Psychology of intimate relationships

PS68CH15-Finkel ARI 16 November 2016 8:6

The Psychology of Close
Relationships: Fourteen
Core Principles
Eli J. Finkel,1 Jeffry A. Simpson,2

and Paul W. Eastwick3
1 Department of Psychology and Kellogg School of Management, Northwestern University,
Evanston, Illinois 60208; email: [email protected]
2 Department of Psychology, University of Minnesota, Minneapolis, Minnesota 55455;
email: [email protected]
3 Department of Psychology, University of California, Davis, California 95616;
email: [email protected]

Annu. Rev. Psychol. 2017. 68:383–411

First published online as a Review in Advance on
September 1, 2016

The Annual Review of Psychology is online at
psych.annualreviews.org

This article’s doi:
10.1146/annurev-psych-010416-044038

Copyright c© 2017 by Annual Reviews.
All rights reserved

Keywords

relationship science, core principles, attachment theory, interdependence
theory, culinary approach

Abstract

Relationship science is a theory-rich discipline, but there have been no at-
tempts to articulate the broader themes or principles that cut across the
theories themselves. We have sought to fill that void by reviewing the
psychological literature on close relationships, particularly romantic rela-
tionships, to extract its core principles. This review reveals 14 principles,
which collectively address four central questions: (a) What is a relationship?
(b) How do relationships operate? (c) What tendencies do people bring to
their relationships? (d ) How does the context affect relationships? The 14
principles paint a cohesive and unified picture of romantic relationships that
reflects a strong and maturing discipline. However, the principles afford few
of the sorts of conflicting predictions that can be especially helpful in fos-
tering novel theory development. We conclude that relationship science is
likely to benefit from simultaneous pushes toward both greater integration
across theories (to reduce redundancy) and greater emphasis on the circum-
stances under which existing (or not-yet-developed) principles conflict with
one another.

383

Click here to view this article’s
online features:

• Download figures as PPT slides
• Navigate linked references
• Download citations
• Explore related articles
• Search keywords

ANNUAL
REVIEWS Further

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Contents

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
MAJOR THEORIES IN RELATIONSHIP SCIENCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
EXTRACTING PRINCIPLES: A CULINARY METAPHOR . . . . . . . . . . . . . . . . . . . . . 386
THE CORE PRINCIPLES OF RELATIONSHIP SCIENCE . . . . . . . . . . . . . . . . . . . . . 388

Set 1: What Is a Relationship? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
Set 2: How Do Relationships Operate? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
Set 3: What Tendencies Do People Bring to Their Relationships? . . . . . . . . . . . . . . . . 395
Set 4: How Does the Context Affect Relationships? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397

COMBINING THE PRINCIPLES TO REFINE OR DEVELOP THEORIES . . . . 401
Refining Existing Theories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401
Generating New Theories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402

OPTIMIZING RELATIONSHIP SCIENCE: THEORETICAL COHESION
VERSUS CONFLICT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402

CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404

[R]elationships with other humans are both the foundation and the theme of the human condition:
We are born into relationships, we live our lives in relationships with others, and when we die, the
effects of our relationships survive in the lives of the living, reverberating throughout the tissue of their
relationships.

—Ellen Berscheid (1999, pp. 261–262)

INTRODUCTION

Poets, novelists, and philosophers have long recognized the centrality of relationships to human
existence. Yet the coalescence of an integrated science devoted to understanding human rela-
tionships dates back only to the 1980s. Today, relationship science is an interdisciplinary field
that employs diverse empirical methods to understand the initiation, development, maintenance,
and dissolution of interpersonal relationships. This field addresses the structure and trajectory of
relationships, how relationships operate, and how relationship outcomes are influenced by both
the personal characteristics that people bring to their relationships and the broader context in
which relationships are embedded. Relationship scientists investigate many types of relationships,
but the primary emphasis is on close relationships—those characterized by “strong, frequent,
and diverse interdependence that lasts over a considerable period of time” (Kelley et al. 1983,
p. 38)—especially well-established romantic relationships.1 In her classic article on the “green-
ing of relationship science,” Berscheid (1999) discussed the growing coherence and influence of
relationship science on myriad scholarly fields, presciently forecasting the growth of a

1 Close relationships researchers investigate a wide range of relationships, even within the subcase of romantic relationships.
Although there are main effect differences across relationship types (Kurdek 2005), the available evidence suggests that “the
processes that regulate relationship functioning generalize across gay, lesbian, and heterosexual couples” (Kurdek 2004,
p. 880). Thus, we have no reason to believe that the 14 principles discussed below qualitatively differ across different romantic
relationship arrangements. As such, and because the vast majority of research has examined heterosexual romantic relationships,
our examples focus on the heterosexual case.

384 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

flourishing discipline in the twenty-first century (see also Campbell & Simpson 2013,
Reis 2007).

Researchers have written many reviews of the close relationships literature, including in pre-
vious volumes of the Annual Review of Psychology (e.g., Clark & Reis 1988, Gottman 1998). In
this review, we focus on the major theories that guide research in relationship science, with
a particular emphasis on those deriving from social and personality psychology. We seek to
understand what assumptions these theories share, the extent to which they align or conflict,
and how they could be augmented and complemented. Toward those ends, we attempt to ex-
tract from the literature a set of core principles for understanding close relationships and illus-
trate how articulating and organizing these core principles can promote theory refinement and
development.

MAJOR THEORIES IN RELATIONSHIP SCIENCE

Relationship science has produced many strong theories, two of which—interdependence theory
and attachment theory—have been especially influential. Interdependence theory, which began as
a game-theoretic model of dyadic interaction, traces its roots to Thibaut & Kelley’s (1959) book
The Social Psychology of Groups. This theory was first applied to close relationships in the 1970s
(Kelley 1979, Levinger & Snoek 1972) and became a dominant theory of such relationships in the
1980s (Kelley et al. 1983, Rusbult 1983). According to interdependence theory, social situations
vary along several dimensions, and this variation influences relationship processes and outcomes
(Kelley et al. 2003). For example, situations in which a man is more (versus less) dependent on his
girlfriend for rewarding experiences should increase the extent to which he monitors her behavior
for signs that she loves and is committed to him. His high level of dependence puts him in a low-
power position unless she is also highly dependent upon him. High levels of mutual dependence
typically promote cooperative behavior when partners have corresponding interests but conflictual
behavior when they have noncorrespondent interests.

Attachment theory, which initially focused on infant–caregiver relationships, traces its roots to
Bowlby’s (1969, 1973, 1980) trilogy on attachment, separation, and loss. The theory was adapted
to explain the nature of close relationships between adults in the 1980s (Hazan & Shaver 1987), and
it joined interdependence theory as a dominant model of adult relationships in the 1990s (Hazan
& Shaver 1994). According to attachment theory, people develop emotional bonds with significant
others (usually romantic partners in adulthood) and are motivated to maintain these bonds over
time (Mikulincer & Shaver 2007). People seek proximity to their primary attachment figure,
especially when they are stressed, ill, or afraid, and rely on the psychological security provided by
this person when pursuing challenging activities that can promote mastery and personal growth.
Individuals vary along two dimensions of attachment insecurity: (a) anxiety, the extent to which
they need reassurance that their attachment figures love and will stay with them, and (b) avoidance,
the extent to which they are uncomfortable with emotional intimacy and being vulnerable. Secure
individuals, who score low on both dimensions, typically display the most constructive relationship
processes and have the most positive relationship outcomes.

Several other theoretical perspectives have also been influential in relationship science, in-
cluding risk regulation theory (Murray et al. 2006), self-expansion theory (Aron et al. 2013), the
communal/exchange model (Clark & Mills 2011), the interpersonal process model of intimacy
(Reis & Shaver 1988), and the vulnerability-stress-adaptation model (Karney & Bradbury 1995).
The existence of such theories, along with many others, is a major strength of relationship science:
These theories have fruitfully guided thousands of empirical investigations into how people think,
feel, and behave in close relationships.

www.annualreviews.org • Relationship Science 385

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Nevertheless, it is not obvious how, or whether, these theories cohere and what qualities they
have in common. Some theories overlap in intended ways. For example, risk regulation theory
(Murray et al. 2006) deliberately combines elements of attachment theory and interdependence
theory. Other theories overlap in underappreciated ways. For example, the ideal standards model
(Simpson et al. 2001) focuses on standards, whereas the suffocation model (Finkel et al. 2014a) fo-
cuses on expectations, two constructs that are almost synonymous in interdependence theory. Still
other theories discuss processes that are rarely articulated elsewhere. For example, the emphasis
in the vulnerability-stress-adaptation model on stressors external to the relationship (Karney &
Bradbury 1995) is neglected in most other theories (but see Hill 1949, McCubbin & Patterson
1983). Relationship science is fortunate to have this rich assemblage of theories, but their col-
lective depiction is murky because the degree to which the field’s core principles complement,
circumscribe, overlap with, or conflict with one another remains unclear.2

EXTRACTING PRINCIPLES: A CULINARY METAPHOR

The primary goal of this review is to articulate the principles that cut across many of the theories
in relationship science. Consider a culinary metaphor in which each theory is a dish (e.g., a curry)
composed of discrete ingredients (e.g., a grain, a protein, a vegetable, several spices). We set
ourselves the task of extracting the core principles—the basic ingredients—and determining which
principles emerge repeatedly across different theories. Our approach, in other words, involves
temporarily setting the theories aside in order to identify and organize a set of core principles that
characterize relationship science in general. Subsequently, we illustrate how theorists might use
these principles in theory refinement and novel theory development.

