocative and risky predictions of cognitive
dissonance theory – that under certain conditions, people may like a job more the less they
are paid for doing it. This prediction follows directly from the theoretical propositions out-
lined above; yet it runs counter to most people’s intuition, as well as a straightforward
extension of the principles of behaviorism and learning theory, which was the dominant
school of thought in American academic psychology from the 1920s through most of the
1950s and 1960s. Festinger and Carlsmith were about to turn behaviorism on its head.

The basic idea of the induced compliance paradigm is to get participants to do or say
something against that which they would normally say or do (viz., counter-attitudinal
behavior). Dissonance is manipulated by the degree of external pressure brought to bear on
the counter-attitudinal behavior. Festinger and Carlsmith (1959) had each participant in
their study perform a deliberately boring task for 30 minutes, loading and unloading spools
from a tray. Next each participant performed another boring task, turning ornaments on a
pegboard one-quarter turn, one after the other in turn, for another 30 minutes. At this
point, participants were allegedly debriefed, but this “debriefing” was in reality part of the
cover story and the still-unfolding procedure. Participants were told that the study was
really about the effects of expectations on the performance of manual tasks, which had been
designed to simulate assembly-line work, and that they had been in the no expectation
condition. The next participant, however, a female, was to be in the high expectation
condition, but an unanticipated problem had popped up. The student/experimental

Chadee_c02.indd 51Chadee_c02.indd 51 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

52 Paul R. Nail and Kurt A. Boniecki

confederate who was supposed to build up the female participant’s positive expectations
about how fun and enjoyable the experimental tasks were going to be had called in and
would not be able to serve as the confederate that day. As a result, the experimenter was in
a jam and was wondering if the participant might be willing to fill the confederate’s role.
Would the participant be willing to tell the waiting female participant how truly fun and
enjoyable the experimental tasks were?

Next was the manipulation of the independent variable, the external pressure or incen-
tive offered to participants. Some were offered and paid $20 for saying that the tasks were
quite fun and enjoyable, others only $1 (equivalent to $150 and $7.50 in 2010). A third
group of no-dissonance/control participants was not asked to contribute to the manipula-
tion of positive expectations for the next participant. At this point, as the participants were
leaving the experiment, all were told, incidentally, that the Psychology Department was
interested in evaluating the research being conducted by faculty members. Participants
were asked to stop by the office down the hall as they were leaving to help with the evalua-
tions. At the office, participants were asked to rate the experimental tasks regarding how
interesting and enjoyable they had been. The ratings were made on a −5 to +5 scale. Festinger
and Carlsmith (1959) had predicted that $1, unlike $20, would not be sufficient justifica-
tion for participants to say the tasks were fun and enjoyable when they actually were not.
Therefore, $1 participants should be more likely than $20 participants to rate the tasks
favorably. The results supported these predictions, as $1 participants gave the tasks a mod-
erately favorable rating on average (M = +1.35), a rating that was significantly higher than
that of both $20 participants (M = −0.05) and control participants (M = −0.45).

These findings challenged learning theory and the behaviorist establishment. According
to learning theory, the driving forces behind most behavior, human and nonhuman alike,
are reinforcement and punishment. Social psychologists had successfully applied these
same principles to the understanding of attitude formation and change (Hovland et al.,
1953). Simply stated, we should like objects and events that are followed by a reward and
dislike objects and events that are followed by punishment. Thus, learning theorists, as well
as most folk philosophers, would predict that participants paid $20 should rate the tasks
more favorably than those paid $1. However, the evidence was to the contrary, and many
psychologists, believing in the unassailable truth of learning theory, attacked the findings
while offering a litany of alternative explanations (Chapanis & Chapanis, 1964). We address
criticisms and revisions of Festinger’s (1957) theory later in this chapter. Suffice it to say for
the present that science thrives on controversy, and ironically, it was the critics of disso-
nance theory that helped to further refine the theory and establish it as one of the most
well-known and heuristically valuable theories in the history of psychology.

Another provocative aspect of Festinger and Carlsmith’s (1959) research is that it dem-
onstrated that attitudes can be influenced by behavior (behavior → attitudes). At the time,
this finding was outside most psychologists’ ideas about the relationship between attitudes
and behaviors, that attitudes predict or predispose behaviors in a general motivating way
(attitudes → behavior). Ultimately, the research of Festinger and Carlsmith, and other early
research based on the induced compliance paradigm, helped foster the contemporary view
regarding the relationship between behavior and attitudes – that they are reciprocally
related, always potentially both causes and effects of one another. The idea that behavior

Chadee_c02.indd 52Chadee_c02.indd 52 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

Inconsistency in Cognition: Cognitive Dissonance 53

can shape attitudes is especially noteworthy because of its relevance to one of the important
practical applications of dissonance theory we discuss in the final section of this chapter.

Although Festinger and Carlsmith (1959) invented and were the first to use the induced
compliance paradigm, later dissonance researchers have generally relied upon a more
direct variation that was developed by Brehm and Cohen (1962). This variation was pro-
posed in conjunction with one of the early revisions of Festinger’s (1957) theory. Here,
dissonance is induced by having participants write an essay or speech that runs counter
to their true attitudes (e.g., Nel, Helmreich, & Aronson, 1969). In the most frequently
used variation, college students are asked to write a counter-attitudinal essay arguing in
favor of a possible tuition increase at their school (e.g., Cialdini, Trost, & Newsom, 1995;
Steele & Liu, 1983). The experimenter explains to participants that an administrative
committee in charge of making a decision on the matter needs essays on both sides of the
issue so that a balanced judgment can be reached. Because almost all of the essays so far
have been against the increase, the committee really needs essays in favor of the increase.
Importantly, the independent variable is the choice participants are given in writing the
essay. Brehm and Cohen were the first of many subsequent dissonance theorists to empha-
size that one’s counter- attitudinal behavior must be seen by him or her as freely chosen if
it is to arouse much dissonance (e.g., Cooper & Fazio, 1984; Davis & Jones, 1960). Under
high choice, the experimenter politely asks participants, as a favor, to write in support of
the tuition increase. Under low choice, the experimenter just bluntly instructs partici-
pants that their assignment is to write in support of the increase. After writing the essays,
dissonance is usually measured by participants’ post-attitudes regarding the tuition hike.
The typically greater inferred attitude change on the part of high-choice participants as
compared to low-choice participants is taken as evidence of dissonance and its
reduction.

The free choice paradigm

The induced compliance is the most well known of the dissonance paradigms, but Brehm’s
(1956) study using the free choice paradigm was actually the first to test dissonance theory
under controlled conditions. To begin, it is important to note that the word choice has a
different meaning in this context (see below) than it does under the manipulation of choice
in the induced compliance paradigm. Under the guise of a marketing study, Brehm invited
adult women to evaluate eight common household products, a coffee maker, lamp, radio,
etc. The women inspected each product and rank-ordered them from least desirable
(a ranking of 1) to most desirable (a ranking of 8). After the rankings, Brehm told partici-
pants that they could take home one of the products for free as a gift for participating in the
study. However, to ensure that the researcher did not run out of any one product, partici-
pants would need to choose between only two of the products (hence the label, free choice).
In one condition, Brehm gave participants a choice between two highly ranked products,
one ranked 5, 6, or 7 and the other ranked one point lower. In another condition, the choice
was between a highly ranked product (5, 6, or 7) and a product ranked two or three points
lower. Thus, in the former condition, participants faced a difficult choice between two
products that they had ranked nearly equally, whereas in the latter condition, participants

Chadee_c02.indd 53Chadee_c02.indd 53 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

54 Paul R. Nail and Kurt A. Boniecki

faced an easier choice between two, more disparately ranked items. After participants made
their choice, Brehm wrapped the gift for the participants to take home and continued the
study. Participants read consumer reports of four products they had evaluated, including
the two products from which they had to choose only one. After reading the consumer
reports, Brehm asked participants to rank all eight products a second time. This procedure
provided participants with a context and rationale for reevaluating the products as well as
a variety of positive and negative qualities about the chosen and rejected items that might
facilitate participants’ justification of their choice (i.e., reducing their dissonance).

Brehm (1956) predicted that the participants who had made the more difficult choice
would show a greater spread in the rankings of the chosen and rejected items than those
who made the easier choice (the so-called spread of alternatives), and that is exactly what
happened. Participants in the difficult-choice condition elevated their rankings of the cho-
sen items (M = 0.4) and lowered their rankings of the rejected items (M = −0.4), and they
did so significantly more than participants in the easy-choice condition (M = 0.13 and 0,
respectively). In cognitive dissonance terms, the women who were forced to make a tough
choice between two highly ranked products experienced more dissonance than those who
did not, and thus they were more motivated to change their evaluations of the chosen and
rejected items after the fact, apparently to justify their decision and reduce their dissonance.
Although these results did not ignite the same level of interest and controversy as Festinger
and Carlsmith’s (1959) study, Brehm had provided the first experimental evidence in sup-
port of cognitive dissonance theory. He also established the free choice paradigm as a valid
means of measuring dissonance and its effects in future research (e.g., Steele, Spencer, &
Lynch, 1993; Stone, 1999).

Development of the Theory

Whenever a theory starts to develop a reputation for making correct predictions, espe-
cially perhaps when the predictions are counterintuitive, novel, unexpected, or counter to
accepted knowledge, it is terribly tempting to think that the theory is correct, that all
bases have been covered, and that the theory must therefore be true. But this thinking is
wrong. All theories are false. The main reason is because nature, including human nature,
is infinitely subtle and complex. There is simply no way that any human being can antici-
pate all relevant circumstances with 100% accuracy, creating a set of abstract proposi-
tions before the fact, that will apply to all evidence that might have bearing on a theory.
The best a theorist can do is closely approximate the empirical world he or she is trying
to describe and explain, hopefully with closer approximations over time. As new evidence
comes in, all theories must change. And so it is with cognitive dissonance theory. As a
master theorist, Festinger (1987/1999, pp. 382–383) understood this fact about theories
better than most:

No theory is going to be inviolate. Let me put it clearly. The only kind of theory that can be

proposed and ever will be proposed that absolutely will remain inviolate for decades, certainly

centuries, is a theory that is not testable. If a theory is at all testable, it will not remain unchanged.

Chadee_c02.indd 54Chadee_c02.indd 54 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

Inconsistency in Cognition: Cognitive Dissonance 55

It has to change. All theories are wrong. One doesn’t ask about theories, can I show that they

are wrong or can I show that they are right, but rather one asks, how much of the empirical

realm can it handle and how must it be modified and changed as it matures?

Proposed revisions of Festinger’s (1957) theory started appearing almost before the
ink was dry on the first articles that reported support for the theory (e.g., Aronson,
1960, cited in Aronson, 1999; Brehm & Cohen, 1962; Rosenberg, 1965b). One reason
for these early revisions is that it soon became apparent to researchers that cognitive
inconsistencies do not always lead to measurable dissonance effects. “About 1958, the
standing joke among Festinger’s research assistants was, ‘If you really want to be sure
whether A is dissonant with B, ask Leon’ ” (Aronson, 1999, p. 109). The revisions of
Aronson, Brehm and Cohen, Rosenberg, and others (e.g., Cialdini et al., 1995; Tedeschi,
Schlenker, & Bonoma, 1971) were designed to tighten the conditions under which
empirical dissonance effects would most likely occur. Once more, for example, Brehm
and Cohen emphasized that participants’ counter-attitudinal behavior must be per-
ceived as freely chosen.

The Aronson (1960), Brehm and Cohen (1962), Tedeschi et al. (1971), and other early
revisions, however, are properly regarded as just that – revisions – because each retained
Festinger’s (1957) core idea that psychological inconsistency is the driving force that
underlies dissonance motivation. Other investigators, in contrast, have constructed
entirely or substantially new theories that attempt to explain dissonance effects with psy-
chological constructs other than the aversive tension of psychological inconsistency (e.g.,
Bem, 1967; Cooper & Fazio, 1984; Steele, 1988). Bem’s self-perception theory, for exam-
ple, is not a motivational theory at all, fitting as it does equally well within either the
information-processing or behavioral camps (see below). What is more, many of the early
and more recent efforts have extended dissonance-related work to domains that far exceed
the boundaries of dissonance theory as originally conceived by Festinger (e.g., Adams,
1965; Higgins, 1987; McGregor, Nail, Marigold, & Kang, 2005; Swann, 1984, 1996; Tesser,
1988; Wicklund & Gollwitzer, 1982). Other efforts have restricted dissonance theory to
the point that some theories only include a small fraction of the empirical world that
Festinger (1957) had originally envisioned (e.g., Cooper & Fazio, 1984; Rosenberg, 1965b;
Tedeschi et al., 1971). Yet all of these theories have generated their own bodies of sup-
portive and non-supportive evidence. Each theory can explain certain aspects of the
empirical world that no other theory can handle; still, no one theory can explain every-
thing. Complete consideration of all of this work would likely require an entire book
(cf. Cooper, 2007) and is therefore well beyond the scope of the present chapter. Accordingly,
we focus here on what we believe are a few of the most important developments in
dissonance theory and related work over the years. In order to fairly illustrate the level,
subtlety, and complexity of the theoretical issues involved, we have chosen to focus in
depth on a few issues rather than to have a necessarily shallow coverage of many issues. We
restrict ourselves to either (a) theories that use Festinger (1957) as a starting point and
propose various modifications therefrom or (b) entirely new theories, but ones that use
alternative interpretations of standard dissonance effects in the literature as a point of
departure from Festinger (1957).

Chadee_c02.indd 55Chadee_c02.indd 55 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

56 Paul R. Nail and Kurt A. Boniecki

Self-consistency theory (Aronson, 1960, cited in Aronson, 1999)

The earliest revision of dissonance theory was proposed by Elliot Aronson (1960, cited in
Aronson, 1999), who was a student of Festinger’s at Stanford beginning in the fall of 1956.
As a result of Aronson’s own research and his ongoing debates with Festinger, Aronson
eventually concluded that cognitive dissonance is more concerned with a person’s cogni-
tions about his or her self-concept and the need for consistency within one’s attitudes,
beliefs, and behaviors than it is with any need for a rationale or logic-like consistency, as
Festinger (1957) had originally posited. To clarify, dissonance in Aronson’s theory is not just
a conflict between any two relevant, important cognitions. Rather, dissonance theory makes
its clearest and strongest predictions when the inconsistent cognitions involve important
elements of a person’s self-concept. Looking back over the early work that had supported
dissonance theory, Aronson discovered in each case that the manipulations had challenged
widely desired self-conceptions, such as the desire to see one’s self as fair, honest, reasonable,
competent, and in control of one’s own behavior.

Consider the behavior of participants in Festinger and Carlsmith’s (1959) experiment,
for example. It is true enough that participants’ telling the incoming participant that the
experimental tasks were really fun and enjoyable is inconsistent with the fact that the
tasks were actually quite boring. What caused the dissonance from Aronson’s perspective,
however, was that telling a lie for no good reason ($1, insufficient justification) is incon-
sistent with most people’s valued self-conception that they are basically good and decent
human beings. Simply stated, good and honorable people do not do such morally ques-
tionable things. Yet a positive self-image can be essentially maintained in this context after
the fact if $1 participants say to themselves, in effect, “Come to think of it, the study
wasn’t that bad. Actually, there were all those spools and gadgets; being in the study was
really sort of fun when you think about it.” Note that in such thinking, $1 participants are
changing their attitude, thereby reducing the degree of the inconsistency of their lie, jus-
tifying their behavior to themselves, and thus reducing the magnitude of their dissonance
(i.e., self-inconsistency).