In general, the goal of the extraction process is not to replace current theories, nor to generate a
comprehensive list of every theoretical idea ever introduced within the relevant research domain.
Rather, the goal is to identify the key—most widespread and influential—principles that have
influenced theory development and hypothesis generation in the field. This assessment can help
determine whether and how various theories align, perhaps through redundancy or by emphasizing
different features of a phenomenon (akin to the proverbial blind men examining different parts
of an elephant). Additionally, it fills the theoretical pantry with the main ingredients required for
the theory development (cooking) process.

In applying this culinary approach to relationship science, we began by examining psycholog-
ically oriented handbook volumes, textbooks, and review articles to identify the major theories
and models within the research domain and to extract an initial list of core principles. We then
obtained feedback from 16 leading relationship scientists in psychology to refine this initial list,
ultimately producing the 14 core principles discussed below (see Table 1).

Each principle is described at a fairly high level of abstraction so that it can align with mul-
tiple theories; our goal is to capture the general thrust of how each theory characterizes a given
principle, even if there is minor variation across theories in the principle’s precise specification.
Each principle can be used to develop empirical hypotheses, but no one principle specifies how
particular constructs should be operationalized (i.e., there is no gold-standard measure required
by a particular principle). Reflecting the current state of the field, the principles exist at somewhat

2 The evolutionary psychology of human mating (Buss 2008) developed alongside mainstream relationship science. By and
large, however, these two fields have developed in parallel. They address some overlapping topics but tend to employ different
research methods and exhibit modest cross-fertilization of ideas (see Durante et al. 2016, Eastwick 2016). Toward the end
of this review (see Optimizing Relationship Science: Theoretical Cohesion Versus Conflict), we address some ways in which
relationship science could benefit from greater incorporation of ideas from evolutionary psychology.

386 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Table 1 The 14 principles extracted from the psychology literature on relationship science

Seta Number Name Definition

1 Uniqueness Relationship outcomes depend not only on the specific qualities of each partner but also
on the unique patterns that emerge when the partners’ qualities intersect

1
2 Integration Opportunities and motivations for interdependence tend to facilitate cognitive, affective,

motivational, or behavioral merging between partners

3 Trajectory The long-term trajectories of relationship dynamics are affected by each partner’s
continually updated perceptions of the couple’s relationship-relevant interactions and
experiences

4 Evaluation People evaluate their relationships and partners according to a set of positive and negative
constructs, which tend to be moderately negatively correlated

5 Responsiveness Responsive behaviors promote relationship quality for both the self and the partner

2 6 Resolution The manner in which partners communicate about and cope with relationship events
affects long-term relationship quality and stability

7 Maintenance Partners in committed relationships exhibit cognitions and behaviors that promote the
relationship’s persistence over time, even if doing so involves self-deceptive biases

8 Predisposition People bring certain basic qualities of personality and temperament to their relationships,
some of which influence their own and their partners’ relationship wellbeing

3
9 Instrumentality People bring certain goals and needs to their relationships, and the dynamics between the

two partners affect the extent to which they succeed in achieving these goals and meeting
these needs

10 Standards People bring certain standards to their relationships and tend to experience greater
relationship wellbeing when their relationships exceed these standards

11 Diagnosticity Situations vary in the extent to which they afford opportunities to evaluate a partner’s true
goals and motives regarding the relationship

12 Alternatives The presence of attractive alternatives to a current relationship—including the option of
not being in a relationship at all—threatens relationship quality and persistence

4
13 Stress High demands external to the relationship predict worse relationship outcomes, especially

if the demands exceed the two partners’ (individual or combined) resources for coping

14 Culture Relationships are embedded in social networks and a cultural milieu—including norms,
practices, and traditions—that shape the nature and trajectory of those relationships

a Set refers to the four major theoretical questions that the principles address: (1) What is a relationship? (2) How do relationships operate? (3) What
tendencies do people bring to their relationships? (4) How does the context affect relationships?.

different levels of analysis. Some, for example, apply to a person at a single moment in time, whereas
others apply to a person in general or across time; some imply a particular causal process (e.g.,
responsive behaviors increase relationship quality), whereas others specify only that a construct
accounts for variance in a process or outcome (e.g., culture accounts for variance in the quality
of relationship functioning).3 Consistent with the culinary metaphor, most theories incorporate
or address only some of the 14 principles, in the same way that specific dishes do not use all the
ingredients in the pantry; it is unlikely that a cogent theory could incorporate all 14 principles,
especially at this rather early stage of the field’s development.

By necessity, our extraction process involved many subjective judgments. For example, what
counts as a theory? Which theories are most relevant to relationship science? Is a given principle

3 All of the principles we discuss can be disconfirmed, though it may be easier to disconfirm principles that specify a particular
causal process than principles that specify that a construct should predict an outcome in general.

www.annualreviews.org • Relationship Science 387

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

sufficiently core to warrant inclusion? Thus, we make no claim that our conclusions reflect the
absolute truth regarding the key principles that define relationship science from a psychological
perspective; other scholars might make different decisions or draw different conclusions about
the discipline’s core principles. However, the subjectivity of this approach does not render the
conclusions arbitrary. The conclusions are constrained by the theories in our research domain,
which means that all competent extraction efforts ought to generate principles that are reasonably
compatible with one another. We hope that our synthesis starts a dialogue about the core principles
that anchor relationship science and about how these principles might be used both to refine
current theories and to generate new ones.

THE CORE PRINCIPLES OF RELATIONSHIP SCIENCE

Once we extracted the 14 principles (see Table 1), we embedded them within a sensible, albeit
post hoc, organizational structure. We settled on a four-set structure in which each set was built
around a central theoretical question in relationship science: (a) What is a relationship? (b) How
do relationships operate? (c) What tendencies do people bring to their relationships? (d ) How does
the context affect relationships? Figure 1 depicts an organizational framework for conceptualizing
the 14 principles within this four-set structure.

Set 1: What Is a Relationship?

Relationship scientists have written extensively about the definitions of terms such as close and
relationship (e.g., Berscheid & Regan 2005, Kelley et al. 1983). One pervasive concept that char-
acterizes all attempts to define close relationships is that partners are dependent on one another to
obtain good outcomes and facilitate the pursuit of their most important needs and goals (see Finkel

Set 1: What is a relationship?

Set 4: How does the context affect relationships?

Set 3: What tendencies do people
bring to their relationships?

8. Predisposition
9. Instrumentality

10. Standards

1. Uniqueness 2. Integration 3. Trajectory

4. Evaluation
5. Responsiveness

6. Resolution
7. Maintenance

Set 2: How do relationships operate?

13. Stress
14. Culture

11. Diagnosticity
12. Alternatives

Figure 1
A psychological perspective of the core principles in relationship science. Although the image depicts a
heterosexual couple, the available evidence suggests that all 14 core principles generalize to other relationship
structures, including those involving gay men or lesbians (see Footnote 1). Picture credit: Pixabay.

388 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

& Simpson 2015). Beyond these broad definitional efforts, theories have extensively explored the
nature of close relationships. Set 1 contains three core principles that address how and why a re-
lationship becomes more than the sum of its parts (the uniqueness principle), the merging of two
partners into a single psychological entity (the integration principle), and the way relationships
change over time (the trajectory principle).

Uniqueness. Relationship outcomes depend not only on the specific qualities of each partner but also on
the unique patterns that emerge when the partners’ qualities intersect.

According to relationship scientists, a relationship functions as its own entity that is distinct
from and irreducible to the two constituent partners (Berscheid 1999). For example, even if two
individuals tend to be low self-disclosers, their idiosyncratic personal characteristics may mesh in
a unique way that leads both of them to self-disclose a great deal to each other. From a statistical
standpoint, uniqueness effects are evident in the degree of relationship variance in social relations
model studies and in actor × partner interaction effects in actor–partner interdependence model
studies (Kenny & Kashy 2011).

Various theories address the uniqueness principle in distinct ways, with most emphasizing
certain characteristics of relationship partners or specific interpersonal outcomes. Interdependence
theory (Kelley et al. 2003) proposes that the qualities of each partner influence how the two partners
interact in particular situations and, consequently, the outcomes they reap from those interactions.
Transactive goal dynamics theory (Fitzsimons et al. 2015) argues that successful goal attainment
depends on features of the self (e.g., a man’s desire to lose weight) in conjunction with those of
the partner (e.g., his wife’s training as a dietician). Relational regulation theory (Lakey & Orehek
2011) posits that the extent to which social interaction successfully regulates affect, behavior, and
cognition depends on the idiosyncratic traits, preferences, and personal tastes of each partner (e.g.,
the two partners find it soothing if she plays guitar while he cooks).

There are many empirical examples of uniqueness. For example, relationship variance explains
most of the total variance in perceptions of mate value and long-term attraction, indicating that
beauty (as well as other desirable qualities of a mate) really is largely in the eye of the beholder
(Eastwick & Hunt 2014). Mutuality of commitment—the degree to which both partners report
comparable levels of commitment to the relationship—predicts unique variance in relationship
wellbeing above and beyond the two partners’ levels of commitment (Drigotas et al. 1999). Capital-
ization discussions—in which one partner attempts to savor positive news with the other (Gable &
Reis 2010)—tend to be especially difficult and unsatisfying if the person sharing the news is high
in attachment anxiety and the partner is high in attachment avoidance (Shallcross et al. 2011). De-
pressive symptoms during the transition to parenthood are particularly pronounced when highly
neurotic individuals have highly disagreeable spouses (Marshall et al. 2015). In short, relationships
cannot be understood fully by studying main effects involving the two partners; consideration of
the unique dyadic context generated by the two partners is also required.

Integration. Opportunities and motivations for interdependence tend to facilitate cognitive, affective,
motivational, or behavioral merging between partners.