For $20 participants, in contrast, the situation is very different. These participants have
a good reason for lying – the $20 (sufficient justification). Thus, their self-concepts are not
threatened, and there is no need to justify their behavior after the fact. These hypothesized
processes explain why $1 participants changed their attitudes about the experiment,
reporting a moderately favorable reaction to it (M = +1.35), whereas $20 participants did
not (M = −0.05). In Aronson’s version of dissonance theory, dissonance arousal is centered
upon inconsistency with one’s self-concept, and dissonance reduction is all about self-
justification. Accordingly, the theory is frequently referred to as self-consistency theory. The
core aspects of most people’s self-concept according to self-consistency theory is that they
see themselves as competent, as morally good, and as a coherent whole (Aronson, 1968,
1999; Thibodeau & Aronson, 1992). Information that is at variance with people’s desired
self-views that they are fair, honest, reasonable, and competent causes dissonance; infor-
mation that does not, does not.

Because Aronson’s (1968, 1999) theory represents only a modification of Festinger’s
(1957) theory, it is important to note that Aronson’s theory incorporates other key aspects

Chadee_c02.indd 56Chadee_c02.indd 56 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

Inconsistency in Cognition: Cognitive Dissonance 57

of Festinger’s theory unaltered. For example, the total magnitude of dissonance in Aronson’s
theory can still be represented by the D* = D/D + C equation, and D* can be affected by both
the perceived importance of each element and the degree of social support. The only differ-
ence is that the D in Aronson’s theory concerns inconsistency with one’s self-concept,
whereas in Festinger’s theory D is more of a logic-like inconsistency.

Individual differences

One critical aspect of Aronson’s (1968, 1999) modified theory is that it led to testable pre-
dictions that would not have been readily apparent or very likely under Festinger’s (1957)
original formulation. In particular, Aronson’s self-consistency theory pointed to the poten-
tial for individual differences to become easily integrated within the theory. Consideration
of individual differences has led to a great deal of new, interesting, and enlightening research
(e.g., McGregor et al., 2005; Nail, Misak, & Davis, 2004; Steele et al., 1993). Festinger (1957)
had acknowledged from the outset that individual differences in sensitivity to cognitive dis-
sonance likely existed, but he gave little indication as to what these individual differences
might be or how to measure them. Under Aronson’s revised theory, however, the implica-
tions were clear. “Festinger’s original statement and all of the early experiments rested on
the implicit assumption that most people have a reasonably positive self-concept. But if a
person considered himself or herself to be a ‘schnook,’ he or she might expect to do sch-
nooky things” (Aronson, 1999, p. 111). The idea is that “schnooks” should not experience as
much dissonance in typical dissonance-arousing circumstances as “non-schnooks.” This
hypothesis has been supported in a number of studies (e.g., Aronson & Carlsmith, 1963;
Epstein, 1969; Swann & Read, 1981; Szabo, 1985).

Manipulators versus non-manipulators. We regard Aronson’s (1968, 1999) provision for
the inclusion and integration of individual differences within dissonance theory as one of
the singular achievements in the dissonance literature over the years (e.g., Bishop, 1967;
Cialdini et al., 1995; Glass, 1964; McGregor et al., 2005; Steele et al., 1993). In one particu-
larly relevant study, Szabo (1985) manipulated dissonance by having participants write
counter-attitudinal essays for a relatively large versus small monetary incentive. Importantly,
participants had been pre-classified as “manipulators” or “non-manipulators” based on a
Machiavellianism scale. The results indicated that non-manipulators showed the standard
insufficient justification effect; they displayed significantly greater attitude change in the
direction of their counter-attitudinal behavior under small incentive (M = 9.03) than large
(M = 7.6). For manipulators, in contrast, there was no reliable difference between the small
and large incentive conditions (M = 8.4 vs. 8.6, respectively). Apparently, because lying is
not atypical for manipulators, doing so is not at odds with their self-image and therefore
produces no particular dissonance that needs reducing. Interestingly, based on Aronson’s
(1968, 1999) theory and Szabo’s (1985) results (see also Epstein, 1969; Snyder & Tanke,
1976), manipulators should only experience dissonance under conditions where they might
get what they desire by lying, but do not lie, and have no good reason for not doing so! This
provocative hypothesis has not been put to the test to our knowledge, however.

Preference for consistency (PFC). Perhaps the most important body of work concerning
individual differences and cognitive dissonance is that of Cialdini and colleagues (e.g.,

Chadee_c02.indd 57Chadee_c02.indd 57 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

58 Paul R. Nail and Kurt A. Boniecki

Bator & Cialdini, 2006; Cialdini et al., 1995; Guadagno, Asher, Demaine, & Cialdini, 2001;
see also Nail, Bedell, & Little, 2003; Nail et al., 2001). Across a number of studies, Cialdini
and colleagues have shown that individuals scoring high on the Cialdini et al. (1995) PFC
scale generally behave consistently with dissonance theory, showing the typical dissonance-
reducing attitude change following counter-attitudinal behavior under high choice, whereas
those scoring low in PFC either show significantly less attitudinal-behavioral consistency
than high PFCs (Nail et al., 2001), none at all (e.g., Cialdini et al., 1995), or even anti-
consistency (Guadagno et al., 2001). Festinger (1957) originally conceived the need for
cognitive consistency as a nomothetic motive, meaning one applying to people, in general.
Work on PFC indicates, however, that Festinger’s hypothesized need for consistency may
apply strongly to as little as one-third of the population, thus significantly reducing the
scope of dissonance theory. However, in proposing alternative accounts of dissonance
phenomena, other theorists have suggested that standard dissonance effects in the litera-
ture have little or nothing to do with the need for either cognitive or self- consistency
(Bem, 1967; Cooper & Fazio, 1984; Steele, 1988). We turn now to the first of two such
accounts – self-perception theory and self-affirmation theory – that we will consider.

Self-perception theory (Bem, 1967)

Unlike Aronson’s (1968, 1969) self-consistency theory, Bem’s (1967) self-perception theory
is more than just a revision of Festinger’s (1957) theory. Further, steeped in the tradition of
American behaviorism, Bem’s theory has nothing to do with psychological drives of any
sort, much less the hypothesized tension that comes with either cognitive inconsistency à la
Festinger or self-inconsistency à la Aronson. Rather, the theory assumes that people are
often unsure of their own attitudes, feelings, and motives. To discover these, as the name of
the theory suggests, people simply watch or monitor their own behavior and then infer,
after the fact, what their attitudes and feelings must be, given the overall context in which
their behavior occurred. Self-perception processes are evident in the report of John Dean,
who was White House Counsel to President Nixon in the early 1970s before the Watergate
scandal. One day in July of 1970, when Dean was still working at the Justice Department, he
received a phone call from a White House staffer. The caller informed Dean that if he wanted
to go work for the President, he should make arrangements to catch the next flight to
California where he would meet Nixon, be interviewed by Nixon’s chief of staff, and spend
the weekend with some of Nixon’s top aides at the “Western White House.” In the made-for-
TV version of his book, Dean (1976) reports that he was never aware how very much
he was interested in politics and in climbing the ranks of power in Washington until after
he received that phone call. Judging from his behavior, however – how fast he booked the
next available flight and “dashed home to pack” (p. 13) – he guessed he had never wanted
anything so badly in all his life.

To further illustrate self-perception theory, consider once more the behavior of Festinger
and Carlsmith’s (1959) participants. Recall that after being in the study and telling the next
participant how interesting and enjoyable the tasks were, participants were asked to rate the
tasks. According to Bem, in the $1 condition participants would think something like this:
“Let’s see. I just told that girl that the tasks were really fun and enjoyable, and I only got paid $1.

Chadee_c02.indd 58Chadee_c02.indd 58 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

Inconsistency in Cognition: Cognitive Dissonance 59

So, I must have thought that the tasks were pretty enjoyable at least because I wouldn’t say
something that is untrue for only $1. I’ll give the tasks a moderately favorable rating.”
In the $20 condition, however, the hypothesized internal dialogue goes more like this: “Let’s
see. I just told that girl that the tasks were really fun and enjoyable, but I got paid $20, which
to me is a lot of money. I guess I probably told the girl that so I could get the $20. I would
have to rate the tasks, therefore, as not very enjoyable.” Support for self-perception theory
was obtained in a series of studies by Bem and his students in which the Festinger and
Carlsmith (1959) procedure was merely described to groups of observers. Importantly, the
observers produced the same pattern of results as the original participants by simply guess-
ing what the attitudes of original participants must have been in the $1, $20, and control
conditions (M = +0.52, −1.96, and −1.56, respectively; Bem, 1968; see also Bem & McConnell,
1970). Because the observers engaged in no attitudinal-behavioral or self inconsistency,
Festinger’s (1957) and Aronson’s (1968) versions of dissonance theory can not easily explain
these findings.

Dissonance and physiological arousal

The debate between Festinger’s (1957), Aronson’s (1968), and other consistency-based
accounts of dissonance phenomena on the one hand (e.g., Rosenberg, 1965b; Tedeschi
et al., 1971) and a self-perception account on the other raged in the literature for a good
part of the late 1960s and 1970s (Fazio, Zanna, & Cooper, 1977). The debate eventually
subsided, however, aided largely by the discovery that dissonance manipulations are often
accompanied by physiological arousal. This possibility was first suggested in the work of
Pallak and Pittman (1972), but with only an indirect measure of arousal (viz., consistent
with Zajonc’s, 1965, drive theory of social facilitation, dissonance manipulations had the
effect of energizing whatever responses were dominant in a particular setting). However,
the idea that dissonance manipulations cause physiological arousal has been supported in
numerous subsequent dissonance studies based on direct measures of arousal (e.g., GSR or
heart rate; Croyle & Cooper, 1983; Elkin & Leippe, 1986; Nail, Harton, & Decker, 2003).
Festinger (1957), his students, and other early dissonance researchers had often used terms
such as drive, arousal, drive-like, and drive reduction to describe dissonance processes, but
before Pallak and Pittman (1972), no one had ever gone so far as to suggest that the hypoth-
esized psychological arousal of dissonance might be accompanied by literal physiological
arousal as well. Nevertheless, this idea fit the theory neatly, and it had a great deal of intui-
tive appeal in retrospect. The presence of physiological arousal in dissonance research was
regarded by Festinger (1987/1999) as “an extremely important finding” (p. 384). Moreover,
it provided support for drive-based accounts of dissonance phenomena (Aronson, 1968;
Festinger, 1957) over self-perception theory (Bem, 1967).

Ever since the late 1970s, a truce of sorts between the drive-based and self-perception
accounts of dissonance phenomena has prevailed. It appears that self-perception processes
can and do occur but that these processes apply more to situations where people’s baseline
attitudes are either (a) not yet formed or (b) unclear and ambiguous. In situations where
people violate their clearly defined, preexisting attitudes, in contrast, psychological and
physiological arousal can and do occur, leading frequently to attitude change. In short,

Chadee_c02.indd 59Chadee_c02.indd 59 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

60 Paul R. Nail and Kurt A. Boniecki

self-perception theory applies fundamentally to attitude formation (e.g., John Dean),
whereas dissonance theory applies more to attitude change (Fazio et al., 1977). Further,
based on the Cialdini et al. (1995, pp. 325–326) conclusions regarding individual differences
in PFC and attitude change, it appears that the self-perception process may apply more to
people scoring low in PFC, whereas dissonance processes apply more to high PFCs.

Near-death and rebirth

At some point in the 1960s, cognitive dissonance theory became the dominant theoretical
orientation in social psychology. It remained so through much of the 1970s, and some still
regard Festinger (1957) as the single most important work in the history of social psychol-
ogy (Aronson, 1999; Nail, 2006). Yet, with the resolution of the conflict with Bem’s (1967)
self-perception theory (Fazio et al., 1977), and the rise of attribution theory and the infor-
mation-processing paradigm of the 1970s and 1980s, interest in cognitive dissonance the-
ory eventually waned. A key reason for this general lack of interest was stated succinctly by
“Ned” Jones (1985), one of the preeminent social psychologists from the 1960s until his
death in 1993: “Because the main propositions of dissonance theory have been confirmed
with sufficient regularity, there is not a great deal to be gained from further research in this
area” (p. 57). Uncharacteristically, in this instance, time has proven Jones wrong. At about
the time that Jones’s words were going to press, a number of different theorists working
independently began to consider more seriously and in divergent detail fuller implications
of Aronson’s (1968) proposal that dissonance processes reside in the self-concept (e.g.,
Higgins, 1987; Steele, 1988; Swann, 1984, 1996; Tesser, 1988; Wicklund & Gollwitzer, 1982).
Their work has led to renewed interest in dissonance theory, or at least in dissonance-
relevant phenomena, that persists to this day (e.g., Cooper, 2007; Hart et al., 2009; McGregor
et al., 2005; Nail, McGregor, Drinkwater, Steele, & Thompson, 2009). Steele’s (1988) self-
affirmation theory is one of the more prominent of these theories.

Self-affirmation theory (Steele, 1988)

Similar to Bem’s (1967) self-perception theory, Steele’s (1988) self-affirmation theory can
explain much of the dissonance literature, and it stands as an alternative and competing
theory of dissonance phenomena rather than just a modification of Festinger (1957; see
Aronson, Cohen, & Nail, 1999). Unlike Bem’s theory, however, self-affirmation theory can
accommodate the finding that psychological dissonance is physiologically arousing. Steele’s
theory is highly similar to Aronson’s (1968) self-consistency theory in that both theories
point to the critical role of the self in dissonance processes. Consequently, both theories
conceptualize dissonance as being grounded in the ego and not in a free-standing need for
cognitive consistency (Greenwald & Ronis, 1978, p. 55). However, unlike Aronson, but
similar to Bem (1967), Steele maintains that cognitive inconsistency is not at the heart of
dissonance motivation. According to Steele, dissonance is not the aversive tension of logic-
like inconsistency (Festinger, 1957), or even the tension of self-inconsistency (Aronson,
1968); rather, it is the tension of a “threatened sense of self-integrity” (Steele et al., 1993,
p. 893). Steele et al. postulate the existence of a “self system for maintaining a perception of

Chadee_c02.indd 60Chadee_c02.indd 60 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

Inconsistency in Cognition: Cognitive Dissonance 61

global integrity, that is, of overall moral and adaptive adequacy” (p. 885). Dissonance
occurs when one learns something or engages in behavior that threatens this overall
perception of self-integrity. Thus, attitude change in the induced-compliance paradigm
occurs not to reduce inconsistency, but rather to restore one’s overall sense of a good, com-
petent, and stable self-image.

Based on self-affirmation theory, Steele and Liu (1983, Study 1) predicted that partici-
pants would have no problem tolerating inconsistency provided that the experimental pro-
cedure afforded the opportunity to reaffirm values important to participants before
dissonance was measured. Students wrote essays in favor of a large tuition increase at their
university under high- or low-choice conditions. Immediately after writing the essay but
before attitude assessment, some high-choice participants were reminded of a pre- measured,
important component of their self-concept by completing an economic–political values
scale (value-oriented participants). Other high-choice participants went through the same
procedure, but they were chosen for the study precisely because economic–political val-
ues were not important to them (non-value-oriented participants). By definition, values
cannot be self-affirming if they are not important to a person. As predicted, the values scale
eliminated any semblance of attitude change among high-choice/value-oriented partici-
pants (M = 3.6) but not among high-choice/non-value-oriented participants (M = 9.1);
low-choice/control participants actually reported non-significantly greater attitude change
(M = 4.3) than high-choice/value-oriented participants (M = 3.6).1

According to Steele and Liu, these findings support self-affirmation theory over Festinger
(1957) because the values scale did nothing to reduce the inconsistency between attitude
and behavior among high-choice/value-oriented participants; still, these participants
revealed no tendency to rationalize their behavior. Stated differently, high-choice/value-
oriented participants experienced attitudinal-behavioral inconsistency but no threat to the
self because of the self-affirmational intervention. High-choice/non-value-oriented par-
ticipants, however, experienced both attitudinal-behavioral inconsistency and a threat to
the self because the affirmational intervention could not have been affirming to them. The
bottom line across both groups is that attitude change varied as a function of the threat to
the self inherent in writing the counter-attitudinal essay, not with attitudinal-behavioral
inconsistency, per se. Thus, the threat to the self must have been the cause of the dissonance
experienced by high-choice/non-value-oriented participants rather than the inconsistency.