In many close relationships, the psychological boundaries that separate partners are blurry,
making it difficult to discern where one partner ends and the other begins. The self component
in terms such as self-concept and self-regulation takes on a less individualistic focus. Consider
the self-concept, which is often deeply embedded in and possibly altered by close relationships
(Andersen & Chen 2002). As a relationship develops and the desire to maintain it increases, an in-
dividual’s self-concept usually becomes increasingly intertwined with the partner and relationship.
In unstructured relationship thought-listing tasks, for example, higher relationship commitment

www.annualreviews.org • Relationship Science 389

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

predicts greater spontaneous use of plural pronouns such as we, us, and our (Agnew et al. 1998).
Individuals also become confused about whether they or their partner has a given attribute (e.g.,
extraversion), as illustrated by research showing that participants are slower and less accurate
when making me/not me decisions under time pressure if either they or their partner possesses
the relevant attribute than if neither or both of them do (Mashek et al. 2003). Similar effects are
also found in newly formed relationships when an individual desires high interdependence with
his or her potential partner (Slotter & Gardner 2009). Perhaps due to this merging of identities,
the tendency for individuals to exhibit self-enhancing biases generalizes to their close (but not
nonclose) relationship partners (Sedikides et al. 1998).

Self-regulation is also embedded within close relationships. According to transactive goal dy-
namics theory (Finkel et al. 2016, Fitzsimons et al. 2015), relationship partners form a single
self-regulating unit that involves a complex web of goals, pursuits, and outcomes. The optimal
unit of analysis for understanding goal dynamics is the dyad or group, not the individual. For
instance, Alice might set a goal for John, such as losing weight; she might then pursue this goal by
buying healthier snacks, or John might pursue it by forgoing desserts. Alice might also set a goal
for herself, such as submitting a work project on time, which John helps her pursue by doing some
solo parenting so she can complete her project. Depending on how efficiently partners coordinate
their goals and pursuits, goal interdependence can either bolster or undermine each person’s goal
success. When goal coordination is strong, partners can achieve a level of goal success that would
have been impossible if they were single or had a less compatible partner. In fact, research has
shown that individuals assigned to think about ways in which their romantic partner is helpful in
their pursuit of a goal work less hard at pursuing that goal (Fitzsimons & Finkel 2011), which frees
resources for other goal pursuits.

When performing joint tasks, people who desire a communal rather than an exchange relation-
ship tend to behave in ways that obscure rather than accentuate their independent contributions,
which makes overall performance function as a shared dyadic contribution rather than as a com-
bination of two independent contributions. Even at a physiological level, the line separating close
relationship partners is fuzzy (Beckes & Coan 2011, IJzerman et al. 2015, Sbarra & Hazan 2008).
Romantic partners performing laboratory interaction tasks, for example, exhibit increased align-
ment over time in their respiratory sinus arrhythmia (a biomarker of feeling safe), an effect that is
stronger among individuals who are more satisfied with their relationship (Helm et al. 2014).

Trajectory. The long-term trajectories of relationship dynamics are affected by each partner’s continually
updated perceptions of the couple’s relationship-relevant interactions and experiences.

Relationships change over time. Close relationships models from the 1960s and 1970s posited
that change reflected a normative series of stages or filters. For example, the intersection model of
pair relatedness (Levinger & Snoek 1972) proposes that relationship partners move through stages
of escalating interdependence as they become aware of each other, interact, and eventually form a
relationship characterized by a couple-level identity. The relational development model (Knapp
et al. 2014) suggests that couples move through a series of stages both when beginning a relation-
ship (e.g., initiating, then intensifying, then bonding) and when ending a relationship (e.g., differen-
tiating, then stagnating, then terminating). Shifts between stages are often marked by transitions—
turning points where partners’ levels of commitment become explicit (Loving et al. 2009) or life
events that change the relationship, such as the transition to parenthood (Rholes et al. 2011).

Other models focus on how specific relationship constructs ebb and flow across time. Accord-
ing to social penetration theory (Altman & Taylor 1973), relationship partners develop intimacy
as they gradually increase the depth and breadth of their self-disclosures. Attachment theorists
propose that the three behavioral systems associated with pair-bonding develop at different rates,

390 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

with the sexual mating system being particularly important early in a relationship and the at-
tachment and caregiving systems taking on greater importance once the relationship has become
established (Zeifman & Hazan 2008). Interdependence theorists have focused on how the situa-
tions encountered by a given couple produce relationship-specific behavioral tendencies, which
often become reified as injunctive norms (Thibaut & Kelley 1959). For example, if one partner
likes action films and the other likes screwball comedies, the couple might develop a strong turn-
taking norm, which would not exist in couples in which partners had identical film preferences.
People also update their internal working models over time (Bretherton & Munholland 2008).
Research has confirmed that events that produce feelings of greater attachment security lead to
changes in partner-specific attachment models (e.g., expectations that one’s current partner is
reliable), which subsequently change global attachment models (e.g., expectations that partners
in general are reliable) (Fraley 2007, Pierce & Lydon 2001).

Many of the events and experiences that cause relationships to change occur in simple, or-
dinary interactions and commonplace situations. Indeed, relational regulation theory (Lakey &
Orehek 2011) suggests that people’s perceptions of social support originate mainly from everyday
conversations and shared activities with partners rather than in response to major life stressors.
Consequently, relationship outcomes are challenging to predict before a relationship begins, even
if one has considerable knowledge of each partner’s personal characteristics (Finkel et al. 2012).
The ReCAST model (Eastwick et al. 2016) posits that relationships that turn out to be long-term,
committed relationships are often indistinguishable from those that turn out to be short-term and
casual in the early stages as two people get to know each other. Long-term and short-term rela-
tionships prove difficult to differentiate primarily because people do not know if they want to be in
a committed relationship with a specific person until they can fully gauge the relationship’s emo-
tional and sexual chemistry. However, relationships do not remain unpredictable forever: Once
relationship partners progress to advanced relationship stages (e.g., marriage), latent strengths
and vulnerabilities presage whether partners’ evaluations of relationship quality will remain high
or deteriorate (Lavner et al. 2012).

Set 2: How Do Relationships Operate?

Besides examining existential and temporal features of relationships, relationship scientists also
investigate how individuals think, feel, and behave with regard to their experiences and interactions
with their partners. Set 2 encompasses four core principles, which address how individuals evaluate
their partners and relationships (the evaluation principle); how partners respond to each other’s
needs (the responsiveness principle); how partners react dyadically to conflict and other important
relationship events (the resolution principle); and how partners (typically) manage to sustain their
relationship, despite challenges (the maintenance principle).

Evaluation. People evaluate their relationships and partners according to set of positive and negative
constructs, which tend to be moderately negatively correlated.

People constantly evaluate the world around them, and their relationships and partners are no
exception. Most people make relationship evaluations on separable positivity and negativity dimen-
sions (Gable & Reis 2001); researchers have used this two-dimensional conceptualization to exam-
ine the effects of ambivalence—simultaneous highly positive and highly negative evaluations—on
relationship processes and outcomes (e.g., Uchino et al. 2013).

Typically, however, individuals who evaluate a relationship more positively also evaluate it
less negatively, so most evaluative variables are bipolar and labeled according to their positive
endpoint (e.g., satisfaction, commitment, trust, etc.). Each of these constructs has its own

www.annualreviews.org • Relationship Science 391

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

definition, timecourse, and measure, and various theoretical perspectives have attempted to
explain how and why these constructs are related or distinct.

According to the triangular theory of love (Sternberg 1986), for example, love has three ele-
ments, which can be present or absent to varying degrees: intimacy (warm feelings of connect-
edness), passion (romantic and sexual attraction), and commitment (the decision to maintain the
relationship). Other scholars have focused on the timecourse of passion and intimacy, finding
that passion is a function of the first derivative of intimacy over time: When intimacy is increas-
ing, passion is high; when it is stable (regardless of its level), passion is low (Rubin & Campbell
2012). These and other positive evaluative constructs are conceptually distinct, but they are often
positively correlated, sometimes quite highly. Although six of the major evaluative constructs in
relationship science—commitment, trust, love, passion, intimacy, and satisfaction—are distinct,
they share considerable variance and form a single, broad dimension reflecting overall relationship
quality (Fletcher et al. 2000a,b).

Most and perhaps all of these constructs include not only reflective/consciously accessible
components but also impulsive/automatic components (Murray et al. 2013). Using distinct mea-
surement approaches (e.g., explicit versus implicit priming), both kinds of components can be
assessed; depending on the context, either explicit or implicit measures may account for more
of the variance in predicting a given relationship process or outcome (Banse & Kowalick 2007,
McNulty et al. 2013).

Responsiveness. Responsive behaviors promote relationship quality for both the self and the partner.
Individuals’ assessments of relationship quality are strongly influenced by their interactions

with their partners, including their degree of mutual responsiveness—the extent to which they
are “cognizant of, sensitive to, and behaviorally supportive of ” each other’s core needs and values
(Reis 2007, p. 9; see also Clark & Lemay 2010, Reis & Clark 2013). Studies have confirmed that
partners’ responsive behaviors across a wide range of negative and positive experiences predict
greater personal and relationship wellbeing (Debrot et al. 2012, Gable et al. 2012), above and
beyond the positive effects of other more general forms of support (Otto et al. 2015).

This emphasis on responsiveness to the partner’s core needs and values indicates that one cannot
be responsive by simply learning a set of techniques and applying them in all situations (Finkel
et al. 2014b). Rather, responsiveness requires tailoring one’s actions to the unique needs of one’s
partner in a particular situation. Consider this classic example: Having someone immediately repay
your favor is responsive if you want an exchange relationship with him or her, but it is unresponsive
if you desire a communal relationship (Clark & Mills 1979).

When individuals believe their partner is responsive to their needs, they typically feel good
about themselves and are more willing to place themselves in emotionally vulnerable positions,
which can enhance the quality of their relationship (Murray et al. 2006). People vary, of course, in
how comfortable they are being emotionally vulnerable, an individual difference that influences
not only how responsive they are but also how they are likely to react to responsive behavior from
their partner. When in a support-provision role, insecurely attached individuals tend to provide
less responsive support when their partners are upset (Feeney & Collins 2001, Simpson et al. 1992).
When in a support-recipient role, such individuals report feeling greater insecurity and display
relationship-destructive behaviors, although receiving more responsive support tailored to their
needs buffers them from experiencing these adverse states (Lemay & Dudley 2011, Simpson &
Overall 2014). Responsiveness, in other words, plays a crucial role—frequently in conjunction
with the partners’ individual qualities—in social support contexts in which one partner helps the
other cope with negative experiences or stressors. The extent to which individuals perceive that

392 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

they have high-quality support available predicts greater wellbeing and better health outcomes
(Robles et al. 2014).

Responsiveness is also important in capitalization situations (Gable & Reis 2010) in which
responsive reactions typically involve active, constructive behaviors such as excitement or enthu-
siasm (Feeney & Collins 2015). These reactions often yield positive outcomes, such as increases
in the discloser’s positive mood or self-esteem, which in turn lead the discloser to feel closer to
the responsive partner. Conversely, passive or destructive responses such as apathy or envy signal
a lack of responsiveness and frequently elicit distancing responses from disclosers.

Resolution. The manner in which partners communicate about and cope with relationship events affects
long-term relationship quality and stability.

Certain relational events stand out, reverberating with psychological resonance for one or both
partners. These events may be commonplace, such as the fifteenth fight over chores in a month,
or infrequent, such as the birth of a child. The ways in which these events affect a relationship
often hinge on how both partners behave in response to them (Overall & McNulty 2017).

The range of these resonant events is vast, but because negative relational events have stronger
consequences for relationship wellbeing than do positive events (Gottman 1998), conflictual inter-
action tends to be especially significant. Communication often becomes fraught during conflict,
and relationship satisfaction and stability largely depend on how partners construe and respond
to each other’s behavior. In addition, the effectiveness and pace with which partners recover from
conflict episodes independently predict relationship satisfaction and stability (e.g., Gottman &
Levenson 1999).

The response options to conflict reside within a constructive/destructive × active/passive be-
havioral space (Rusbult et al. 1982; for a similar model, see Overall & McNulty 2017). Responses
within this space have downstream effects on relationship quality; active/constructive responses,
for example, tend to predict higher satisfaction and lower breakup likelihood (Rusbult et al. 1982).
Observational research has documented four conflict-relevant behavioral patterns that forecast
relationship distress and propensity to divorce: globally criticizing your partner’s personality, re-
sponding defensively to your partner’s criticism, conveying the belief that your partner is beneath
you, and refusing to engage with your partner’s concerns (Gottman 1998).

Major relationship problems must sometimes be directly addressed to resolve persistent,
nagging issues that, if left unattended, could further destabilize the relationship. For example,
compared to partners who are more passive or destructive, partners who directly and openly
confront major problems in active, constructive ways experience greater distress during and im-
mediately following conflict discussions, but they and their partners are more likely to resolve
these problems and have happier relationships over time (Overall et al. 2009). It appears that
direct opposition is beneficial when serious problems must be addressed and when partners can
make changes, but it is often harmful when partners do not have the traits or skills to be ade-
quately responsive to one another (Overall & McNulty 2017). Indirect (i.e., passive) cooperative
communication, on the other hand, appears to be harmful when major problems must be resolved
but can be beneficial when (a) problems are minor, (b) things cannot be changed, or (c) one or
both partners are too defensive to resolve the problem effectively.

Forgiveness research has revealed that the ways in which both partners behave and react follow-
ing major interpersonal transgressions alter how resolution unfolds. If, for example, transgressors
make stronger amends and victims forgive them more wholeheartedly following a major rela-
tionship transgression, victims tend to develop greater self-respect and clearer self-views (Luchies
et al. 2010), which should result in better and more constructive interactions later on. Indeed,
the expression of greater forgiveness by the victim predicts more constructive patterns of marital

www.annualreviews.org • Relationship Science 393

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

conflict resolution and greater long-term relationship stability across time (Fincham et al. 2004).
If, however, transgressors fail to make adequate amends or are verbally aggressive, victims who
are highly (versus modestly) forgiving tend to respect themselves less and feel less satisfied in the
marriage (Luchies et al. 2010, McNulty 2008).

Maintenance. Partners in committed relationships exhibit cognitions and behaviors that promote the
relationship’s persistence over time, even if doing so involves self-deceptive biases.

Forgiveness is one of many processes that protect and promote relationships over time; these
processes are collectively called relationship maintenance mechanisms. Many of these mechanisms
involve a transformation process in which partners override their immediate self-interests in favor
of behaviors that are more beneficial to the partner or the relationship (Kelley & Thibaut 1978).
One of the most robust predictors of the tendency to enact relationship maintenance mechanisms
is relationship commitment, which emerges from feelings of satisfaction and investment in the
relationship and from the belief that the alternatives to involvement in the relationship are less
desirable (Le & Agnew 2003). Greater relationship commitment, in turn, is associated with en-
acting relationship-maintaining cognitions and behaviors, such as perceiving one’s relationship
as better than others’ (including in the sexual arena; see de Jong & Reis 2015), ignoring or men-
tally derogating romantic alternatives, making sacrifices to benefit the relationship, and forgiving
partner transgressions (Rusbult et al. 2001).

Some of the relationship-promoting effects of commitment stem from motivated biases. For
example, the positive association of commitment with perceptions that one’s relationship is better
than others’ relationships is stronger when one’s relationship is threatened (Rusbult et al. 2000).
The negative association of commitment with the assessment of romantic alternatives as desirable
is stronger when alternatives are objectively more appealing; in fact, the negative association
disappears when alternatives are objectively unappealing, presumably because they do not threaten
the relationship and do not require derogation ( Johnson & Rusbult 1989, Simpson et al. 1990).
Similarly, although romantically unattached men tend to find a novel woman more desirable when
she is at the most fertile stage of her ovulatory cycle, men who are involved in a committed romantic
relationship show the opposite pattern (Miller & Maner 2010). That is, men in relationships
actually find a woman other than their partner less attractive when she is highly fertile, presumably
because she is especially tempting and threatening to their existing relationship and they are
therefore motivated to perceive her negatively.

When engaged in relationship maintenance activities, one partner’s commitment becomes
more closely tied to the other’s trust (Wieselquist et al. 1999). It is, after all, fairly easy to trust
someone who forgives your transgressions and derogates attractive alternatives. Trust, in turn,
is associated with less monitoring of the partner’s behavior as the trusting individual develops
more faith that the partner has his or her best interests at heart (Holmes & Rempel 1989). Indeed,
individuals who place greater trust in their partners exhibit relationship-promoting biases in which
they misremember their partner’s relationship transgressions as being more benign than they
actually were (Luchies et al. 2013).

Relationship scientists have also drawn from other theoretical frameworks to identify rela-
tionship maintenance mechanisms. For example, research has shown that positive illusions about
the partner predict salutary relationship outcomes over time (Murray et al. 2011), and making
more generous attributions about the causes of a partner’s behavior predicts higher relationship
satisfaction (Bradbury & Fincham 1990). Of course, reality does act as a constraint on an individ-
ual’s rose-colored glasses (Fletcher & Kerr 2010, West & Kenny 2011), and biases may be more
or less pronounced depending on the specific features of a situation. For example, the strength of
a person’s positive biases is more pronounced when he or she is pursuing important relationship

394 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

goals rather than deliberating about which goals to pursue (Gagné & Lydon 2004). Relationship-
promoting effects are also found when partners engage in novel and arousing (rather than merely
pleasant) activities with each other (Aron et al. 2000) and when they adopt the perspective of a
neutral, benevolent third party when thinking about relationship conflict (Finkel et al. 2013).

Set 3: What Tendencies Do People Bring to Their Relationships?

Thus far, our discussion has focused predominantly on relationship functioning. We have largely
sidestepped both the normative or idiosyncratic tendencies that individuals bring to their relation-
ships (Set 3) and the contextual factors that might influence relationship processes (Set 4). Set 3
contains three core principles that address how and why relationship functioning is influenced
by the partners’ personality qualities (the predisposition principle), their needs and goals (the in-
strumentality principle), and the benchmarks they use to evaluate the relationship (the standards
principle).

Predisposition. People bring certain basic qualities of personality and temperament to their relationships,
some of which influence their own and their partners’ relationship wellbeing.

The most basic tendencies that people bring to their relationships are tied to their personality
and temperament. The effects of personal strengths (e.g., high self-esteem, attachment security,
approach goals) or vulnerabilities (e.g., neuroticism, rejection sensitivity, avoidance goals) can be
amplified by events that transpire within relationships or in the wider environment. For exam-
ple, John, who has low self-esteem and adopts avoidance goals in his relationship (e.g., avoiding
conflict), may not worry much about the status of his relationship with Alice if they are getting
along well and everything is fine at work. However, when either the relationship or work generates
stress, his personal vulnerabilities may rise to the fore and make him think, feel, and behave in
relationship-damaging ways, which adversely affect Alice and their later interactions (Gable &
Impett 2012, Murray et al. 2006).