Is self-affirmation theory necessary? Part I. The basic self-affirmation effect – that partici-
pants show no need to rationalize their counter-attitudinal behavior if they are reminded of
important, self-relevant values before their post attitudes are measured – has been repli-
cated many times (e.g., McGregor et al., 2005, Study 3; McGregor, Zanna, Holmes, &
Spencer, 2001; Steele & Liu, 1983, Study 3; Tesser & Cornell, 1991). Yet traditional disso-
nance/inconsistency theorists have questioned the need for a new and separate theory to
explain the dissonance-reducing benefits of self-affirmation (e.g., Beauvois & Joule, 1996;
Simon et al., 1995; Thibodeau & Aronson, 1992). After all, as Thibodeau and Aronson
(1992) have pointed out, adding new, consonant cognitions to the dissonant elements was
one of Festinger’s (1957) originally hypothesized modes of dissonance reduction. Such cog-
nitions are hypothesized to reduce dissonance because, once more, the absolute magnitude
of dissonance in both Festinger’s (1957) and Aronson’s (1968) theories is defined by the

Chadee_c02.indd 61Chadee_c02.indd 61 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

62 Paul R. Nail and Kurt A. Boniecki

ratio of dissonant elements to dissonant plus consonant elements (D* = D/D + C). Thibodeau
and Aronson asserted that once participants’ central values have been affirmed, they have
little need for attitude change because the affirmational intervention reminds participants
of valued, self-relevant cognitive elements that are consonant with a positive self-concept
overall. Given Aronson’s (1968, 1999) theoretical position that all dissonance processes
reside within one’s self-concept, once these consonant elements are added to the dissonance
equation, the magnitude of dissonance is quite obviously reduced.

The role of self-esteem. Steele et al. (1993) subsequently created a more pointed test between
self-affirmation theory and any of the consistency-based accounts (Aronson, 1968; Festinger,
1957; Swann, 1984). This research was based on the idea that consistency-based theories and
self-affirmation theory make conflicting predictions when considering individuals with
high versus low self-esteem. According to self-consistency theory, for example, high self-
esteem individuals should experience more conflict in dissonance-arousing situations than
low-esteem individuals (Aronson & Mettee, 1968). Assuming that self-inconsistency is the
driving force in dissonance motivation, it follows that the possibility of a bad choice as in the
free choice paradigm, for example, would be more inconsistent with the self-concept of high
self-esteem individuals than with that of low self-esteem individuals. Thus, high self-esteem
individuals should display greater dissonance-reducing rationalization – a greater spread of
alternatives – than lows. In sharp contrast, self-affirmation theory predicts that high self-
esteem individuals will experience less dissonance than individuals with low self-esteem.
High self-esteem individuals have positive self-concepts precisely because they have large
reservoirs of positive information, attributes, and thoughts regarding the self. Assuming that
a threatened sense of self-integrity is the driving force in dissonance motivation, it follows
that high self-esteem individuals would have greater self-resources to draw upon than low
self-esteem individuals. Thus, high self-esteem individuals should display a smaller spread
of alternatives following a difficult choice than low self-esteem individuals.

Steele et al. (1993, Study 2) tested these ideas in a 2 × 2 design that crossed trait high/low
self-esteem × focus/no-focus on one’s self-concept. Trait high/low self-esteem was deter-
mined based on a median split of scores on the Rosenberg (1965a) scale, which was admin-
istered several weeks before the experimental sessions began. Self-focus/no-self-focus was
manipulated by having randomly selected participants retake the Rosenberg scale at the
start of the experimental session or not. Dissonance was evoked employing a choice between
two similarly valued music CDs in the free-choice paradigm. The results were clear. They
showed that in the no-self-focus condition, there was no difference in the spread of alterna-
tives between high (M = 0.62) and low self-esteem (M = 0.56) participants. In the self-focus
condition, however, high self-esteem participants (M = −0.29) displayed a significantly
lower spread of alternatives than low self-esteem participants (M = 1.14). These results sup-
port self-affirmation theory over self-consistency theory. Presumably, the self-focus manip-
ulation caused participants to reflect on their self-concepts and self-resources. Self-reflection
resulted in low self-esteem participants rationalizing their choice more than high self-
esteem participants. Apparently, with their self-systems primed by the Rosenberg (1965a)
scale, high self-esteem participants simply had no need to rationalize their choice. In a con-
ceptual replication of Steele et al., similar results favoring self-affirmation theory were
reported by Nail et al. (2004).

Chadee_c02.indd 62Chadee_c02.indd 62 12/13/2010 2:17:37 PM12/13/2010 2:17:37 PM

Inconsistency in Cognition: Cognitive Dissonance 63

Not all studies examining trait self-esteem, however, are consistent with self-affirmation
theory. Stone (1999, Experiment 1) replicated the Steele et al. (1993, Study 2) research but
varied the order of critical events in one condition. When Stone repeated the Steele et al.
ordering of events exactly, he found the same results; low self-esteem individuals rational-
ized their CD choice (M = 2.48) more than high self-esteem individuals (M = 0.55), thus
supporting self-affirmation theory. When Stone had some participants retake the Rosenberg
(1965a) scale after choosing one of the CDs but before the re-rating, however, the pattern
was reversed. High self-esteem participants showed greater self-justifying rationalization
(M = 1.35) than low self-esteem participants (M = 0.52), thus supporting self-consistency
theory. Further, at least three studies of manipulated rather than trait self-esteem have simi-
larly found that high self-esteem individuals display greater dissonance-reducing attitude
change than low self-esteem individuals, again supporting self-consistency theory (Aronson &
Mettee, 1968; Gerard, Blevans, & Malcolm, 1964; Glass, 1964).

To explain the complex relationship between self-esteem and the experience of cognitive
dissonance, Stone and Cooper (2001) have proposed an integration between self-affirma-
tion theory and self-consistency theory, the self-standards model of cognitive dissonance.
Stone and Cooper posit that people’s self-concepts are flexible enough to produce either a
self-affirmation effect à la Steele (1988) or a self-discrepancy effect à la Aronson (1968, 1999),
depending on the details of the situation. More specifically, the self can serve either as (a) a
resource that can buffer self-threats or (b) a standard that one may or may not have lived
up to. In essence, Stone and Cooper propose that if a study’s procedures and manipulations
remind participants of positive self-attributes that are irrelevant to the dissonance in ques-
tion, then the self should serve as a resource and high self-esteem individuals should display
less dissonance-reducing behavior than low self-esteem individuals – a self-affirmation
effect in line with self-affirmation theory (e.g., Nail et al., 2004; Steele et al., 1993; Stone,
1999). On the other hand, if a study’s procedures and manipulations remind participants of
positive self-attributes that are relevant to the dissonance in question but that they have
not lived up to, then the self should serve as a standard and high self-esteem individuals
should display more dissonance-reducing behavior than low self-esteem individuals – a
self-discrepancy effect in line with self-consistency theory (e.g., Aronson & Metee, 1968;
Glass, 1964; Stone, 1999).

Is self-affirmation theory necessary? Part II. Instead of attempting to integrate self-
affirmation theory and self-consistency theory, Simon et al. (1995) have proposed an expla-
nation of the self-affirmation effect that supports the Festinger (1957) and Aronson (1968)
consistency-based theories over self-affirmation theory. Simon et al. propose that self-
affirmation interventions may eliminate the need for dissonance reduction not because
they restore one’s self-image but because they establish a trivializing frame of reference
whereby the relative importance of dissonant elements can be reduced. As noted in the
third major section of this chapter, one of Festinger’s (1957) key, originally proposed meth-
ods of dissonance reduction is to decrease the importance of dissonant cognitions. Simon
et al. found that reminding high-choice participants of a generally important issue (e.g., world
hunger) following counter-attitudinal behavior eliminated the need for dissonance reduc-
tion in the form of attitude change whether or not the issue was of high personal importance
to participants (M = 3.25 and 3.27, respectively). These means did not differ significantly

Chadee_c02.indd 63Chadee_c02.indd 63 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

64 Paul R. Nail and Kurt A. Boniecki

from low-choice control participants (M = 2.55), whereas standard high-choice partici-
pants showed significantly greater attitude change than all three groups (M = 5.75). These
findings create problems for self-affirmation theory because, by definition, and consistent
with Steele and Liu (1983), an issue cannot be self-affirming unless it is of high personal
importance. At the same time, these data provide evidence consistent with dissonance proc-
esses as originally conceived by Festinger (1957) and as modified by Aronson (1968, 1999).

Evaluation of the Theory and Conclusions

In evaluating cognitive dissonance theory, it is sometimes difficult to separate Festinger’s
(1957) theory from Aronson’s (1968, 1999) version. First, with the exception of the distinc-
tion between logic-like inconsistency versus self-inconsistency, the theories are identical.
Second, given the prominence of the self-concept in contemporary social psychology,
Aronson’s theory is what many authors and researchers mean these days when they refer to
cognitive dissonance theory. Fundamentally in this section, we will treat the two versions as
one. When beneficial, however, we will differentiate between the two.

There are numerous recognized criteria by which any theory may be evaluated. One is a
theory’s usefulness or applicability in solving real-world problems. Here cognitive dissonance
theory gets high marks as the theory has been successfully brought to bear on numerous
real-world, important, and practical problems. These include fostering more successful child
rearing (e.g., Lepper, 1973; Mills, 1958), increasing condom use among sexually active college
students (Aronson et al., 1991; Stone, Aronson, Crain, Winslow, & Fried, 1994), promot-
ing pro-community behaviors such as water conservation (Dickerson, Thibodeau,
Aronson, & Miller, 1992) and waste-product recycling (Fried & Aronson, 1995), explaining
consumer behavior (Cialdini, 2009), increasing community service and charitable dona-
tions (e.g., Howard, 1990; Kraut, 1973; Sherman, 1980), reducing racial prejudice (e.g.,
Leippe & Eisenstadt, 1994; Son Hing, Li, & Zanna, 2002), and providing insight into the
treatment of phobias (Cooper, 1980) and the success of therapeutic interventions more
generally (e.g., Axsom & Cooper, 2004; Cooper & Axsom, 1982). In most of these applica-
tions, the basic idea is that since dissonance is unpleasant and aversive, less adaptive and
desirable behaviors can be decreased by associating them in people’s minds with cognitive
dissonance (i.e., times in the past where one has not lived up to one’s own standards [hypoc-
risy induction]), whereas more adaptive behaviors can be increased by associating them
with dissonance reduction. Some of this work is particularly impressive in that follow-up
measures have shown dissonance/hypocrisy interventions to still yield significant outcomes
even two and three months following the original induction of dissonance (Aronson et al.,
1991; Stone et al., 1994).

A quite different but potentially even more important application of dissonance theory
is more personal in nature and concerns what Aronson, Wilson, and Akert (1997) refer to
as the rationalization trap: “the potential for dissonance reduction to produce a succession
of self-justifications that ultimately result in a chain of stupid, immoral, [unethical, or ille-
gal] actions” (p. 222). A key problem with the experience of dissonance and the human
tendency to justify one’s behavior is that, over time, one’s self-regulatory boundaries can

Chadee_c02.indd 64Chadee_c02.indd 64 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

Inconsistency in Cognition: Cognitive Dissonance 65

expand. For example, few people start off stealing Post-it notes from their employer with
the thought they might eventually wind up embezzling thousands of dollars, getting caught,
and going to jail. Yet, once one begins stealing small things and rationalizing this stealing as
justifiable, it becomes easier and easier to graduate, over time, to higher and higher levels of
theft and impropriety (Lepper, 1973; Mills, 1958). Importantly, as documented by Tavris
and Aronson (2007), such escalation can occur with respect to any questionable behavior
whether the boundaries are personal, ethical, moral, or legal (e.g., lying, gambling, cheating
on one’s spouse or taxes, using illegal drugs or steroids). An anonymous quote is relevant to
this application of dissonance theory: “Be careful what you think because your thoughts
become your words, your words become your actions, your actions become your character,
and your character determines your destiny.” Translating this quote into the more direct
language of dissonance theory, it becomes more poignant and to the point: “You must be
careful what you do because your behavior can influence your attitudes, your attitudes can
influence your values, and your values will determine your destiny.”

The efficiency of a theory can be thought of as the ratio of phenomena explained by the
theory to the theory’s complexity. Here again, dissonance theory gets high marks because,
as documented in this chapter, there are literally scores of seemingly diverse findings that
are adequately, neatly, even compellingly, explained and integrated by the theory
(cf. Aronson, 1999, p. 108). Yet the essence of the theory remains relatively simple, capable
as it is of being summarized with a handful of sentences. Still, no one version of dissonance
theory can explain all of the data. What about the relative efficiency of Festinger’s (1957)
theory versus Aronson’s (1968, 1999)? In principle, Festinger’s theory is potentially more
comprehensive because Aronson’s self-inconsistency can be correctly regarded as a special
case of Festinger’s cognitive inconsistency. On the other hand, Aronson’s theory more easily
and elegantly accommodates individual differences than Festinger’s theory. If dissonance
is caused by inconsistency within one’s self-concept, it follows directly that dissonance will
be highly subjective in nature and that what causes dissonance in one person may not do
so in another (Aronson, 1999; e.g., Epstein, 1969; Nail et al., 2001, 2004; Snyder & Tanke,
1976; Szabo, 1985).

The heuristic value of a theory refers to its ability to generate testable, empirical hypoth-
eses. Here, dissonance theory is perhaps second to none. What is more, the Aronson (1968,
1999), Steele (1988), and other self-versions of the theory (e.g., Tesser, 1988) have added
tremendously to the heuristic value of Festinger (1957), which was already very high. It is
interesting to note, however, that Festinger did not think much of any of the revisions of his
theory. For example, he regarded the Brehm and Cohen (1962) emphasis that counter-
attitudinal behavior must be perceived as freely chosen as merely a special case of his more
general concept of minimal external pressure:

It says in the original book that in order for dissonance to be large enough to exist there has to

be minimal pressure on the person to do what the person does. If there is too much pressure,

there is too much justification for having done it, and if it is all consonant with having done it,

there is no dissonance. Doesn’t that encompass the idea of choice? Isn’t choice one part of that?

I said to myself, these wonderful people whom I like and respect are taking one operation and

elevating it to the status of a construct, and it’s terrible. (Festinger, 1987/1999, p. 383)

Chadee_c02.indd 65Chadee_c02.indd 65 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

66 Paul R. Nail and Kurt A. Boniecki

According to Aronson (1999), Festinger was also not pleased with Aronson’s (1968)
self-consistency revision. Festinger felt that it narrowed the theory too greatly. Further, once
more, can not the concept of self-inconsistency be encompassed and conceptualized as
merely a special case of the more general concept of cognitive inconsistency? Perhaps more
importantly, logic-like inconsistency that does not involve one’s self-concept is clearly
excluded by Aronson’s (1968, 1999) theory.

In this same vein, Festinger (1957) was clear that only important inconsistencies should
arouse measurable dissonance effects. From Festinger’s point of view, can it not be rea-
sonably argued that one of the surest ways to make an inconsistency important to some-
one is to arrange things so that the inconsistency is relevant to that person’s self-concept?
If so, as Greenwald and Ronis (1978) have argued, dissonance research may have been
inadvertently pulled in the direction of investigating inconsistencies with the self- concept
only because significant dissonance effects in this domain were relatively easy to obtain
and replicate in comparison to dissonance effects that might be obtained from the purely
logic-like inconsistencies hypothesized by Festinger (1957). Yet, it is not logically possible
to accept the hypothesis that important logic-like inconsistencies cannot motivate the
type of dissonance-reducing behavioral and attitudinal changes hypothesized by Festinger
only because they have not heretofore been demonstrated. Attempting to do so would
involve the same logical fallacy as accepting the null hypothesis. Thus, it is possible that
a true test of Festinger’s theory – one where a purely logic-like inconsistency is not
confounded with self-inconsistency – has yet to be performed. Ironically, there is never-
theless ample evidence in the literature that supports dissonance processes as originally
conceived by Festinger, although this evidence is indirect and hidden to most observers.
We refer to the behavior of the many theorists who have proposed revisions of dissonance
theory to better square with the evidence the theory has generated, thus attempting to
reduce a purely logic-like inconsistency. “The behavior of the theorists doing the revising
is a near-perfect illustration of dissonance reduction of the sort intended in the original
statement of the theory” (Greenwald & Ronis, 1978, p. 55) but excluded by the Aronson
(1968, 1999) and Steele (1988) ego-based versions, as well as all other versions (e.g.,
Cooper & Fazio, 1984).

With final respect to evaluation, the originality of a theory refers to the novelty of its
constructs and hypotheses. Here, dissonance theory only gets mixed marks. If dissonance
processes are grounded in the ego or self-system as Aronson (1968, 1999), Steele (1988),
and others maintain (e.g., Higgins, 1987; Wicklund & Gollwitzer, 1982), then there is a
considerable body of theoretical work that highlighted inconsistencies within, or threats to,
one’s self-concept or ego system that antedated Festinger, Aronson, and Steele by several
years, even decades (e.g., Allport, 1943; Freud, 1924/1968; Hilgard, 1949; Rogers, 1951).
Rogers, for example, theorized how existing inconsistencies between a person’s actual and
ideal self-images can cause internal psychological tensions that are of critical importance
with respect to one’s personal growth, development, and mental health. On the other hand,
if dissonance refers only to the need for logic-like consistency, as Festinger (1957) proposed,
at least two social-psychological cognitive consistency theories were published before
Festinger (1957): the A–B–X model (Newcomb, 1953; Newcomb & Svehla, 1937) and con-
gruity theory (Osgood & Tannenbaum, 1955). Still, there is no question that Festinger’s

Chadee_c02.indd 66Chadee_c02.indd 66 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

Inconsistency in Cognition: Cognitive Dissonance 67

theory is far superior at generating testable hypotheses than any of the early self-theories or
any of the alternative cognitive consistency theories (e.g., Heider, 1958).

With further respect to the lack of originality of Festinger (1957), students of social
psychology who learn of Piaget’s (1937/1971) theory of cognitive development cannot
help but be struck by the parallels between the two systems. Whereas Festinger (1957)
hypothesized how inconsistencies among cognitive elements cause cognitive dissonance,
which motivates a person to make changes to restore consonance, Piaget (1937/1971)
hypothesized how inconsistencies among mental structures (viz., schemas) cause psy-
chological disequilibrium, which motivates a person to make changes (viz., accommoda-
tions) to restore equilibrium. These parallels suggest one potential place to look for
direct evidence that might support dissonance theory as originally conceived by
Festinger – the Piagetian tasks, which Piaget (1929/1979, 1941/1997) developed to test
various aspects of his theory (e.g., the object permanence, conservation of number and
volume, and displacement of liquid tasks). Our observations of preschool children,
clearly experiencing and overcoming cognitive dissonance en route to correct solutions
on these tasks, suggest that the solutions might occur either too early in a child’s devel-
opment, or too quickly and spontaneously, to involve the self-concept. Careful experi-
mentation might settle the issue. Evidence that success on these tasks is not mediated or
moderated by the self, we argue, would support Festinger’s (1957) original theory over
any of the competing versions. Importantly, such evidence could potentially reestablish
Festinger’s (1957) theory as the dominant, most generally adequate account of disso-
nance phenomena.

Adams, J. S. (1965). Inequity in social exchange. In

L. Berkowitz (Ed.), Advances in experimental social psy-

chology (Vol. 2, pp. 267–299). New York: Academic

Press.

Allport, G. W. (1943). The ego in contemporary psychology.

Psychological Review, 50, 451–478.

Aronson, E. (1960). The cognitive and behavioral consequences

of the confirmation and disconfirmation of expectancies.

Grant proposal, Harvard University.

Aronson, E. (1968). Dissonance theory: Progress and

problems. In R. P. Abelson, E. Aronson, W. J. McGuire,

T. M. Newcomb, M. J. Rosenberg, & P. H. Tannenbaum

(Eds.), Theories of cognitive consistency: A source book

(pp. 5–27). Skokie, IL: Rand McNally.

Aronson, E. (1999). Dissonance, hypocrisy, and the self-

concept. In E. Harmon-Jones & J. Mills (Eds.) Cognitive

dissonance: Progress on a pivotal theory in social psy-

chology (pp. 103–126). Washington, DC: American

Psychological Association.

Aronson, E. (2007). The evolution of cognitive dissonance

theory. In A. R. Pratkanis (Ed.), The science of social influ-

ence: Advances and future progress (pp. 115–135).

New York: Psychology Press.

Aronson, E., & Carlsmith, J. M. (1963). Effect of severity of

threat on the valuation of forbidden behavior. Journal of

Abnormal and Social Psychology, 66, 178–182.

Aronson, E., Fried, C. & Stone, J. (1991). Overcoming denial

and increasing the intention to use condoms through

References

1 These means were transformed so that higher numbers indicate greater attitude change and lower

numbers lesser attitude change.

Note

Chadee_c02.indd 67Chadee_c02.indd 67 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

68 Paul R. Nail and Kurt A. Boniecki

the induction of hypocrisy. American Journal of Public

Health, 81, 1636–1638.

Aronson, E., & Mettee, D. (1968). Dishonest behavior as a

function of different levels of self-esteem. Journal of

Personality and Social Psychology, 9, 121–127.

Aronson, E., & Mills, J. (1959). The effect of severity of ini-

tiation on liking for a group. Journal of Abnormal and

Social Psychology, 59, 177–181.

Aronson, E., Wilson, T. D., & Akert, R. M. (1997). Social

psychology (2nd ed.). New York: Longman.

Aronson, J., Cohen, G., & Nail, P. R. (1999). Self-affirmation

theory: An update and appraisal. In E. Harmon-Jones &

J. Mills (Eds.), Cognitive dissonance: Progress on a pivotal

theory in social psychology (pp. 127–147). Washington,

DC: American Psychological Association.

Axsom, D., & Cooper, J. (2004). Cognitive dissonance and

psychotherapy: The role of effort justification in inducing

weight loss. In R. M. Kowalski & M. R. Leary (Eds.), The

interface of social and clinical psychology: Key readings

(pp. 313–323). New York: Psychology Press.

Bator, R. J., & Cialdini, R. B. (2006). The nature of con-

sistency motivation: Consistency, aconsistency, and

anticonsistency in a dissonance paradigm. Social

Influence, 1, 208–233.

Batson, C. D. (1975). Rational processing or rationalization?:

The effect of disconfirming information on a stated reli-

gious belief. Journal of Personality and Social Psychology,

32, 176–184.

Beauvois, J. L., & Joule, R. V. (1996). A radical dissonance

theory. London: Taylor & Francis.

Bem, D. J. (1967). Self-perception: An alternative interpreta-

tion of cognitive dissonance phenomena. Psychological

Review, 74, 183–200.

Bem, D. J. (1968). Dissonance reduction in the behaviorist.

In R. P. Abelson, E. Aronson, W. J. McGuire, T. M.

Newcomb, M. J. Rosenberg, & P. H. Tannenbaum (Eds.),

Theories of cognitive consistency: A source book (pp. 5–27).

Skokie, IL: Rand McNally.

Bem, D. J., & McConnell, H. K. (1970). Testing the self-

perception explanation of dissonance phenomena: On

the salience of premanipulation attitudes. Journal of

Personality and Social Psychology, 14, 23–31.

Bendersky, C., & Curhan, J. R. (2009). Cognitive dissonance

in negotiation: Free choice or justification? Social

Cognition, 27, 455–474.

Bishop, F. V. (1967). The anal character: A rebel in the dis-

sonance family. Journal of Personality and Social

Psychology, 6, 23–36.

Blass, T. (2004). The man who shocked the world: The life

and legacy of Stanley Milgram. New York: Basic Books.

Brehm, J. (1956). Postdecision changes in the desirability of

alternatives. Journal of Abnormal and Social Psychology,

52, 384–389.

Brehm, J. W., & Cohen, A. R. (1962). Explorations in cogni-

tive dissonance. New York: John Wiley.

Bugliosi, V. (2007). Reclaiming history: The assassination of

President John F. Kennedy. New York: W. W. Norton.

Bushman, B. J. (2002). Does venting anger feed or extin-

guish the flame? Catharsis, rumination, distraction,

anger, and aggressive responding. Personality and Social

Psychology Bulletin, 28, 724–731.

Chapanis, N. P., & Chapanis, A. (1964). Cognitive disso-

nance: Five years later. Psychological Bulletin, 61, 1–22.

Cialdini, R. B. (2009). Influence: Science and practice (5th

ed.). Boston, MA: Pearson.

Cialdini, R. B., Trost, M. R., & Newsom, J. T. (1995).

Preference for consistency: The development of a valid

measure and the discovery of surprising behavior impli-

cations. Journal of Personality and Social Psychology, 69,

318–328.

Cooper, J. (1980). Reducing fears and increasing assertive-

ness: The role of dissonance reduction. Journal of

Experimental Social Psychology, 16, 199–213.

Cooper, J. (2007). Cognitive dissonance: Fifty years of a clas-

sic theory. Thousand Oaks, CA: Sage.

Cooper, J., & Axsom D. (1982). Effort justification in psy-

chotherapy. In G. Weary & H. Mirels (Eds.), Integrations

of clinical and social psychology (pp. 214–230). New York:

Oxford University Press.

Cooper, J., & Fazio, R. H. (1984). A new look at dissonance

theory. In L. Berkowitz (Ed.). Advances in experimental

social psychology (Vol. 17, pp. 229–266). New York:

Academic Press.

Croyle, R., & Cooper, J. (1983). Dissonance arousal:

Physiological evidence. Journal of Personality and Social

Psychology, 45, 782–791.

Davis, K., & Jones, E. E. (1960). Changes in interpersonal

perception as a means of reducing cognitive dissonance.

Journal of Abnormal and Social Psychology, 61,

402–410.

Dean, J. W. (1976). Blind ambition: The White House years.

New York: Simon & Schuster.

Dickerson, C., Thibodeau, R., Aronson, E., & Miller, D.

(1992). Using cognitive dissonance to encourage water

conservation. Journal of Applied Social Psychology, 22,

841–854.

Chadee_c02.indd 68Chadee_c02.indd 68 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

Inconsistency in Cognition: Cognitive Dissonance 69

Dunn, J. A., Auden, J. B., Gosh, A. M. N., & Roy, S. C. (1981).

The Bihar–Nepal earthquake of 1934. Memoir Geological

Society of India, Survey of India, Calcutta, 73, 391. (Original

work published 1939)

Elkin, R. A., & Leippe, M. R. (1986). Physiological arousal, dis-

sonance, and attitude change: Evidence for a dissonance–

arousal link and a “don’t remind me” effect. Journal of

Personality and Social Psychology, 51, 55–65.

Epstein, G. F. (1969). Machiavelli and the devil’s advocate.

Journal of Personality and Social Psychology, 11, 38–41.

Fazio, R. H., Zanna, M. P., & Cooper, J. (1977). Dissonance

and self-perception: An integrative view of each theory’s

proper domain of application. Journal of Experimental

Social Psychology, 13, 464–479.

Festinger, L. (1957). A theory of cognitive dissonance.

Stanford, CA: Stanford University Press.

Festinger, L. (1987/1999). Reflections on cognitive disso-

nance: 30 years later. In E. Harmon-Jones & J. Mills

(Eds.), Cognitive dissonance: Progress on a pivotal theory

in social psychology (pp. 381–385). Washington, DC:

American Psychological Association.

Festinger, L., & Carlsmith, J. M. (1959). Cognitive conse-

quences of forced compliance. Journal of Abnormal and

Social Psychology, 58, 203–210.

Festinger, L., Riecken, H. W., & Schachter, S. (1956). When

prophecy fails. New York: Harper & Row.

Freud, S. (1924/1968). A general introduction to psychoa-

nalysis (J. Riviere, Trans.). New York: Washington Square

Press.

Fried, C. B., & Aronson, E. (1995). Hypocrisy, misat-

tribution, and dissonance reduction: A demon-

stration of dissonance in the absence of aversive

conse quences. Personality and Social Psychology Bulletin,

21, 925–933.

Gerard, H. B., Blevans, S. A., & Malcolm, T. (1964). Self-

evaluation and the evaluation of choice alternatives.

Journal of Personality, 32, 395–410.

Gerard, H. B., & Mathewson, G. C. (1966). The effects of

severity of initiation on liking for a group: A replication.

Journal of Experimental Social Psychology, 2, 278–287.

Glass, D. (1964). Changes in liking as a means of reducing

cognitive discrepancies between self-esteem and aggres-

sion. Journal of Personality, 32, 531–549.

Greenwald, A. G., & Ronis, D. L. (1978). Twenty years of

cognitive dissonance: Case study of the evolution of a

theory. Psychological Review, 85, 53–57.

Gleick, E. (1997, February 10). Marine blood sports. Time,

p. 30.

Guadagno, R. E., Asher, T., Demaine, L. J., & Cialdini, R. B.

(2001). When saying yes leads to saying no. Preference

for consistency and the reverse foot-in-the-door effect.

Personality and Social Psychology Bulletin, 27, 859–867.

Harmon-Jones, E., & Mills, J. (Eds.). (1999). Cognitive

dissonance: Progress on a pivotal theory in social psy-

chology. Washington, DC: American Psychological

Association.

Hart, W., Albarracin, D., Eagly, A. H., Brechan, I., Lindberg,

M. J., & Merrill, L. (2009). Feeling validated versus being

correct: A meta-analysis of selective exposure to infor-

mation. Psychological Bulletin, 135, 555–588.

Heider, F. (1958). The psychology of interpersonal rela-

tions. New York: Wiley.

Higgins, E. T. (1987). Self-discrepancy: A theory relating

self and affect. Psychological Review, 94, 319–340.

Hilgard, E. R. (1949). Human motives and the concept of

self. American Psychologist, 4, 374–382.

Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953).

Communication and persuasion: Psychological studies of

opinion change. New Haven, CT: Yale University Press.

Howard, D. J. (1990). The influence of verbal responses to

common greetings on compliance behavior: The foot-

in-the-mouth effect. Journal of Applied Social Psychology,

20, 1185–1196.

Jones, E. E. (1985). Major accomplishments in social psy-

chology during the past five decades. In G. Lindzey &

E. Aronson (Eds.), The handbook of social psychology (3rd

ed., Vol. 1, pp. 47–108). New York: Random House.

Kahn, M. (1966). The physiology of catharsis. Journal of

Personality and Social Psychology, 3, 278–298.

Klugel, J. R., & Smith, E. R. (1986). Beliefs about inequality:

Americans’ views of what is and what ought to be.

New York: Aldine De Gruyter.

Kraut, R. E. (1973). Effects of social labeling on giving to

charity. Journal of Experimental Social Psychology, 9,

551–562.