Several relationship theories are relevant to the predisposition principle. Attachment theory
(Bowlby 1973), for instance, proposes that the ways in which an individual is treated by significant
others (attachment figures) across the course of his or her life—and especially during childhood—
produce internal working models of the self and others, which then guide how he or she thinks,
feels, and behaves in later interpersonal contexts, particularly stressful ones. Securely attached
individuals, who have received nurturing and sensitive care, develop positive models of the self and
others and, therefore, behave more positively and constructively toward their partners (Mikulincer
& Shaver 2007), especially when one or both of them are upset (Collins & Feeney 2004, Simpson
et al. 1992). Anxiously attached individuals, who have received unpredictable or inconsistent care,
develop negative models of the self (viewing themselves as unworthy of love), which motivates
them to be hypervigilant to signs that their partner might be pulling away. Avoidantly attached
individuals, who have been rebuffed or rejected, develop negative models of others (viewing others
as uncaring), which motivates them to keep their attachment systems deactivated by being self-
reliant, especially in stressful situations (Simpson & Rholes 2012).

The predisposition principle is prominent in other theories and bodies of research, as well.
For example, according to evolutionary models of social development (reviewed in Simpson &
Belsky 2016), stressful circumstances (e.g., early unpredictable environments) result in poorer
parenting, which creates enduring vulnerabilities (e.g., attachment insecurity) that eventually affect
the quality and stability of an individual’s romantic relationships years later (Szepsenwol et al.
2016). According to the communal/exchange model (Clark & Mills 2011), people who bring
greater communal strength to a relationship typically provide benefits to their partners that are

www.annualreviews.org • Relationship Science 395

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

costly to themselves. According to the intimacy process model (Reis & Shaver 1988), the degree to
which an individual discloses important personal information to his or her partner—and how the
partner then perceives this information and reacts to it—is shaped by the unique motives, needs,
goals, and fears (the working models) that each partner brings to the relationship (Laurenceau et al.
1998). Certain personality traits—especially neuroticism, which develops early in life—predict a
host of negative relationship outcomes later in life (McNulty 2013).

Instrumentality. People bring certain goals and needs to their relationships, and the dynamics between
the two partners affect the extent to which they succeed in achieving these goals and meeting these needs.

Beyond personality differences, people also bring many needs and goals to relationships. Some
of these motivational elements are species typical. Attachment theory, for example, contends that
humans have an innate need to develop attachment bonds (Bowlby 1969), whereas self-expansion
theory suggests that humans have an innate need to expand the self (Aron et al. 2013). Applications
of self-determination theory in the domain of relationships indicate that people look to their sig-
nificant others to help them achieve their innate psychological needs for autonomy, competence,
and relatedness (La Guardia et al. 2000). Other goals, such as the desire to reduce carbohydrate
consumption, are more idiosyncratic. Relationship scientists investigate the ways in which
relationships influence how much individuals are able to fulfill these needs and achieve their goals.

One foundational need relevant to the formation and maintenance of close relationships is
attachment, the need to establish an emotionally close relationship that fosters feelings of secu-
rity (Baumeister & Leary 1995, Bowlby 1969). Attachment theory’s central idea is that human
adults (in contrast to adults of our closest genetic relatives, chimpanzees and bonobos) evolved to
form deep, long-term emotional attachments with other adults, presumably because such bonds
promoted survival of our species’ altricial infants ancestrally (Eastwick 2009, Finkel & Eastwick
2015, Fletcher et al. 2015). Human adults also seek out their primary attachment figures (e.g.,
romantic partners) for subsidiary attachment-related needs, such as comfort when they are upset
or strength when pursuing challenging goals (Feeney & Collins 2015). Brain imaging research
reveals that people subjected to physical pain exhibit stronger reductions in the activation of neural
systems supporting emotional and behavioral threat responses when they are randomly assigned
to hold their spouse’s hand, especially if they have a higher-quality marriage (Coan et al. 2006).
Other studies demonstrate that merely viewing a photo of one’s romantic partner when enduring
physical pain activates brain regions linked to safety signaling, especially among those who believe
their partner is highly supportive (Eisenberger et al. 2011).

Close others influence an individual’s goal pursuit processes in diverse ways. For example, when
the opportunity arises, people tend to outsource their goal-related activities to their significant
others, which may reduce the effort they exert when pursuing their goals (Fitzsimons & Finkel
2011), and they draw closer to those who help them achieve their high-priority goals (Fitzsimons &
Shah 2008). As noted in the section Integration, above, the degree to which goal interdependence
bolsters or undermines an individual’s goal success is partially determined by how effectively
partners can coordinate, such as by pooling and efficiently allocating their goal-relevant resources
across the many goals that both partners possess (Fitzsimons et al. 2015). When things go well,
partners not only achieve better goal-related outcomes on a daily basis but also move toward their
ideal selves across time (Rusbult et al. 2009).

Close others also play a major role in existential outcomes. Perhaps the most remarkable
evidence of this is that both marital status (married versus single) and marital quality (higher
versus lower) predict lower morbidity and mortality rates (Holt-Lunstad et al. 2010, Robles et al.
2014).

396 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Standards. People bring certain standards to their relationships and tend to experience greater relationship
wellbeing when their relationships exceed these standards.

A third tendency that people bring to relationships is their personal standards, a construct that
assumes a prominent role in many relationship theories. Thibaut & Kelley (1959) express this
idea in their concept of comparison level (CL), which refers to individuals’ overall assessments of
the outcomes they believe they deserve in a particular relationship. According to interdependence
theory (Kelley & Thibaut 1978, Thibaut & Kelley 1959) and its offshoots (Rusbult 1983), people
are more satisfied with a relationship when the outcomes (rewards minus costs) it provides exceed
their CL.

Many domain-specific relationship theories also focus on standards or on similar concepts, such
as expectations, ideals, or preferences. For example, the triangular theory of love posits that greater
relationship quality is indexed by smaller discrepancies between an individual’s ideal level of each
component of love (intimacy, passion, commitment) and the actual amount of love that he or
she experiences in each component (Sternberg 1986). The ideal standards model (Simpson et al.
2001) claims that individuals should experience higher relationship quality when they perceive
greater alignment between their ideals for particular traits in a romantic partner (warmth/loyalty,
vitality/attractiveness, and status/resources) and their partner’s actual traits. The suffocation model
(Finkel et al. 2014a,b) suggests that people have varied historically in the degree to which they
expect their spouses to fulfill needs that are low (e.g., safety) versus high (e.g., self-actualization)
in Maslow’s hierarchy of needs; the extent to which their relationships meet these expectations is
theorized to predict marital quality more strongly for higher-level than lower-level needs.

High standards bode poorly for relationship wellbeing when they are unattainable (McNulty
2016a,b); people tend to be less happy when their actual partners and relationships do not fulfill
their lofty standards. Conversely, high standards bode well for relationship wellbeing when such
standards motivate individuals to engage in behaviors that improve relationship outcomes, such
as when molding a less-than-ideal partner into an ideal one (Murray et al. 1996). Although high
standards often motivate prorelationship cognitions and behaviors in people who have strong
relationship skills, they also produce disappointment in those with poor relationship skills
(McNulty & Karney 2004).

People sometimes deviate from strict veridicality when comparing their standards with reality.
For example, people who are in a relationship characterized by aggression but nonetheless remain
committed to the relationship adopt more tolerant standards for partner aggression (Arriaga et al.
2016). In addition, an individual’s ideal partner preferences change over time to match the desirable
qualities possessed by his or her current partner (Fletcher et al. 2000a,b; Neff & Karney 2003).
People also have difficulty comparing, on a trait-by-trait basis, the concrete features of a flesh-
and-blood partner with their abstract standards. Consequently, the match between ideals and a
partner’s traits is typically irrelevant to relationship outcomes if ideals are measured on single traits
(e.g., attractiveness) isolated from the partner’s complete suite of traits (Eastwick et al. 2014a,b).

Set 4: How Does the Context Affect Relationships?

Consistent with classic person × situation models (e.g., Lewin 1936), the close relationships liter-
ature complements its analysis of the tendencies that people bring to their relationships with an
analysis of the situational and contextual factors that influence relationship processes and outcomes
(McNulty 2016a,b). Set 4 includes four core principles, which range from the micro to the macro
level of analysis. These principles address how and why partners navigate situations in which their
interests diverge (the diagnosticity principle), how they respond to appealing alternatives (the

www.annualreviews.org • Relationship Science 397

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

alternatives principle), how stressors affect relationship dynamics (the stress principle), and how
the broader social network and culture influence relationship dynamics (the culture principle).

Diagnosticity. Situations vary in the extent to which they afford opportunities to evaluate a partner’s
true goals and motives regarding the relationship.

Some situations provide better opportunities than others for revealing a partner’s relationship-
relevant goals and motives. For example, an individual’s behavior in noncorrespondent situations
such as strain tests—in which a good outcome for one partner produces a bad outcome for the other
partner—can reveal his or her relationship goals, motives, and orientations more clearly than does
his or her behavior in correspondent situations (Holmes 1981, 2002; Kelley & Thibaut 1978). If
John agrees to quit his dream job and leave his friends and family so Alice can pursue her dream job
in a faraway city, his willingness to make these sacrifices reveals how much he cares about her and is
committed to their relationship. Alice’s ability to make relatively unambiguous attributions about
John’s motives would have been diminished if the situation were more correspondent, as would
be the case if he disliked his current job and did not have close social ties where they were living.
Because highly noncorrespondent situations allow individuals to demonstrate their willingness to
make significant sacrifices for their partner and relationship, such prorelationship behavior from
one partner tends to promote the other partner’s trust (Shallcross & Simpson 2012, Wieselquist
et al. 1999), among other relationship benefits (Simpson 2007).