Leippe, M. R., & Eisenstadt, D. (1994). The generalization

of dissonance reduction: Decreasing prejudice through

induced compliance. Journal of Personality and Social

Psychology, 67, 395–413.

Lepper, M. R. (1973). Dissonance, self-perception, and

honesty in children. Journal of Personality and Social

Psychology, 25, 65–74.

Lewin, K. (1943). Defining the “field at a given time.”

Psychological Review, 50, 292–310.

Manchester, W. (1967). The death of a president. New York:

Harper & Row.

Chadee_c02.indd 69Chadee_c02.indd 69 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

70 Paul R. Nail and Kurt A. Boniecki

McGregor, I., Nail, P. R., Marigold, D. C., & Kang, S.-J.

(2005). Defensive pride and consensus: Strength in

imaginary numbers after self-threat. Journal of Personality

and Social Psychology, 89, 978–996.

McGregor, I., Zanna, M. P., Holmes, J. G., & Spencer, S. J.

(2001). Compensatory conviction in the face of personal

uncertainty: Going to extremes and being one self.

Journal of Personality and Social Psychology, 80, 472–488.

Mills, J. (1958). Changes in moral attitudes following temp-

tation. Journal of Personality, 26, 517–531.

Nail, P. R. (2006, April). Explorations in cognitive dissonance:

Whatever that means? Presidential Address, Southwestern

Psychological Association, Austin, TX.

Nail, P. R., Bedell, K. E., & Little, C. D. (2003). Should

President Clinton be prosecuted for perjury?: The effects

of preference for consistency, self-esteem, and political

party affiliation. Personality and Individual Differences,

35, 1821–1831.

Nail, P. R., Correll, J. S., Drake, C. E., Glenn, S. B., Scott, G.

M., & Stuckey, C. (2001). A validation study of the pref-

erence for consistency scale. Personality and Individual

Differences, 31, 1193–1202.

Nail, P. R., Harton, H. C., & Decker, B. P. (2003). Political

orientation and modern versus aversive racism: Tests of

Dovidio and Gaertner’s integrated model. Journal of

Personality and Social Psychology, 84, 754–770.

Nail, P. R., & MacDonald, G. (2007). On the develop-

ment of the social response context model: New

directions for social influence theory and research. In

A. R. Pratkanis (Ed.), The science of social influence:

Advances and future progress (pp. 193–221). New York:

Psychology Press.

Nail, P. R., MacDonald, G., & Levy, D. A. (2000). Proposal of

a four-dimensional model of social response. Psychological

Bulletin, 126, 454–470.

Nail, P. R., McGregor, I., Drinkwater, A., Steele, G., &

Thompson, A. W. (2009). Threat causes liberals to think

like conservatives. Journal of Experimental Social

Psychology, 45, 901–907.

Nail, P. R., Misak, J. E., & Davis, R. M. (2004). Self-

affirmation versus self-consistency: A comparison of

two competing self-theories of dissonance phenomena.

Personality and Individual Differences, 36, 1893–1905.

Nel, E., Helmreich, R., & Aronson, E. (1969). Opinion

change in the advocate as a function of the persuasibility

of his audience: A clarification of the meaning of disso-

nance. Journal of Personality and Social Psychology, 12,

117–124.

Newcomb, T. M. (1953). An approach to the study of com-

municative acts. Psychological Review, 60, 393–404.

Newcomb, T. M., & Svehla, G. (1937). Intra-family rela-

tionships in attitudes. Sociometry, 1, 180–205.

Osgood, C. E., & Tannenbaum, P. H. (1955). The principle

of congruity in the prediction of attitude change.

Psychological Review, 62, 42–55.

Pallak, M., & Pittman, T. (1972). General motivation effects

of dissonance arousal. Journal of Personality and Social

Psychology, 21, 349–358.

Piaget, J. (1929/1979). The child’s conception of the world

(J. Tomlinson & A. Tomlinson, Trans.). New York:

International Universities Press.

Piaget, J. (1937/1971). The construction of reality in the child

(M. Cook, Trans.). New York: Ballantine Books.

Piaget, J. (1941/1997). The child’s conception of number

(C. Gattegno & F. M. Hodgson, Trans.). London & New

York: Routledge.

Prasad, J. (1950). A comparative study of rumours and

reports in earthquakes. British Journal of Psychology, 41,

129–144.

Rogers, C. R. (1951). Client-centered therapy: Its current prac-

tice, implications, and theory. Boston: Houghton Mifflin.

Rosenberg, M. (1965a). Society and the adolescent self-

image. Princeton, NJ: Princeton University Press.

Rosenberg, M. J. (1965b). When dissonance fails: On elimi-

nating evaluation apprehension from attitude measure-

ment. Journal of Personality and Social Psychology, 7,

28–42.

Scher, S. J., & Cooper, J. (1989). Motivational basis of dis-

sonance: The singular role of behavioral consequences.

Journal of Personality and Social Psychology, 56,

899–906.

Sherman, S. J. (1980). On the self-erasing nature of errors

of prediction. Journal of Personality and Social Psychology,

39, 211–221.

Simon, L., Greenberg, J., & Brehm, J. (1995). Trivialization:

The forgotten mode of dissonance reduction. Journal of

Personality and Social Psychology, 68, 247–260.

Snyder, M., & Tanke, E. D. (1976). Some people are more

consistent than others. Journal of Personality, 44,

501–517.

Son Hing, L. S., Li, W., & Zanna, M. P. (2002). Inducing

hypocrisy to reduce prejudicial responses among aver-

sive racists. Journal of Experimental Social Psychology, 38,

71–78.

Stafford, E. (Producer). (2009, November 3). Inside a cult.

Viewed from National Geographic Channel website:

Chadee_c02.indd 70Chadee_c02.indd 70 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

Inconsistency in Cognition: Cognitive Dissonance 71

http://channel.nationalgeographic.com/channel/videos/

player.html?channel=60850.

Steele, C. M. (1988). The psychology of self-affirmation:

Sustaining the integrity of the self. In L. Berkowitz (Ed.),

Advances in experimental social psychology (Vol. 21,

pp. 261–302). San Diego, CA: Academic Press.

Steele, C. M., & Liu, T. J. (1983). Dissonance processes as

self-affirmation. Journal of Personality and Social

Psychology, 45, 5–19.

Steele, C. M., Spencer, S. J., & Lynch, M. (1993). Self-image

resilience and dissonance: The role of affirmational

resources. Journal of Personality and Social Psychology,

64, 885–896.

Stone, J. (1999). What exactly have I done? The role of self-

attribute accessibility in dissonance. In E. Harmon-Jones

& J. Mills (Eds.), Cognitive dissonance: Progress on a piv-

otal theory in social psychology (pp. 175–200). Washington,

DC: American Psychological Association.

Stone, J., Aronson, E. Crain, A. L., Winslow, M. P., &

Fried, C. B. (1994). Inducing hypocrisy as a means of

encouraging young adults to use condoms. Personality

and Social Psychology Bulletin, 20, 116–128.

Stone, J., & Cooper, J. (2001). A self-standards model of

cognitive dissonance. Journal of Experimental Social

Psychology, 37, 228–243.

Styron, W. (1977). A farewell to arms. New York Review of

Books, 24, 3–5.

Swann, W. B., Jr. (1984). Quest for accuracy in person per-

ception: A matter of pragmatics. Psychological Review,

91, 457–477.

Swann, W. B., Jr. (1996). Self-traps: The elusive quest for

higher self-esteem. New York: Freeman.

Swann, W. B., Jr., & Read, S. J. (1981). Acquiring self-

knowledge: The search for feedback that fits. Journal

of Personality and Social Psychology, 41, 1119–1128.

Szabo, C. (1985). Effect of role-play on attitude change.

Magyar Pszichologiai Szemle, 42, 495–507.

Tavris, C., & Aronson, E. (2007). Mistakes were made (but

not by me): Why we justify foolish beliefs, bad decisions,

and hurtful acts. Orlando, FL: Harcourt.

Tedeschi, J. T., Schlenker, B. R., & Bonoma, T. V. (1971).

Cognitive dissonance: Private ratiocination or public

spectacle? American Psychologist, 26, 685–695.

Tesser, A. (1988). Toward a self-evaluation maintenance

model of social behavior. In L. Berkowitz (Ed.), Advances

in experimental social psychology (Vol. 21, pp. 181–227).

San Diego, CA: Academic Press.

Tesser, A., & Cornell, D. P. (1991). On the confluence of self

processes. Journal of Experimental Social Psychology, 27,

501–526.

Thibaut, J. W., & Kelley, H. H. (1959). The social psychology

of groups. New York: John Wiley.

Thibodeau, R., & Aronson, E. (1992). Taking a closer look:

Reasserting the role of the self-concept in dissonance

theory. Personality and Social Psychology Bulletin, 18,

591–602.

Wicklund, R. A., & Gollwitzer, P. M. (1982). Symbolic self-

completion. Hillsdale, NJ: Erlbaum.

Zajonc, R. B. (1965). Social facilitation. Science, 149,

269–274.

Chadee_c02.indd 71Chadee_c02.indd 71 12/13/2010 2:17:38 PM12/13/2010 2:17:38 PM

In social psychology, the term attribution has two primary meanings. The first refers to
explanations of behavior (i.e., answers to why questions); the second refers to inferences or
ascriptions (e.g., inferring traits from behavior, ascribing blame to a person). What the two
meanings have in common is a process of assigning: in attribution as explanation, a behav-
ior is assigned to its cause; in attribution as inference, a quality or attribute is assigned to the
agent on the basis of an observed behavior. Despite the connection between these phenom-
ena, they have distinct psychological characteristics (Hamilton, 1998; Hilton, Smith, & Kin,
1995; Malle, in press). This chapter will focus on attribution as behavior explanation because
it is a far-reaching cognitive and social phenomenon that is embedded in the larger human
search for meaning (Malle, 2004).1

The discussion will begin with the undisputed founder of attribution work, Fritz Heider,
then briefly visit Jones and Davis’s contribution, and move on to Harold Kelley’s theoretical
model. Because many excellent reviews of the standard views on these theories are available
(see note 1), I will spend relatively little time recounting them. My goal is rather to point out
aspects of classic attribution theories that are not generally emphasized, highlight historical
misunderstandings, and bring to light theoretical difficulties that have not been adequately
addressed. In the second half of the chapter I then introduce an alternative theory of behavior
explanations that builds on previous theories but tries to overcome their major shortcomings.

Heider’s Theory of Attribution

Fritz Heider developed models of attribution for both object perception and person per-
ception. His theory of object perception (first described in Heider, 1920, his dissertation) is
rarely cited today, but it serves as the foundation for his later theory of person perception.

3

Attribution Theories: How People
Make Sense of Behavior

Bertram F. Malle

Theories in Social Psychology, First Edition. Edited by Derek Chadee.
© 2011 Blackwell Publishing Ltd. Published 2011 by Blackwell Publishing Ltd.

Chadee_c03.indd 72Chadee_c03.indd 72 12/13/2010 2:17:56 PM12/13/2010 2:17:56 PM

Attribution Theories: How People Make Sense of Behavior 73

Heider attempted to solve one of the core philosophical problems of phenomenology:
the relation between sensory information and real objects. That is, he asked how it was pos-
sible that humans perceive qualities of objects in the world even though all they have are
sensations in the mind. Heider argued that real objects shape “media” such as air pressure,
light reflections, and sense organs. These media have a considerable degree of variance (for
one thing, they reflect many real objects), but the perceptual apparatus reconstructs real
objects from their characteristic effects on the media. Heider labeled this reconstruction
attribution – a process that generates inferences of the relatively invariant qualities of objects
from the characteristic variance patterns they cause in their media. Perceivers faced with
sensory information thus experience perceptual objects as “out there” because they attribute
the sensory data to their underlying cause in the world (Heider, 1920).2

After his early work on object perception Heider turned to the domain of social interac-
tions, wondering how people perceive each other in interaction and especially how they
make sense of each others’ behavior. Heider proposed that a process of attribution is
involved in person perception as well, but he recognized that person perception is more
complex than object perception – due to the manifold observational data available and the
various causes (e.g., beliefs, desires, emotions, traits) to which these data can be attributed.
In addition, it was clear to Heider that persons are very different targets of perception than
inanimate objects. Persons are “perceived as action centers and as such can do something to
us. They can benefit or harm us intentionally, and we can benefit or harm them. Persons
have abilities, wishes and sentiments; they can act purposefully, and can perceive or watch
us” (Heider, 1958, p. 21). Note that Heider repeatedly refers to the intentionality of persons,
which he considered a core assumption in the conceptual framework that underlies social
perception. With the help of such concepts as intentionality and the inference of wishes,
purposes, sentiments, and other mental states, Heider argued, perceivers bring order and
meaning to the massive stream of behavioral data.

Even though in one sense person perception is like object perception – a process of
extracting invariance out of variance – Heider emphasized two distinct features of person
perception. The first is that in the social domain, variance refers to the agent’s stream of
ongoing behavior and invariance refers to the inferred perceptions, intentions, motives,
traits, and sentiments. Heider (1958) sometimes used the term disposition to refer to these
invariances, and even though he occasionally referred to traits and abilities when talking
about dispositions (e.g., pp. 30, 80), he considered “motives, intentions, sentiments … the
core processes which manifest themselves in overt behavior” (p. 34). It was the agent’s
motives that occupied a special role in Heider’s model: “The underlying causes of events,
especially the motives of other persons, are the invariances of the environment that are rele-
vant to [the perceiver]; they give meaning to what he experiences” (Heider, 1958, p. 81;
emphasis added). In social perception, then, Heider’s terms disposition and invariance
referred primarily to mental states.

The second distinct feature of person perception is that when people perform a causal
(i.e., attributional) analysis of human behavior, their judgments of causality follow one of
two conceptual models (Heider, 1958, chap. 4). The first is a model of impersonal causality,
applied to unintentional human behaviors (such as sneezing or feeling sad) and physical
events (such as waves splashing or leaves falling). The second is a model of personal causality,

Chadee_c03.indd 73Chadee_c03.indd 73 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

74 Bertram F. Malle

which is invoked whenever a human agent performs an intentional action (such as clean-
ing the kitchen or inviting someone to dinner). “Personal causality,” Heider wrote, “refers
to instances in which p causes x intentionally. That is to say, the action is purposive”
(Heider, 1958, p. 100).

Unfortunately, subsequent attribution research misrepresented both of these crucial
features of person perception. First, even though mental states make up the majority of
what Heider subsumed under dispositional properties (Heider, 1958, pp. 31–34, chap. 4,
passim), most scholars portrayed Heider’s notion of disposition as referring to stable
personality factors (i.e., traits, attitudes, or abilities). For example: “Heider began by
assuming that just as objects have enduring qualities that determine their appearances, so
people have stable psychological characteristics that determine their behavior” (Gilbert,
1998, p. 94; emphasis added). This interpretation of dispositions as necessarily stable can
be traced to two early sources. The first is Kelley’s (1960) review of Heider’s (1958) book,
according to which “Heider’s central theme is that perception leaps over the raw data
presented and enables the person to understand the stable, dispositional properties …
that account for them” (p. 2). The second is Jones and Davis’s (1965) influential paper
“From acts to dispositions,” in which they used the term disposition to refer to character
traits and attitudes only. Hence research on dispositional attribution became research on
trait inferences.

The second misunderstanding was that many scholars mistook Heider’s distinction
between personal and impersonal causality – the terms he used to characterize intentional
versus unintentional behavior (Heider, 1958, pp. 100–101) – for a distinction between per-
son causes and situation causes. Kelley (1967) famously posited that in all explanations,
“the choice is between external attribution and internal … attribution” (p. 194). Attribution
research applied this person–situation dichotomy to all behaviors alike, whether intentional
or unintentional, and thereby eliminated Heider’s central concepts of intention, purpose,
and motive from later models of social perception.

Ironically, Heider was credited for – and indeed identified with – “discovering” the simple
person–situation dichotomy himself:

How do we search for the causal structure of interpersonal events? According to Heider, we do

so by reliance upon attributions to the environment (external factors) or to something about

the other person (internal factors). (Weary, Edwards, & Riley, 1994, p. 292)

Central to Heider’s entire theoretical position is the proposition that man perceives behavior as

being caused, and that the causal locus can be either in the perceiver or in the environment.

(Hastorf, Schneider, & Polefka, 1970, p. 63)

But why the perception that Heider proposed a person–situation dichotomy in attribu-
tion? One small section of his book appears to have spawned this claim (Heider, 1958, pp.
82–84). There Heider characterizes any “action outcome” (the result of an action, not the
action itself ) as “dependent upon a combination of effective personal force and effective
environmental force” (p. 82). In elaborations of this claim (pp. 83–87), Heider offered
a complex picture. He argued that for an action outcome to occur, there needs to be a
concomitance of two elements: the agent’s attempt to perform the action (trying) and

Chadee_c03.indd 74Chadee_c03.indd 74 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

Attribution Theories: How People Make Sense of Behavior 75

supporting factors (can) that lie in the agent (effort, ability) or in the environment
(e.g., opportunity, luck, favorable conditions). Trying is the execution of an intention,
so Heider stayed true to his analysis of action in terms of intentionality (personal causal-
ity). Only for the can forces did Heider apply the distinction between person factors and
environmental factors, and these can forces play a very circumscribed role: they are the
necessary elements for an intentional action to be successful, the elements that enable the
desired outcome to occur (Heider, 1958, p. 109). Such enabling factor explanations (Malle,
1999) answer the specific explanatory question of how it was possible that the action
outcome was attained (Malle, Knobe, O’Laughlin, Pearce, & Nelson, 2000; McClure &
Hilton, 1997). Only for this explanation mode did Heider introduce the distinction
between internal and external causal factors. There is no indication in the text that Heider
thought people use the internal–external distinction when explaining behavior in gen-
eral. On the contrary, Heider stated that people explain why a person is trying to do
something by referring to the “reasons behind the intention” (Heider, 1958, p. 110; see
also pp. 125–129).

The contrast between these two types of explanations – why the action was chosen and
what enabled it to succeed – can be illustrated with the following passage from Gilbert
(1998, p. 96):

If a pitcher who wishes to retire a batter (motivation) throws a burning fastball (action) directly

into the wind (environmental influence), then the observer should conclude that the pitcher

has a particularly strong arm (ability). If a batter tries to hit that ball (motivation) but fails

(action), then the observer should conclude that the batter lacked coordination (ability) or was

blinded by the sun (environmental influence).

The observer’s reasoning here is entirely focused on accounting for successful or failed out-
comes; the question of why the batter and the pitcher acted as they did is not answered by
reference to either arm, wind, or sun. This why question is already answered by mentioning
the pitcher’s obvious desire to retire the batter and the batter’s obvious desire to hit the ball
and get a run. Talking about the arm, wind, and sun, by contrast, answers the question of
how the outcome was attained (Malle et al., 2000).

In an interview with Bill Ickes (1976, p. 14), Heider explicitly distinguished between these
two questions and hence between two types of explanation:

1. the attribution of outcomes to causal factors (i.e., enabling factor explanations);
2. the attribution of intentional actions to the actor’s motives (i.e., reasons for acting).

Heider himself never developed a model of motive attributions or reason explanations,
and he felt that these explanations had not been adequately treated by contemporary attri-
bution work (Ickes, 1976, p. 14). By contrast, Heider felt that outcome attributions
(e.g., what made a student fail or succeed on a test) were well developed in Weiner’s work
(e.g., Weiner, 1986; Weiner et al., 1972).

The misperception that Heider proposed the external–internal dichotomy as the funda-
mental dimension of explanation may thus stem from a confounding of outcomes

Chadee_c03.indd 75Chadee_c03.indd 75 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

76 Bertram F. Malle

(explained by enabling factors) and actions (explained by motives or reasons). The follow-
ing passages from Hastorf et al. (1970) illustrate this confusion:

Presumably the outcomes of action are caused by some combination of personal characteris-

tics and environmental forces [outcome attribution]. The person may have done something

because he had to do it … or because he wanted to do it [action explanation]. (p. 64)

When we infer that the combination of ability and effort was stronger than the external forces,

we infer that internal causality was present [outcome attribution]. Only then do we say such

things as “he did it because he wanted to” [action explanation]. (p. 89)

In both of these passages, the authors treat two different explanatory questions as if they
were one and the same. The judgment whether “he did it because he wanted to” or “because
he had to do it” clarifies the agent’s reasons for an action. These reasons can be given even before
the agent tries to perform the action because reasons explain the intention, whether or not
the intention gets fulfilled. By contrast, the judgments about ability, effort, or external forces
clarify how it was possible that the action outcome was attained (an enabling factor explana-
tion). This explanation can be given only after the agent tried to perform the action – if he
succeeded – whereas reason explanations track the motivation for the action in the first place.

Thus, whereas Heider has been consistently credited with introducing the person–situation
dichotomy in attribution theory, Heider’s actual theory was predicated on the distinction
between personal causality (which accounts for intentional events) and impersonal causality
(which accounts for unintentional events) – later recognized as a core constituent of social
cognition (Malle, Moses, & Baldwin, 2001; Zelazo, Astington, & Olson, 1999). Heider’s analy-
sis of personal causality revealed a complex set of mental states centered on intention; and
people, when trying to make sense of intentional behavior, attempt to infer those states (such
as beliefs, wishes, and sentiments), which were the reasons that guided the actor’s behavior.

Jones and Davis’s Abandoned Theory of Explanation

Jones and Davis (1965) were the first to introduce a theory of dispositional inference, speci-
fying conditions under which a perceiver infers a stable disposition (personality trait or
attitude) from an agent’s behavior. But before they introduced this theory, the first few
pages of their famous chapter appeared to go in a different direction. They targeted just the
issue that Heider had left open: exactly how people explain intentional action by means of
motives or reasons. In a short section, entitled “The Naive Explanation of Human Action:
Explanation by Attributing Intentions,” the authors attempted to account for “a perceiver’s
inferences about what an actor was trying to achieve by a particular action” (p. 222) and the
process of finding “sufficient reason why the person acted” (p. 220), because “the perceiver’s
explanation comes to a stop when an intention or motive is assigned that has the quality of
being reason enough” (p. 220). Despite an apparent plan to present a theory of action expla-
nation by reasons, this was the last that Jones and Davis wrote about action explanations.3
They immediately turned to the conditions under which perceivers infer traits (e.g., arro-
gance or dominance; see p. 223) from single behavioral events. This theory was refined
in several models of dispositional attribution (e.g., Gilbert, Pelham, & Krull, 1988;
Quattrone, 1982), and it inspired research on stereotypes (e.g., Gilbert & Hixon, 1991;

Chadee_c03.indd 76Chadee_c03.indd 76 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

Attribution Theories: How People Make Sense of Behavior 77

Yzerbyt, Rogier, & Fiske, 1998), judgment biases (e.g., Ross, 1977), and impression forma-
tion (e.g., Reeder, 1997; Trope, 1986). However, Jones and Davis (1965) wholly left behind
the social perceiver’s attempt to find the agent’s reasons for acting,4 and it was up to the next
attribution theorists to tackle this question.

Kelley’s Theory of Attribution as Causal Judgment

Kelley’s (1967) paper on attribution theory in social psychology is generally considered the
first systematic and general treatment of lay causal explanations. Kelley’s self-ascribed goal
in the paper was “to highlight some of the central ideas contained in Heider’s theory”
(Kelley, 1967, p. 192). Specifically, the two central ideas on which Kelley focused were:

1. In the attribution process “the choice is between external attribution and internal …
attribution” (Kelley, 1967, p. 194).

2. The procedure of arriving at these external or internal attributions is analogous to
experimental methodology.5

Two questions must be considered here, one historical, one substantive. First, were these
two ideas really central to Heider’s theory, as Kelley claimed? Second, do the two ideas
together provide a strong foundation for a theory of behavior explanation?

As argued earlier, Heider applied the external–internal dichotomy to action outcomes
that either succeeded or failed. Explanations of intentional action, by contrast, involve the
agent’s reasons, and for Heider an external–internal distinction was not useful in this
domain. Nonetheless, to support the notion that Heider considered the process of explain-
ing as analogous to experimental methodology (what Kelley termed the covariation princi-
ple), Kelley quoted a passage from the very end of Heider’s book (Heider, 1958, p. 297),
which is itself largely based on the section “Attribution of Desire and Pleasure” (Heider,
1958, pp. 146–160). In this section, however, Heider focused entirely on the attribution of
impersonal or unintentional events such as enjoyment. Heider never claimed that all
behavioral events are explained by means of covariation. Historically, then, Heider provided
at best partial backing for Kelley’s two postulates.

Whether Kelley correctly represented Heider or not, the question remains whether
Kelley’s theory generally accounts for folk explanations of behavior – and not merely for
explanations of unintentional events. As a starting point, consider the following example
that Kelley offers to illustrate the attribution process:

Am I to take my enjoyment of a movie as a basis for an attribution to the movie (that it is

intrinsically enjoyable) or for an attribution to myself (that I have a specific kind of desire

relevant to movies)? The inference as to where to locate the dispositional properties responsi-

ble for the effect is made by interpreting the raw data (the enjoyment) in the context of subsidi-

ary information from experiment-like variations of conditions. (Kelley, 1967, p. 194)

This example features an actor’s attempt to explain enjoyment – an unintentional event.
Indeed, throughout his chapter Kelley applies this attribution analysis to “effects such as
experiences, sensations, or responses” (p. 196) and “impressions” (p. 197), as well as arousal

Chadee_c03.indd 77Chadee_c03.indd 77 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

78 Bertram F. Malle

states and evaluative reactions (pp. 231–232). All of these events are unintentional. Kelley
occasionally claimed that his model also extended to the case of “inferring a person’s inten-
tions from knowledge of the consequences of his actions” (p. 196, see also p. 193). However,
no theory, empirical data, or examples clarified how this process might work.

Of course, the absence of such clarification is not yet proof that Kelley’s model fails to
account for intentional actions. But difficulties emerge quickly when one applies either the
person–situation dichotomy or covariation reasoning to intentional actions.

Consider the following scenario:

Having just arrived in the department as a new Assistant Professor, Pauline finds in her mail-

box a note that says “Let’s have lunch tomorrow. Faculty club at 12:30? – Fred.” Pauline is a bit

surprised. She met Fred W. during her interview, but she wouldn’t have expected him to ask her

out for lunch.

Pauline now tries to explain Fred’s (by assumption intentional) action of leaving the note
in her mailbox. Kelley’s attribution model would suggest that Pauline’s choice is between a
person attribution (something about Fred caused the action) and a situation attribution
(something about her or the circumstances caused the action). But right away, this is a confus-
ing choice. Surely something about Fred must have been present in order for him to put the
note in her mailbox: motives, an intention, a fairly controlled movement – all inescapable
implications of Fred’s action being intentional. In some basic sense, intentional actions are
always caused by the person (D’Andrade, 1987; Heider, 1958; Kruglanski, 1975; Malle &
Knobe, 1997). At the same time, the situation had to play some role in Fred’s choice as well –
but the situation as subjectively represented by Fred. He wouldn’t have put this note in Pauline’s
mailbox if he hadn’t expected her to check her mailbox and if he hadn’t thought about a good
time and place for the lunch and had not hoped her response to the invitation to be positive.
If Pauline knew Fred’s deliberations, she would at once understand and be able to explain why
he wrote the note. By contrast, she would not be able to explain it with the obvious conclusion
that “something about Fred caused the action” or the vague insight that “something about the
situation caused the action.”

Is there a sense in which the “experimental methodology” Kelley has in mind could prove
useful? If Pauline engaged in covariation reasoning, she would have to ask the three famous
questions about consensus, distinctiveness, and consistency to arrive at a plausible explana-
tion of Fred’s action. If we play it by the books, however, this will generate few answers.
Assume that no other faculty member has so far, on Pauline’s first day on the job, left a note
in her mailbox (low consensus). What can she conclude from that? Fred may have wanted
to welcome her, or go out on a date with her, or discuss some common research ideas with
her – there are just too many possibilities. All of these explanations might be labeled
“person attributions,” because they are possible goals/desires Fred had when leaving the
note. But making a sheer person attribution is uninformative in this case. Pauline does not
doubt that Fred had some goal; she rather wonders what goal Fred had.

Similar problems arise with the other covariation questions: has Fred performed this
kind of action before? Pauline won’t know, but assuming she finds out that this is the first
time Fred did it (low consistency), she learns only that his action has something to do with

Chadee_c03.indd 78Chadee_c03.indd 78 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

Attribution Theories: How People Make Sense of Behavior 79

her or with this particular point in time, while still not knowing why he did it. Finally, con-
sider that Fred has left this kind of note with other people as well (low distinctiveness). In
this case, Pauline may conclude that Fred shows some habit, which is also of limited use. She
would want to know specifically whether his habit is to invite all new faculty members, or
only women, or members of her research area, etc. Systematic collection of such covariation
information (if available) may sometimes prove helpful by ruling out some possible expla-
nations of the action in question. For example, if Pauline somehow learns that Fred doesn’t
act this way with others and that others don’t act this way, it can’t just be a habit or depart-
ment culture. But there are still critical steps to go, and in those steps the explainer will not
try to choose between “person” and “situation” but rather infer specific goals, beliefs, and
the like, that were – in the explainer’s assessment – the reasons for the agent’s action.

Over the years, Kelley’s covariation model and its refinements have been tested empirically
and appeared to receive reasonable support (e.g., Försterling, 1992; McArthur, 1972; Sutton &
McClure, 2001). However, these studies suffered from a pair of heavy demands. First, they
used person–situation ratings as dependent measures; people were not actually asked to
provide natural explanations. Second, the experimenter always presented covariation infor-
mation to the participants, and unsurprisingly they made use of it. There is no evidence that
people spontaneously search for covariation information when trying to explain behavior.
Lalljee, Lamb, Furnham, and Jaspars (1984) asked their participants to write down the kind
of information they would like to have in order to explain various events, and covariation
information was in low demand under these conditions. Similarly, Ahn, Kalish, Medin, and
Gelman (1995) allowed people to choose between receiving covariation information or some
other information, and explainers were less interested in covariation information than in
information about generative forces or mechanisms. Nonetheless, we can safely assume that
people occasionally seek out covariation information for such unintentional events as head-
aches or moods and such outcomes as success or failure. These events occur repeatedly, in
different contexts, and for many people, thus making covariation information more readily
available. However, as we have seen, covariation reasoning about person and situation causes
is ineffectual in the case of explaining intentional actions (Knobe & Malle, 2002).

To summarize, Kelley’s (1967) model of attribution contains two core propositions: (a) that
attribution is a choice between external and internal causes and (b) that the cognitive pro-
cedure by which people arrive at this choice is covariation assessment. Both propositions are
problematic. First, the internal–external dimension may be a relevant distinction in explana-
tions of unintentional events, but it does not capture people’s explanations of intentional
action. Second, covariation assessment is used far less than has been commonly assumed, and
it is not at all useful as a method to generate explanations of intentional actions.