Diagnostic situations are central to several theories in relationship science. According to inter-
dependence theory (Holmes 1981, 2002; Rusbult & Van Lange 2003), when relationship partners
find themselves in noncorrespondent situations such as strain tests, the partner being asked to
make a sacrifice must try to set aside his or her personal desires and transform his or her moti-
vation to do what is best for the partner and relationship. He or she must then coordinate plans
and actions with his or her partner to help achieve the partner’s important goals. This explains
why strain tests in particular are such powerfully diagnostic situations: They leave little attribu-
tional ambiguity regarding the extent and nature of the sacrificing partner’s transformation of
motivation. But if partners fail strain tests because they do not engage in prorelationship trans-
formation of motivation, relationships run the risk of becoming unstable (Rusbult et al. 2001).

Diagnosticity is also a key element of both the risk regulation model (Murray et al. 2006) and
the mutual responsiveness model (Murray & Holmes 2009). Individuals are typically motivated
to connect emotionally with their partners while protecting the self from excessive vulnerability.
Compared to correspondent situations, noncorrespondent situations highlight the fundamental
conflict between (a) seeking connection and allowing the self to be vulnerable and (b) protecting
the self and avoiding potential rejection. This is a basic conflict that both partners must struggle
to resolve because relationships cannot fully develop unless the two of them are willing to take
leaps of faith (Murray et al. 2006) and reciprocally disclose intimate information (Reis & Shaver
1988), actions that make them vulnerable to possible exploitation (Cavallo et al. 2009).

Alternatives. The presence of attractive alternatives to a current relationship—including the option of
not being in a relationship at all—threatens relationship quality and persistence.

The existence and extent of options that make it desirable for an individual to leave an existing
relationship play a vital role in interdependence theory (Kelley & Thibaut 1978, Thibaut & Kelley
1959) and the investment model (Rusbult 1983). Specifically, the comparison level for alternatives
(CLAlt) concept reflects the outcomes that individuals would experience in their best alternative to
being in the current relationship, including being single. Interdependence theorists hypothesize
that relationship stability is more closely aligned with CLAlt (the extent to which a person can

398 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

achieve better outcomes in another relationship) than with CL (the extent to which a person’s
current relationship outcomes exceed his or her standards). That is, CLAlt determines the extent
to which a person is dependent on his or her partner to achieve his or her needs, goals, and other
desirable outcomes.

Most CLAlt studies have emphasized individuals’ subjective (rather than objective) perceptions
of the degree to which current alternatives are appealing. Research has shown, for example, that
the better people perceive their alternatives to be, the more likely their relationships are to dissolve
(Le et al. 2010). Research has also tested this association using experimental manipulations: When
participants are randomly assigned to believe that their own sex is in the numerical minority (versus
majority), they report lower relationship quality with their current partners, presumably because
the abundance of opposite-sex people suggests that better options may be available (Kim 2013).

Most relationship models characterize desirable alternatives as threats that individuals should
be motivated to ignore, downplay, or derogate to mitigate negative effects on their relationships
(Durante et al. 2016, Lydon & Karremans 2015). This process is evident in controlled, conscious
responses to attractive alternatives, such as the explicit evaluation of the desirability of opposite-sex
alternative partners ( Johnson & Rusbult 1989, Simpson et al. 1990). However, it is also evident
in automatic, spontaneous responses, such as the amount of time individuals spend looking at
attractive alternatives (Maner et al. 2009, Miller 1997) or displaying affiliative nonverbal behaviors
in response to them (Karremans & Verwijmeren 2008).

Some theoretical analyses linked to these findings adopt an evolutionary perspective. Pair-
bonds most likely evolved in humans because such relationships offer adaptive benefits for offspring
(Eastwick 2009). However, it takes considerable time and energy to cultivate a strong pair-bond,
which suggests that the derogation process may be an evolved adaptation that motivates commit-
ted romantic partners to train their attention on each other to preserve the existing pair-bonded
relationship (Maner et al. 2008). The process of derogating desirable alternatives is also consis-
tent with cognitive dissonance perspectives: Once people have made a difficult-to-reverse choice
(e.g., committing to a partner), they become motivated to perceive nonchosen alternatives as less
desirable (Brehm 1956).

However, in some cases, people can sustain committed relationships while also pursuing sexual,
even loving, relationships with other partners, as in the case of polyamory, a relationship structure
in which individuals have “consensual loving and romantic relationships with more than one
partner” (Conley et al. 2012, p. 126). In fact, polyamorous relationships may be stable precisely
because people do not construe their additional sexual partners as true alternatives that would
replace a current partner.

Stress. High demands external to the relationship predict worse relationship outcomes, especially if the
demands exceed the two partners’ (individual or combined) resources for coping.

Beyond the immediate situation and romantic alternatives, external factors, especially stressors,
can also affect relationship functioning. It is difficult to sustain a high-quality relationship when
confronting acute or chronic stress external to the relationship (Karney & Bradbury 1995). A
broad spectrum of stressors—including job loss, financial strain, incarceration, chronic illness,
infertility, and natural disasters—predicts myriad adverse relationship outcomes, including low
satisfaction and breakup (Karney & Neff 2015, Randall & Bodenmann 2009).

Research has shown that some couples manage stress better than others. One major factor
in explaining this variation is the level of coping-relevant resources (Hill 1949, McCubbin &
Patterson 1983). According to stress buffering perspectives (Cohen & Wills 1985), the adverse
effects of stressors are especially strong if such resources are low, as is the case when the couple lacks

www.annualreviews.org • Relationship Science 399

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

sufficient money or when one or both partners feel psychologically depleted due to work-related
stress. However, among couples who have good problem-solving skills, navigating moderate levels
of stress early on strengthens the relationship over time, as long as they responded to those stressors
effectively (Neff & Broady 2011).

The most influential framework for conceptualizing the impact of stress on relationship func-
tioning is Karney & Bradbury’s (1995) vulnerability-stress-adaptation model, which emphasizes
the role of interpersonal processes in mediating the effects of stress and resources on relationship
outcomes. According to this model, stress exerts its adverse influence on relationship processes
and outcomes via two routes (Karney & Neff 2015). First, stress alters how much time partners
have for each other and how they use that time. Partners who encounter high levels of stress
have less time to engage in tasks that might increase emotional or physical intimacy, and they use
more of their winnowed time dealing with stressful, challenging situations (Neff & Karney 2009).
Second, stress depletes the self-regulatory resources that partners need to respond constructively
to relationship challenges (Repetti 1989). Partners whose self-regulatory resources have been de-
pleted are especially prone to retaliation in response to provocation (Finkel et al. 2009), and the
subjective experience of self-regulatory depletion mediates the association of stress with both neg-
ative marital behaviors and diminished marital satisfaction (Buck & Neff 2012). These effects are
particularly strong when individuals are tempted to lash out at their partners but are smaller or
nonexistent in the absence of such temptation (Finkel et al. 2012).

Culture. Relationships are embedded in social networks and a cultural milieu—including norms, practices,
and traditions—that shape the nature and trajectory of those relationships.

As we broaden the contextual lens to consider cultural and subcultural effects on relationships,
we turn to social ecological models, which posit that environmental contexts have nested layers
(e.g., Bronfenbrenner 1986). More specifically, individuals live within social networks of friends
and family whose approval or disapproval of a given relationship might affect its trajectory. These
social networks are themselves embedded in cultural contexts consisting of norms, values, and
scripts, and relationships are also shaped by these socially shared constructs. Finally, cultures are
embedded in national and historical contexts that can cause relationships to differ across time and
place.

At the level of the social network, approval from friends and family members predicts greater
relationship satisfaction and stability (Felmlee 2001). In some cases, friends and family members
may engage in specific behaviors that help a relationship flourish or flounder; in other cases, simply
hearing a loved one’s positive reaction about one’s current partner can reduce uncertainty and
increase the likelihood of investing more in the relationship (Sprecher 2011).

People also share knowledge about sexual scripts and norms within the local culture, which sub-
sequently guide behavior (Simon & Gagnon 2003). Fraternity membership, for example, predicts
the extent to which sexual activity is part of the script that undergraduates use when describing
a typical date (Bartoli & Clark 2006), and norms about appropriate sexual behaviors vary as a
function of regional levels of education and religiosity (Laumann et al. 1994).

Finally, relationships can be influenced by the national and historical context. For example,
the degree to which individuals are willing to engage in casual sexual activity is linked to national
indicators such as the rate of infectious diseases and women’s economic power (Schaller & Murray
2008, Schmitt et al. 2005). People in the United States expect their marriage to help them fulfill
certain needs, but these needs have varied over time; Americans were especially likely to prioritize
needs like safety and food production circa 1800, needs like intimacy and sexual fulfillment circa
1900, and needs like self-discovery and self-expression circa 2000 (Finkel et al. 2014a). The rise of

400 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

the postindustrial economy in Western cultures during the second half of the twentieth century
facilitated a “grand gender convergence” (although certainly not an equalization) in men’s and
women’s social roles (Goldin 2014), which profoundly influenced relationship dynamics, especially
in marriages (Finkel et al. 2014a).

Evolutionary models of culture connect these different context levels to specific psychological
processes. Transmitted cultural models, for instance, explore how and why people share beliefs,
practices, and knowledge, usually emphasizing the processes of adopting, changing, and improving
these products of shared culture (Richerson & Boyd 2005). As an example, college administrators
frequently hold workshops that increase students’ sensitivity to issues surrounding sexual consent
and that change how the students engage in sexual behavior. Evoked cultural models posit that
encountering a particular environmental cue that was prevalent in our ancestral past, such as the
presence versus absence of a responsive caregiver (Simpson & Belsky 2016), triggers adaptive
cognitive and behavioral responses. For example, environments containing more pathogens may
trigger preferences for romantic partners who carry genes associated with better health (Gangestad
et al. 2006). These two forms of culture may influence psychological functioning in tandem or
independently (Eastwick 2013).