Other Attribution Research

The three classic works by Heider, Jones and Davis, and Kelley were not the only important
contributions to the study of attribution. In particular, Weiner made seminal contribu-
tions to our knowledge of outcome attribution (Weiner, 1986; Weiner et al., 1972). In the
domain of achievements, he analyzed the emotions and evaluations people have of others

Chadee_c03.indd 79Chadee_c03.indd 79 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

80 Bertram F. Malle

who succeed or fail. He highlighted the causal dimension of stability as complementing
externality– internality and showed that people who failed because of lack of effort (unsta-
ble internal) were evaluated more negatively than those who failed because of inability
(stable internal). Later Weiner also analyzed other outcomes that happen to people, such as
sickness or stigma, and he focused on the dimension of controllability to account for social
perceivers’ responses to such outcomes (Weiner, 1995). For example, people are more angry
at agents who suffer negative outcomes brought about by controllable causes (e.g., illness
because of risky behavior) than agents who suffer negative outcomes brought about by
uncontrollable causes (e.g., illness because of a genetic precondition).

An overlooked aspect of Weiner’s (1986, 1995) work is that it provides considerable evi-
dence against Kelley’s theory, because two additional attribution dimensions are needed
besides external–internal to account for people’s emotional and moral responses to out-
comes, and Weiner’s predictions were developed without reference to covariation reasoning.

However, like models of dispositional attribution and covariation reasoning, Weiner’s
models of outcome attribution do not speak to Heider’s demands for a theory of action
attribution. Folk explanations of intentional action remain to be accounted for.

More in Heider’s spirit, Buss (1978) argued that ordinary people do not explain all
behavior with causes (as Kelley had suggested) but rather use reasons to explain inten-
tional behavior. Reasons and causes are fundamentally different types of explanation, so
Buss’s attribution theory confounded the two. Buss’s (1978) article drew rather negative
responses (e.g., Harvey & Tucker, 1979; Kruglanski, 1979), and mainstream attribution
theory remained unaffected by his critique. Over the next decade, other scholars launched
similar critiques, arguing that reasons are an autonomous form of explanation (Locke &
Pennington, 1982) and that attribution theories must incorporate reasons and goals
into their conceptual repertoire (Lalljee & Abelson, 1983; Read, 1987; for a review see
McClure, 2002). However, such an integration proved difficult, in part because it was not
made sufficiently clear why reasons are used to explain intentional behavior in the first
place. What makes intentional behaviors so special that they require a unique mode of
explanation?

An important contribution to attribution theory emerged in a series of papers on the
conversational nature of explanations (Hilton, 1990; Kidd & Amabile, 1981; Turnbull, 1986),
which characterized explanations as answers to why questions. Such question–answer pairs
sometimes occur in people’s own heads, but more often they occur as an actual conversa-
tional exchange between a questioner and an explainer. This conversational analysis comes
with the important implication that, in answering a why question, explainers must take into
consideration (a) exactly what the questioner finds puzzling (Hilton & Slugoski, 1986;
Turnbull, 1986) and (b) what information the questioner already has available (Slugoski,
Lalljee, Lamb, & Ginsburg, 1993). In a sense, the explainer anticipates what kinds of possible
answers the questioner had in mind when asking the question (Bromberger, 1965).

The insight that explanations are subject to conversational processes was a minor revolu-
tion, because it pulled attributions out of their cognitive rabbit hole and highlighted the
fundamentally social nature of explanations. However, research into the conversational fea-
tures of explanations still did not challenge the conceptual apparatus of person–situation
causes inherited from Kelley (see, e.g., Norenzayan & Schwarz, 1999).

Chadee_c03.indd 80Chadee_c03.indd 80 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

Attribution Theories: How People Make Sense of Behavior 81

The preceding overview has pointed to a number of shortcomings in classic and con-
temporary attribution theories. First, after Heider, the conceptual framework underlying
people’s explanations was grossly simplified, ignoring such important concepts as inten-
tion, goal, and reason. As a result, no adequate accounts of intentional action explana-
tions were offered. Second, covariation analysis was the only psychological process
postulated to drive people’s construction of explanations, but because it is used much less
than was previously assumed, other processes need to be explored. Third, the conversa-
tional aspects of explanations must be integrated with the unique conceptual and cogni-
tive aspects of intentional action explanations.

To resolve these problems, we need to first locate behavior explanations in their proper
folk-conceptual framework – the network of concepts that people use to make sense of
human behavior. Second, we need to specify the psychological processes that enable people
to construct explanations. And third, we need to track the communicative and linguistic
manifestations of explanations because they are clearly used for a variety of social-interac-
tive goals, and language is the tool that accomplishes these goals. The folk-conceptual the-
ory of behavior explanation (Malle, 1999, 2004) tries to meet all three of these objectives.

The Folk-Conceptual Theory of Behavior Explanation

Folk concepts of mind and behavior

Past attribution theories ascribed to laypeople a simple framework of “effects” (behaviors,
outcomes, or events) and “causes,” with the latter falling into personal (dispositional or
internal) and situational (external) causes. This framework is incompatible with what we
know about children’s emerging theory of mind and behavior – the conceptual network
that 4–5-year-olds rely on when interpreting human behavior (Gopnik & Meltzoff, 1997;
Perner, 1991; Wellman, 1990). In children’s theory of mind, we see the importance of a
concept of intentionality, of the mind contrasted with observable behaviors, and of specific
mental states, such as beliefs and desires, that explain intentional behavior (Bartsch &
Wellman, 1995). It is unlikely that when children grow up they forget these concepts and
suddenly begin to explain behavior using a person–situation dichotomy.

Instead, the adult’s folk conception of mind and behavior is continuous with the child’s
framework, with more numerous and more subtle distinctions but essentially the same
two-part structure. First, people have perceptual and cognitive systems that filter, group,
and integrate certain stimuli into such concepts or categories as agent, intention, belief, and
reason (D’Andrade, 1987; Kashima, McKintyre, & Clifford, 1998; Leslie, 1995; Malle, 2004;
Malle & Knobe, 1997). Second, people make assumptions about these categories and their
relationships (Fodor, 1992; Morton, 1996). For example, coordinated movements of agents
are classified into the category intentional action (Wellman & Phillips, 2001; Woodward,
Sommerville, & Guajardo, 2001), and the concept of intentional action relies on the inter-
play of multiple mental state categories, including belief and desire (Malle & Knobe, 1997).
Intentionality is arguably the core of this framework (Malle et al., 2001). It directly connects
behavior with mind by classifying a behavior as intentional when it is characteristically
generated by certain mental states (such as belief, intention, and awareness). This postulate

Chadee_c03.indd 81Chadee_c03.indd 81 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

82 Bertram F. Malle

had the least impact on subsequent attribution research, and the greatest weakness of past
attribution theories has therefore been the treatment of intentional action. The folk-con-
ceptual theory of explanation tries to address this weakness.

Intentionality. When explaining behavior, people distinguish sharply between intentional
and unintentional events (Malle, 1999, 2004, 2007). Social perceivers show a high level of agree-
ment in their intentionality judgments (Malle & Knobe, 1997), and they do so by relying on a
shared folk concept of intentionality (Figure 3.1). This concept includes five requirements for
an action to be judged as intentional: the action must be based on a desire for an outcome,
beliefs about the action’s relationship with this outcome, a resulting intention to perform the
action, and skill and awareness when actually performing it (Malle & Knobe, 1997, Studies 2–4).
Subsequent studies showed that the core concept of intention is seen as a choice and commit-
ment to act that flows from a reasoning process, in which the agent weighs a number of beliefs
and desires and settles on a course of action (Malle & Knobe, 2001; Monroe & Malle, 2010).

Modes of explanation. Unintentional events are explained by referring to “mechanical”
causal factors (e.g., physical objects and events, but also traits or others’ behaviors), and we
may label them cause explanations (top of Figure 3.2). That is because people assume no
other link that mediates between behavior and explanation besides causality (i.e., no com-
ponents of intentionality such as awareness or intention). Consider the following examples,
extracted from original transcripts:

I almost failed my exams. – Why? – Oh, ’cuz I didn’t really prepare for them.

My dad got mad with me because something was wrong with my computer and he did not

know how to fix it.

A friend cried on the phone. – Why? – She felt that no one loved her.

In contrast, explanations of intentional behavior are far more complex because the folk
concept of intentional action is far more complex. As a result, explanations of intentional
action break down into three modes (bottom of Figure 3.2): reason explanations, causal
history of reason explanations, and enabling factor explanations.

Reason explanations. Reason explanations are the most frequently used mode, compris-
ing about three-quarters of all action explanations (Malle, 2004; Malle, Knobe, & Nelson,
2007). They link directly to the heart of the intentionality concept – the reasoning process

Belief Desire

Skill Intention Awareness

Intentionality

Figure 3.1 A model of the folk concept of intentionality.

Adapted from “The folk concept of intentionality,” by B. F. Malle and J. Knobe, 1997, Journal of Experimental Social

Psychology, 33, 101–121. © Lawrence Erlbaum Associates, Inc., with permission.

Chadee_c03.indd 82Chadee_c03.indd 82 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

Attribution Theories: How People Make Sense of Behavior 83

leading up to an intention. The concept of intentionality specifies two paradigmatic types
of reasons that precede the formation of an intention: the agent’s desire for an outcome and
a belief that the intended action leads to that outcome. For example, a student explained
why she chose psychology as her major by saying, “I want to go to graduate school in
counseling psychology [desire]; I think psychology is the right major to have as background
for counseling psychology [belief].” In many naturally occurring explanations, other rea-
sons are mentioned in addition to or instead of the paradigmatic reasons, such as desires for
avoiding alternative outcomes, beliefs about the context, beliefs about consequences, and
expressions of valuing the action itself.

Reasons have two defining features: subjectivity and rationality. Subjectivity refers to the
fact that reason explanations are designed to capture the agent’s subjective reasons for act-
ing. That is, social perceivers normally try to reconstruct the considerations the agent
underwent when forming an intention, and they thus take the agent’s subjective viewpoint
when explaining the action. For example, the explanation “She thought she was late for
her class” in response to the question “Why did she rush off?” illustrates the subjectivity
assumption, because the explainer subtly distances himself from the agent’s belief and
implies that, in reality, she probably was not late. But it was that subjective belief, not objec-
tive reality, that guided the agent’s action and thus explains it.

Rationality, the second defining feature of reason explanations, refers to the fact that the
contents of mental states that are cited as reasons have to hang together so as to offer sup-
port for the “reasonableness” of the intention and action they brought about. For an inten-
tional action to be adequately explained by reasons, the action must fulfill the agent’s
predominant desire in light of her beliefs. Philosophers often speak of a “practical reasoning
argument,” in which reasons are the premises and the decision to act is its conclusion.
Consider the following explanation: “Why did you go running?” – “Because I wanted to get
in better shape, and … I figured that I can do that by going running.” Schematically:

X wanted O [to get in better shape].

X believed that A [going running] leads to O [getting in better shape].

Therefore, X decided to A [go running].

Causes
Unintentional

Behavior

Intentional
Behavior

Intention

Enabling
Factors

Reasons
Causal History

of Reasons

Figure 3.2 Four modes of explanation for unintentional and intentional behavior.

Adapted from “How people explain behavior: A new theoretical framework,” by B. F. Malle, 1999, Personality and

Social Psychology Review, 3, 23–48. © Lawrence Erlbaum Associates, Inc, with permission.

Chadee_c03.indd 83Chadee_c03.indd 83 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

84 Bertram F. Malle

In everyday conversation, the logic of rationality will be somewhat cursory and
the premises will not always be made explicit. In the earlier example of an agent rushing off,
the action was rationally supported by her belief that she was late for class and the
unmentioned but implied desire to be on time for class. The action would not have been
rationally supported if the agent had thought there was plenty of time left or if she had had
no desire to be on time. Social perceivers would consider it irrational to rush off in this case,
or they would assume that some other reason can explain the action in a rational way.

Causal history of reason explanations. Even though people explain most intentional behav-
iors by reference to the agent’s reasons, they explain some of them by pointing to factors that
lay in the background of those reasons. These factors can be subsumed under the label causal
history of reasons (CHR) and include such forces as the agent’s unconscious mental states,
personality, upbringing, culture, and the immediate context (Malle, 1994, 1999; Malle et al.,
2000; O’Laughlin & Malle, 2002). Whereas reason explanations try to capture what the agent
herself considered and weighed when deciding to act, causal history explanations take a step
back and try to capture what led up to the agent’s reasons in the first place. For example, when
clarifying why Kim didn’t vote, an explainer might say “She is lazy” or “Her whole family is
apolitical.” These statements provide explanations of an intentional action, but they do not
pick out Kim’s subjective reasons for not voting. Causal history of reason explanations help
explain an intentional action by citing causal antecedents to the agent’s reasoning and her
decision to act, but there is no assumption that the agent actively considered those anteced-
ents in her reasoning process. Hence, when an explainer states that “Kim didn’t vote because
she is lazy,” he does not imply that Kim reasoned: “I am lazy; therefore I shouldn’t vote.”

Thus, even though CHR explanations help explain intentional actions, they are not sub-
ject to the constraints of subjectivity and rationality as reason explanations are (Malle et al.,
2000). The agent need not have considered or even been aware of the causal history factors
cited in the explanation, nor do CHR factors provide rational support for an explained
action. Offering CHR explanations is therefore not an act of perspective taking but rather
takes attention away from the agent’s subjective processes of choice and intention and
locates the agent’s action in a broader causal nexus.

Enabling factor explanations. The third, and relatively rare, mode of explaining intentional
action refers to factors that enabled the action to come about as it was intended. These ena-
bling factor explanations refer to the agent’s skill, effort, opportunities, facilitating circum-
stances, and the like. Whereas reason explanations and CHR explanations focus on clarifying
what motivated the agent’s intention and action, enabling factor explanations take it as a
given that the agent had motives and attempt to clarify how it was possible that the action
was successfully performed. For example, “She hit her free throws because she had practiced
all week.” This is not a motivational account of why the agent decided to hit the free throws;
rather, the agent’s practicing is identified as the critical factor that allowed her to perform
the action the way she had intended. Consequently, enabling factor explanations are offered
primarily for difficult actions (Malle et al., 2000, McClure & Hilton, 1997).

The folk-conceptual framework for explaining behavior thus contains four modes of
explanation: people explain unintentional behavior with causes, and they explain intentional
behavior by either referring to the agent’s reasons, citing causal factors that lay in the history
of those reasons, or clarifying what factors enabled an agent to successfully perform an

Chadee_c03.indd 84Chadee_c03.indd 84 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

Attribution Theories: How People Make Sense of Behavior 85

intended action. Before we move on to the psychological processes that guide people’s choices
among these modes, we need to briefly examine reason explanations more closely, for they
have two further features of psychological significance: explainers can offer either belief rea-
sons or desire reasons, and they can mark those reasons as subjective mental states by using
appropriate mental state verbs (such as “thinks,” “wanted”) or leave them unmarked.

Belief and desire reasons. In people’s folk concept of intentionality, both beliefs and desires
serve as necessary conditions of an intention to act (Malle & Knobe, 1997), and both are
frequently cited in explanations of intentional action. For example, when explaining why
Ian has been working so much lately, one might cite a desire such as “He wants that promo-
tion” or a belief such as “He realizes the project is due in a week.” At times it may not matter
whether the explainer mentions a belief reason or a desire reasons because one implies the
other (“He thinks hard work will get him the promotion” ⇔ “He wants that promotion”).
But at other times it matters quite a bit. For one thing, belief reasons, more than desire
reasons, provide idiosyncratic details about the agent’s decision-making process, such as
rejected options, specific plans of action initiation, and considered long-term consequences.
For another, belief reasons refer to the agent’s thinking and knowledge, drawing attention
to the agent’s rational, deliberative side, whereas desire reasons highlight what the agent
wants, needs, hence lacks (Malle et al., 2000).

Mental state markers. A reason explanation can be linguistically expressed in two different
ways. The explainer may use a mental state verb to mark the type of reason cited (i.e., a belief
or desire), or the explainer may omit such a verb and directly report the content of that rea-
son. Suppose our explainer is faced with the question “Why did she go to the Italian café ?”
If he chose to cite a desire reason, he could use the marked form of desire ascription:

She went to the café because she wants to have an authentic cappuccino.

Or he could use the unmarked form:

She went to the café [ ____ ] to have an authentic cappuccino.

Likewise, if the explainer chose to cite a belief reason, he could use the marked form of
belief ascription:

She went to the café because she thinks they have the best cappuccino.

Or he could use the unmarked form:

She went to the café because [ ____ ] they have the best cappuccino.

Marked or unmarked reasons do not express two different hypotheses about why the
action was performed; rather, they express the same hypothesis in two different ways. This
difference is not trivial, however. Citing or omitting mental state markers can serve signifi-
cant social functions, both for explainers’ self-presentation and for expressing their opinion
about the agent (Malle, 1999; Malle et al., 2000).

In the context of several examples of CHR explanations and reason explanations, Table 3.1
shows both types of reasons, marked and unmarked. (More examples of all modes and
types can be found in Malle, 2004.) Note that causal history factors can refer to the person,

Chadee_c03.indd 85Chadee_c03.indd 85 12/13/2010 2:17:57 PM12/13/2010 2:17:57 PM

86 Bertram F. Malle

the situation, other people, traits, mental states, or interactions among them. In principle,
this choice could have psychological significance, but in our studies we have found no
interesting correlates of “causal locus” – neither within CHR explanations nor within cause
explanations (e.g., Malle et al., 2007; O’Laughlin & Malle, 2002).

Interlude: measurement concerns

Previously offered alternatives to attribution theory were not able to unseat the traditional
theory. This may have been in part because those alternatives were not comprehensive
enough – focusing, for example, only on goals or only on conversational structures. But
another obstacle was researchers’ deeply ingrained habit of using only rating scales to assess
attributions. Participants were typically asked to express their explanations on predefined
person–situation scales rather than in the more natural form of verbal utterances. As a
result, participants had to transform their complex explanatory hypotheses into simple rat-
ings, and the fact that they were willing and able to do so was mistaken for evidence that
they actually thought in those simple terms.

A few studies analyzed free-response explanations, but the coding was again limited to
person–situation categories. This person–situation coding reflects the linguistic surface of
explanations, such as the use of mental state markers (McGill, 1989; Nisbett, Caputo, Legant, &
Marecek, 1973; for evidence and discussion, see Malle, 1999, Study 4, and Malle et al., 2000,
Study 4). Predictable distortions result from treating intentional action explanations within
the person–situation frame, illustrated by the following example (cf. Antaki, 1994):

Why are you going to Iceland for your holidays? – Because it’s cool there.

The standard attribution treatment of such an explanation would be to call it a “situa-
tion cause.” However, the cool weather in Iceland can hardly cause the agent from afar to

Table 3.1 Reason explanations and causal history of reason (CHR) explanations for two
behaviors.

Behavior Reason explanation CHR explanation

Kim chose not to vote

in the last election.

She thought none of the candidates

was trustworthy.

She doesn’t realize that every

vote counts.

[marked belief reason]

She didn’t want to support the system. She is lazy.

[marked desire reason]

Brian used heavy drugs

last Sunday at the party.

He was curious what it would feel

like.

A bunch of others used them.

[marked desire reason]

He thought it would be cool.

[marked belief reason]

He grew up in a drug-dealing

home.

Note. The terms marked and unmarked refer to reason explanations that are expressed either with or without a

mental state marker (“he thought,” “she wanted”). Adapted from “How people explain behavior: A new

theoretical framework,” by B. F. Malle, 1999, Personality and Social Psychology Review, 3, 23–48. © Lawrence

Erlbaum Associates, Inc., with permission.

Chadee_c03.indd 86Chadee_c03.indd 86 12/13/2010 2:17:58 PM12/13/2010 2:17:58 PM

Attribution Theories: How People Make Sense of Behavior 87

go there; rather, the agent thinks that it is pleasantly cool in Iceland, and that is her (belief )
reason for going there. (It would still be her belief reason even if it were in fact warm in
Iceland.) Instead of trying to diagnose the locus of some “cause” in the vague space
between person and situation, we have to recognize that most folk explanations of inten-
tional behavior refer to the agent’s reasons in light of which and on the grounds of which
she acted.

Once the proper folk-conceptual categories (of reasons, causal histories, etc.) are applied to
naturally uttered explanations, they can be reliably classified and analyzed (Malle, 1998/2009);6
and these classifications show more sensitivity to important psychological variables than either
person–situation codings or person–situation ratings (e.g., Malle et al., 2007; O’Laughlin &
Malle, 2002). The researcher does the theoretical work here, and participants simply do what
they do in everyday life – express explanations of behavior in their own words.

The lesson should be clear: if theoretical progress is to be made in the attribution domain,
both theory and measurement must reflect the complexity of people’s explanations. Multiple
modes, types, and linguistic forms have to be described, conceptually analyzed, and ade-
quately measured.

Psychological processes underlying behavior explanations

The conceptual layer of the folk-conceptual theory of explanation showed that explainers
have a number of choices to make when explaining behavior (Figure 3.3): they need to deter-
mine whether the behavior in question is intentional or not; if intentional, whether to explain
it with reasons, causal history factors, or enabling factors; and if a reason is chosen, whether
to cite the agent’s beliefs or desires and whether to use a mental state marker or not.7

if unintentional if intentional

Determine intentionality of behavior

offer cause offer EF offer reason offer CHR

belief desire

marked

unmarked

marked

unmarked

Figure 3.3 Choice points in constructing a behavior explanation.

Adapted from B. F. Malle, 2004, How the mind explains behavior: Folk explanations, meaning, and social

interaction. Cambridge, MA: MIT Press. Note: EF = enabling factor explanation. CHR = causal history of reason

explanation.

Chadee_c03.indd 87Chadee_c03.indd 87 12/13/2010 2:17:58 PM12/13/2010 2:17:58 PM

88 Bertram F. Malle

Adding a process layer now to the theory, a first important question is what psychological
processes determine the explainer’s choices among these various explanatory tools. A fair amount
of research is available to answer this question, and I will discuss it below. A second process
question is how the explainer goes about selecting a specific explanatory content (e.g., the belief
reason, “because she thought the door was locked”). No theory of behavior explanation will be
able to predict specific contents of explanations, but the psychological processes people rely on
when searching for those contents may be identifiable. Currently, very little research is available
on this issue, in part because the process of covariation reasoning has been so dominant in the
research literature. It is evident, however, that many more processes guide people’s search for
specific explanation contents – among them knowledge structures, projection, simulation,
explicit perspective-taking, and contrast analysis (Ames, 2004; Hilton, 2007; Lalljee & Abelson,
1983; Malle, 2004, 2008). All of these processes deserve greater attention in future research.8

What do we know about the psychological determinants of people’s explanatory choices?
At least three such determinants must be considered:

1. Judgments of behavior attributes
2. Pragmatic goals
3. Information resources

Judgments of behavior attributes. Before offering an explanation, social perceivers make
several (often implicit) judgments about the behavior to be explained. One such judgment
obviously concerns the intentionality of the behavior. If considered unintentional, explana-
tions will refer to a cause; if considered intentional, explanations will cite a reason, causal
history, or enabling factor (Malle, 1999).

Among intentional behaviors, a second judgment concerns the difficulty of the action.
If the action is considered difficult to produce, explainers will often choose enabling factors;
otherwise, they are likely to choose reasons or causal histories (Malle et al., 2000; McClure &
Hilton, 1997).

A third judgment determines whether the to-be-explained behavior is singular or repre-
sents a trend (across time or across multiple agents). If the behavior is judged to be a trend,
the rate of CHR explanations increases (O’Laughlin & Malle, 2002), primarily because mul-
tiple instances (or agents) can have multiple reasons, and a CHR explanation can some-
times tie these reasons together by citing an antecedent for all of them (e.g., “Why did you
go shopping so many times this week?” – “Because I have three children”).

Pragmatic goals. Pragmatic goals refer to the social projects that explainers try to accom-
plish with their explanations, such as lessening another person’s confusion, managing their
own status in the interaction, or fending off blame. Two groups of goals can be distin-
guished by their primary beneficiary: audience design, which chiefly benefits the conversa-
tion partner, and impression management, which chiefly benefits the explainer.

Audience design consists in tailoring an explanation so the audience can learn what it
wants to know. The clearest case of such design is when the explainer matches an explana-
tion mode to the type of question asked (Malle et al., 2000; McClure & Hilton, 1998). This
question can inquire either about the agent’s immediate motivation (“What for?”, which
triggers reasons), about the background of that motivation (“How come?”, which triggers

Chadee_c03.indd 88Chadee_c03.indd 88 12/13/2010 2:17:58 PM12/13/2010 2:17:58 PM

Attribution Theories: How People Make Sense of Behavior 89

causal histories of reasons), about the factors that enabled successful action perfor-
mance (“How was this possible?”, which triggers enabling factors). More subtle adjustments
include offering a belief reason when it can be assumed that the audience already knows the
desire reason, or offering a CHR explanation when the reasons for the action are obvious.

Whereas audience design falls under the conversational maxim of relevance (Grice, 1975;
Sperber & Wilson, 1986), impression management does not merely try to optimize com-
munication but rather is an act of social influence. Its impact can reach all explanatory
choices, from both the actor and the observer perspective (Malle et al., 2000). As actors,
people increase their use of reasons, and especially belief reasons, when trying to appear
rational. As observers, they explicitly add a mental state marker to a belief reason when they
want to distance themselves from the agent (e.g., “Why is he looking at apartments?” –
“He thinks I am moving in with him”).

Information resources. Deficits in relevant information about the agent, the behavior, or
the context can limit the explainer’s ability to accomplish his goals, and such deficits often
require adjustments of one’s choices among explanatory tools. This can be seen, for exam-
ple, in the choice of belief reasons vs. desire reasons. We find that observers, compared to
actors, offer more desire reasons than belief reasons (Malle et al., 2007). That is because
desire reasons are easier to infer from contextual and cultural knowledge, whereas belief
reasons are more idiosyncratic and sometimes accessible only to the actor. However, when
observers know the agent well or were present when the action took place, their rate of
belief reasons increases to resemble that of actors (Malle et al., 2007).

A somewhat more complex case is the role of information resources in the choice between
reasons and causal history explanations (Malle et al., 2007; O’Laughlin & Malle, 2002).
In short, when explaining their own intentional behaviors (as actors), people are in the best
position to have every kind of information available and predominantly (in 80% of cases)
offer reasons. When explaining other people’s behaviors (as observers), people lack various
kinds of information and will try to infer or construct reasons, but this construction reaches
a limit, yielding an average of 65% reasons. Interestingly, this lower rate of reasons persists
even when observers are present at the time of action and when observers know the agent
well (Malle et al., 2007). This suggests that the actor–observer asymmetry is due not to
general knowledge differences but to access differences at a more fundamental level. For
example, because actors typically consider their reasons for acting before they decide to act,
they have the ability to directly recall their own reasons. Observers, by contrast, must guess
those reasons or infer them from observable information.

We can systematically relate the discussed psychological determinants to the range of
explanatory choices displayed in Figure 3.2, forming linear equations that connect a par-
ticular explanatory choice to a set of psychological determinants (Table 3.2). These equa-
tions have received some empirical support (Malle, 1999; Malle et al., 2000, 2007), but more
precise relations should emerge from future research.

This theoretical approach of linking the choice of explanatory tools to a limited set of
psychological determinants has been successfully applied to predicting strategies of impres-
sion management in explanation (Malle et al., 2000), differences between explanations of
group and individual behaviors (O’Laughlin & Malle, 2002), and a variety of actor–observer
asymmetries in explanations of behavior (Malle et al., 2007). Other domains are open to

Chadee_c03.indd 89Chadee_c03.indd 89 12/13/2010 2:17:58 PM12/13/2010 2:17:58 PM

90 Bertram F. Malle

investigation as well, such as close relationships, negotiation, psychopathology, and cross-
cultural comparisons (Malle, 2004). To illustrate one application, I describe the death and
rebirth of the well-known actor–observer asymmetry in attribution.

Application: actor–observer asymmetries

Few hypotheses have been as widely accepted in the social-psychological literature as Jones
and Nisbett’s (1972) classic hypothesis of an actor–observer asymmetry in explanations:
“There is a pervasive tendency for actors to attribute their actions to situational require-
ments, whereas observers tend to attribute the same actions to stable personal dispositions”
(p. 80). Surprisingly, a meta-analysis of 173 published studies testing this hypothesis showed
that the average effect size for this pervasive difference between actor and observer attribu-
tions was zero (Malle, 2006). That is, when explanations are analyzed in terms of traditional
attribution concepts – referring to either person (or disposition) vs. situation causes – no
difference exists between actors’ and observers’ explanations of behavior.

But do people really explain their own behavior the same way as they explain other peo-
ple’s behavior? They do not. But to see this, we need to apply a different theory. When
explanations are analyzed within the terms of the folk-conceptual theory, strong and relia-
ble actor–observer differences emerge (Malle et al., 2007). In particular, consideration of
the choices people make when explaining intentional behavior and of the psychological
determinants of these choices leads to three actor–observer hypotheses (for more detail, see
Malle et al., 2007, pp. 495–496).9

First, actors are predicted to use relatively more reason explanations (and fewer CHR
explanations) than observers do. That is because actors have better access to their reasons
(often through direct recall) and engage in stronger impression management to portray
themselves in a positive light (which reasons accomplish better than CHR explanations).

Second, actors are predicted to use relatively more belief reasons (and fewer desire rea-
sons) than observers do. That is because actors’ advantage in information access is espe-
cially pronounced for the more idiosyncratic and context-specific belief reasons.

Third, actors are predicted to use relatively fewer mental state markers for belief reasons
than observers do. That is because actors directly represent the content of their beliefs (e.g.,
“The plants are dry”), not their own mental state of believing (“I believe: the plants are
dry”; Moore, 1993; Rosenthal, 2005). Actors will therefore typically leave their belief reasons

Table 3.2 Linking selected explanatory choices to their psychological determinants.

(1) REA/CHR/EF vs. CAU = b
1
(behavior intentionality)

(2) EF vs. REA/CHR = b
1
(behavior difficulty) + b

2
(explanatory question) +

b
3
(difficulty × question)

(3) CHR vs. REA = b
1
(behavior trend) + b

2
(information) + b

3
(impression management)

(4) belief REA vs. desire REA = b
1
(information) + b

2
(impression management)

(5) marked belief vs. unmarked belief = b
1
(impression management)

Note. REA = Reason explanation. CHR = Causal history of reason explanation. EF = Enabling factor

explanation. CAU = cause explanation.

Chadee_c03.indd 90Chadee_c03.indd 90 12/13/2010 2:17:58 PM12/13/2010 2:17:58 PM

Attribution Theories: How People Make Sense of Behavior 91

unmarked: “Why did you turn the sprinkler on?” – “Because the plants were dry.” Observers,
by contrast, represent the actor as having beliefs and will therefore typically mark those
beliefs (especially when they don’t share them): “Why did she turn the sprinkler on?” –
“Because she thought the plants were dry.”

Tests of these predictions across nine studies provided consistent results (Malle et al.,
2007): Effect sizes for the three novel asymmetries averaged between d = 0.40 and d = 0.69,
whereas effect sizes for the classic asymmetry averaged between d = −0.08 and d = 0.10.
Thus, actors and observers differ in how they explain behavior; but these differences are
revealed only once we give up the traditional person–situation framework and instead assess
behavior explanations as part of people’s folk theory of mind and behavior – centered on
the concept of intentionality and giving rise to multiple modes and types of explanation.