COMBINING THE PRINCIPLES TO REFINE OR DEVELOP THEORIES

Scholars can use these 14 core principles from the psychological literature on relationships (see
Figure 1 and Table 1) to clarify and refine existing theories and perhaps generate new ones.
Using the metaphoric terminology of the culinary approach, the cook (the theorist) can canvas
the pantry (the collection of principles) for particular ingredients (specific principles), prepare the
recipe (select and arrange the ingredients), and then cook the dish (develop the theory).

Refining Existing Theories

To refine an existing theory, theorists might first map each of the 14 principles onto a theory,
retaining the principles that overlap or fit with this theory and setting aside those that do not.
Theorists can then determine whether the addition of one or more of the extra principles—those
that were not part of the original theory—might broaden the explanatory power of the theory
enough to offset the additional complexity that comes with including more principles.

We illustrate this process using transactive goal dynamics theory (Finkel et al. 2016, Fitzsi-
mons et al. 2015). We focus on this theory because we are familiar with it and because it was
developed within the past few years, which means that there are few published articulations of the
core principles. Transactive goal dynamics theory contains elements of integration (Principle 2),
evaluation (Principle 4), responsiveness (Principle 5), predisposition (Principle 8), instrumentality
(Principle 9), and alternatives (Principle 12). Specifically, it proposes that: (a) relationship part-
ners form a shared system of goal pursuits (integration), (b) subjective assessments of relationship
commitment predict increased merging (evaluation), (c) goal success is maximized when partners
support each other in ways tailored to each partner’s idiosyncratic goals and needs (responsive-
ness), (d ) each partner has certain skills and preferences that can be leveraged for optimal goal
functioning at the dyadic level (predisposition), (e) partners influence each other’s degree of goal
success (instrumentality), and ( f ) the relationship is more likely to continue if it results in goal
success that exceeds what the two partners would otherwise experience (alternatives). In short,
we can formulate much of the content of transactive goal dynamics theory with just these six
ingredients.

www.annualreviews.org • Relationship Science 401

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

As with most theories, transactive goal dynamics theory also contains important elements that
do not rise to the level of a core principle. For example, one tenet of the theory is that stronger
goal interdependence in a relationship should predict poorer goal-related recovery following a
breakup. Such idiosyncratic elements are crucial in defining the unique terrain that a given theory
seeks to address.

Transactive goal dynamics theory, however, leaves eight core principles unused. Thus, a scholar
seeking to refine or expand the theory might consider whether adding any additional principles
might benefit or improve the theory enough to offset the complexity of doing so, or whether
incorporating additional principles might generate novel hypotheses. For example, transactive
goal dynamics theory is not a theory of goal content; it primarily takes the two partners’ goals
as a given rather than investigating how or why they adopted these particular goals. By adding
the perspective of culture (Principle 14), a theorist might wonder whether the goal contents that
people bring to their relationships—for example, the desired level or type of interdependence
in the relationship—differ in important ways across cultural or historical contexts and whether
such variation has implications for relationship quality and longevity. This analysis might lead to
the novel hypothesis that emotional responsiveness is more important for such outcomes in the
twenty-first century United States than in Jane Austen’s England (Light & Fitzsimons 2014).

Generating New Theories

Other scholars might want to use the principles not to refine or expand an existing theory but
to guide theory development in a bottom-up manner. Although this process can begin in various
ways (e.g., one might start with observations about relationship dynamics in the surrounding
world), it is likely to entail a systematic consideration of whether each of the principles can inform
thinking, which would lead to the generation of new insights and hypotheses. For example, a
scholar might wish to develop a new theoretical perspective on the circumstances under which
sexual intercourse draws partners closer together versus pushes them apart. Merely looking at the
list of core principles will not yield a new theoretical perspective, but it might be a productive
first step. A scholar can consider whether each principle is likely to yield a deeper, better, or
more nuanced understanding of a topic and can then explore how the most relevant principles
interrelate in theoretically interesting ways. To facilitate this process, he or she might generate
a path diagram that specifies precisely how the variables should interrelate, including processes
such as mediation, moderation, and feedback loops. Because the principles are cast at a relatively
high level of abstraction, they can be exported readily to different research domains. For example,
the researcher might find it easier to apply two or three principles—rather than an entire theory
of relationships—to an existing evolutionary perspective on how sex affects relationship partners.

OPTIMIZING RELATIONSHIP SCIENCE: THEORETICAL COHESION
VERSUS CONFLICT

What has this exercise taught us? Among other things, we have learned that there are few instances
in which a notable principle used in one theory clearly conflicts with a notable principle used in
another.4 Many relationship scientists recall interdependence theorists whispering the objection

4 One reason for this cohesion may be the abstract nature of the 14 principles. When different theories operationalize, test, and
combine constructs associated with specific principles in novel ways, they may generate different or competing predictions. For
example, although scholars agree that the principle of standards (Principle 10) matters, there is debate about the circumstances
under which standards influence relationship outcomes (Eastwick et al. 2014a,b; Schmitt 2014). There could also be conflict

402 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

that attachment theory is too focused on individual differences. However, as attachment theory
complemented research on individual differences with research on normative attachment processes
(Mikulincer & Shaver 2007), the whispers dissipated. Many currently prominent relationship
models derive specific hypotheses on the basis of ideas borrowed from different theories. When
developing the risk regulation model, for example, Murray and colleagues (2006) extracted several
key features of interdependence theory and attachment theory, combined them in novel ways,
and added new theoretical components to generate an important process model that tied together
several major ideas in the field.

In many ways, such strong theoretical cohesion is marvelous. Although relationship scientists
have many theories and models that address distinct relationship processes, we appear to have
something approximating a consensual theoretical paradigm. This paradigm, which encompasses
the 14 principles reviewed above, is compelling and generative. Recent edited volumes (e.g.,
Simpson & Campbell 2013) and journal special issues (e.g., Finkel & Simpson 2015) indicate that
relationship science is thriving.

However, there are also downsides to having such a cohesive discipline. Science often benefits
from competition between conflicting ideas. Although it is pleasant to work in an environment
characterized by consensus, it sometimes takes friction to generate forward motion. We believe
that the current theoretical paradigms in relationship science are excellent, but the field might
benefit from some theoretical conflicts—alternative accounts that might sharpen and hone one
another. For example, our field could explore whether the dominant view that people are best
served by being in a secure relationship with a romantic partner is misguided, at least under
some circumstances, such as when close friends provide a better option (DePaulo & Morris
2005). We could also reexamine the widespread, albeit implicit, assumption that relationship
stability is a good outcome (with abusive relationships being one exception) and breakups are a
bad outcome. Perhaps we could challenge this dominant view by examining the circumstances
under which people are best served by leaving their relationship or seeking to trade up for a
partner who is more compatible. Revisiting broad questions and assumptions such as these ac-
centuates the fact that many relationship scientists have focused quite heavily on the life cycle of
one relationship rather than the multiple relationships that many people develop throughout their
lives.

Evolutionary psychology, for example, potentially poses some serious challenges to certain
theories and models in relationship science, particularly in the realm of mate selection. The evo-
lutionary psychology of human mating adopts foundational assumptions that differ from many
of those in relationship science (Durante et al. 2016, Eastwick 2016). In particular, evolution-
ary perspectives highlight not only the adaptive value of strong pair-bonds but also the potential
adaptive value of behaviors such as sexual infidelity, trading up, and stalking (Buss & Shackelford
1997). If a scholar extracted the core principles in the evolutionary psychological literature on
mating, one of these principles might be that people evolved to seek opportunistic copulations
outside of long-term, committed relationships. The hypotheses that follow from this principle
seem to fundamentally conflict with hypotheses that follow from the maintenance principle dis-
cussed above (Durante et al. 2016). If ancestral humans enjoyed a survival advantage from such
relationship-destructive behaviors, how can relationship scientists reconcile this with the field’s
strong emphasis on the benefits—including the survival benefits—of exclusive romantic relation-
ships (Holt-Lunstad et al. 2010, Robles et al. 2014)? And if humans are best served by having

at the level of abstraction of the 14 principles, but our extensive literature review unearthed minimal evidence of any such
conflict within relationship science.

www.annualreviews.org • Relationship Science 403

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

accurate insights about their partner’s romantic attraction to others so that they can guard against
mate poaching, why do they shield themselves from the truth precisely in those circumstances
when the threat of one’s partner’s extrarelationship temptation is strongest ( Johnson & Rusbult
1989, Simpson et al. 1995)?

By posing these and other questions, evolutionary psychology directly challenges some of
the foundational assumptions and principles in relationship science. If those assumptions and
principles withstand the challenge, the current relationship science paradigm will be solidi-
fied. If they do not, it will need to be altered. Regardless of the outcome, our discipline will
benefit.

CONCLUSION

Relationship science has come a long way in a relatively short period of time. It has become a rich
discipline characterized by strong theories and highly generative research paradigms. According to
our analysis, the field has 14 core principles that address what a relationship is, how relationships
operate, what tendencies people bring to their relationships, and how contextual factors affect
relationship processes and outcomes. At present, the major theories in our field largely align and
rarely conflict.

As we look to the future, it will be interesting to see whether various theories gradually merge
into a single, unified theory of relationships or whether some major disagreements will enter
mainstream relationship science. As the field continues to mature, it is likely to benefit from
simultaneous trends toward greater theoretical unification on the one hand and greater theoretical
disagreement on the other. Such trends should refine, deepen, and extend our understanding of
how and why relationships function as they do in daily life, potentially providing clinicians and
policymakers with more effective tools for helping people achieve deeper and more fulfilling
relationships.

SUMMARY POINTS

1. This review presents the first attempt to discern the core principles that cut across the
major theories in relationship science, especially the theories within psychology.

2. This review of the major theories used a novel procedure called the culinary approach,
which seeks to extract the core principles (the basic theoretical building blocks or ingre-
dients) from a given discipline and address how theorists can use them to refine existing
theories or develop new theories.

3. Applying the extraction process to relationship science revealed 14 core principles, which
help to answer four basic questions in the literature: (a) What is a relationship? (b) How
do relationships operate? (c) What tendencies do people bring to their relationships?
(d ) How does the context affect relationships?

4. The literature review revealed a cohesive discipline with few notable conflicts among the
core theoretical principles.

5. We suggest that relationship science would benefit from both (a) greater recognition of
the principle-level overlap or redundancy across theories and (b) greater effort to adopt
novel perspectives on relationship dynamics, ideally perspectives that raise important
challenges to the dominant paradigm.

404 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

DISCLOSURE STATEMENT

The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

The authors, who contributed equally to this review, thank Galen Bodenhausen, Jim McNulty,
and Nickola Overall for their insightful feedback on a previous draft, as well as the 16 leading
relationship scientists who provided constructive feedback on an early outline of this paper.

LITERATURE CITED

Agnew CR, Van Lange PA, Rusbult CE, Langston CA. 1998. Cognitive interdependence: commitment and
the mental representation of close relationships. J. Personal. Soc. Psychol. 74:939–54

Altman I, Taylor DA. 1973. Social Penetration: The Development of Interpersonal Relationships. New York: Holt,
Rinehart and Winston

Andersen SM, Chen S. 2002. The relational self: an interpersonal social-cognitive theory. Psychol. Rev. 109:619–
45

Aron A, Lewandowski GW Jr., Mashek D, Aron EN. 2013. The self-expansion model of motivation and
cognition in close relationships. See Simpson & Campbell 2013, pp. 90–115

Aron A, Norman CC, Aron EN, McKenna C, Heyman RE. 2000. Couples’ shared participation in novel and
arousing activities and experienced relationship quality. J. Personal. Soc. Psychol. 78:273–84

Arriaga XB, Capezza NM, Daly CA. 2016. Personal standards for judging aggression by a relationship partner:
How much aggression is too much? J. Personal. Soc. Psychol. 110:36–54

Banse R, Kowalick C. 2007. Implicit attitudes towards romantic partners predict well-being in stressful life
conditions: evidence from the antenatal maternity ward. Int. J. Psychol. 42:149–57

Bartoli AM, Clark MD. 2006. The dating game: similarities and differences in dating scripts among college
students. Sex. Cult. 10:54–80

Baumeister RF, Leary MR. 1995. The need to belong: desire for interpersonal attachments as a fundamental
human motivation. Psychol. Bull. 117:497–529

Beckes L, Coan JA. 2011. Social baseline theory: the role of social proximity in emotion and economy of
action. Soc. Personal. Psychol. Compass 5:976–88

Berscheid E. 1999. The greening of relationship science. Am. Psychol. 54:260–66
Berscheid E, Regan P. 2005. The Psychology of Interpersonal Relationships. New York: Prentice-Hall
Bowlby J. 1969. Attachment and Loss, Vol. 1: Attachment. New York: Basic Books
Bowlby J. 1973. Attachment and Loss, Vol. 2: Separation: Anxiety and Anger. New York: Basic Books
Bowlby J. 1980. Attachment and Loss, Vol. 3: Loss. New York: Basic Books
Bradbury TN, Fincham FD. 1990. Attributions in marriage. Review and critique. Psychol. Bull. 107:3–

33
Brehm JW. 1956. Postdecision changes in the desirability of alternatives. J. Abnorm. Soc. Psychol. 52:384–

89
Bretherton I, Munholland KA. 2008. Internal working models in attachment relationships: elaborating a

central construct in attachment theory. In The Handbook of Attachment: Theory, Research, and Clinical
Applications, ed. J Cassidy, PR Shaver, pp. 102–27. New York: Guilford

Bronfenbrenner U. 1986. Ecology of the family as a context for human development: research perspectives.
Dev. Psychol. 22:723–42

Buck AA, Neff LA. 2012. Stress spillover in early marriage: the role of self-regulatory depletion. J. Fam.
Psychol. 26:698–708

Buss DM. 2008. The Evolution of Desire: Strategies of Human Mating. New York: Basic Books. 2nd ed.
Buss DM, Shackelford TK. 1997. From vigilance to violence: mate retention tactics in married couples.

J. Personal. Soc. Psychol. 72:346–61

www.annualreviews.org • Relationship Science 405

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Campbell L, Simpson A. 2013. The blossoming of relationship science. See Simpson & Campbell 2013,
pp. 3–10

Cavallo J, Fitzsimons GM, Holmes JG. 2009. Taking chances in the face of threat: romantic risk regulation
and approach motivation. Personal. Soc. Psychol. Bull. 35:737–51

Clark MS, Lemay EP. 2010. Close relationships. In Handbook of Social Psychology, ed. ST Fiske, DT Gilbert,
G Lindzey, pp. 898–940. New York: Wiley

Clark MS, Mills J. 1979. Interpersonal attraction in exchange and communal relationships. J. Personal. Soc.
Psychol. 37:12–24

Clark MS, Mills JR. 2011. A theory of communal (and exchange) relationships. Handb. Theor. Soc. Psychol.
1:232–50

Clark MS, Reis HT. 1988. Interpersonal processes in close relationships. Annu. Rev. Psychol. 39:609–72
Coan JA, Schaefer HS, Davidson RJ. 2006. Lending a hand: social regulation of the neural response to threat.

Psychol. Sci. 17:1032–39
Cohen S, Wills TA. 1985. Stress, social support, and the buffering hypothesis. Psychol. Bull. 98:310–57
Collins NL, Feeney BC. 2004. Working models of attachment affect perceptions of social support: evidence

from experimental and observational studies. J. Personal. Soc. Psychol. 87:363–83
Conley TD, Ziegler A, Moors AC, Matsick JL, Valentine B. 2012. A critical examination of popular assump-

tions about the benefits and outcomes of monogamous relationships. Personal. Soc. Psychol. Rev. 17:124–41
de Jong DC, Reis HT. 2015. We do it best: commitment and positive construals of sex. J. Soc. Clin. Psychol.

34:181–202
Debrot A, Cook WL, Perrez M, Horn AB. 2012. Deeds matter: daily enacted responsiveness and intimacy in

couples’ daily lives. J. Fam. Psychol. 26:617–27
DePaulo BM, Morris WL. 2005. Singles in society and in science. Psychol. Inq. 16:57–83
Drigotas SM, Rusbult CE, Verette J. 1999. Level of commitment, mutuality of commitment, and couple

well-being. Pers. Relat. 6:389–409
Durante KM, Eastwick PW, Finkel EJ, Gangestad SM, Simpson JA. 2016. Pair-bonded relationships and

romantic alternatives: toward an integration of evolutionary and relationship science perspectives. Adv.
Exp. Soc. Psychol. 53:1–74

Eastwick PW. 2009. Beyond the Pleistocene: using phylogeny and constraint to inform the evolutionary
psychology of human mating. Psychol. Bull. 135:794–821

Eastwick PW. 2013. Cultural influences on attraction. See Simpson & Campbell 2013, pp. 161–82
Eastwick PW. 2016. The emerging integration of close relationships research and evolutionary psychology.

Curr. Dir. Psychol. Sci. 25:183–90
Eastwick PW, Hunt LL. 2014. Relational mate value: consensus and uniqueness in romantic evaluations.

J. Personal. Soc. Psychol. 106:726–51
Eastwick PW, Keneski E, Morgan TA, McDonald M. 2016. What do short-term and long-term relationships

look like? Building the Relationship Coordination and Strategic Timing (ReCAST) model. Work. Pap., Dep.
Psychol., Univ. Calif. Davis. http://papers.ssrn.com/sol3/papers.cfm?abstract_id = 2820704

Eastwick PW, Luchies LB, Finkel EJ, Hunt LL. 2014a. The many voices of Darwin’s descendants: reply to
Schmitt 2014. Psychol. Bull. 140:673–81

Eastwick PW, Luchies LB, Finkel EJ, Hunt LL. 2014b. The predictive validity of ideal partner preferences:
a review and meta-analysis. Psychol. Bull. 140:623–65

Eisenberger NI, Master SL, Inagaki TK, Taylor SE, Shirinyan D, et al. 2011. Attachment figures activate a
safety signal-related neural region and reduce pain experience. PNAS 108:11721–26

Feeney BC, Collins NL. 2001. Predictors of caregiving in adult intimate relationships: an attachment theo-
retical perspective. J. Personal. Soc. Psychol. 80:972–94

Feeney BC, Collins NL. 2015. A new look at social support: a theoretical perspective on thriving through
relationships. Personal. Soc. Psychol. Rev. 19:113–47

Felmlee DH. 2001. No couple is an island: a social network perspective on dyadic stability. Soc. Forces 79:1259–
87

Fincham FD, Beach SRH, Davila J. 2004. Forgiveness and conflict resolution in marriage. J. Fam. Psychol.
18:72–81

406 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Finkel EJ, DeWall CN, Slotter EB, McNulty JK, Pond RS Jr., Atkins DC. 2012. Using I3 theory to clarify
when dispositional aggressiveness predicts intimate partner violence perpetration. J. Personal. Soc. Psychol.
102:533–49

Finkel EJ, DeWall CN, Slotter EB, Oaten M, Foshee VA. 2009. Self-regulatory failure and intimate partner
violence perpetration. J. Personal. Soc. Psychol. 97:483–99

Finkel EJ, Eastwick PW. 2015. Attachment and pairbonding. Curr. Opin. Behav. Sci. 3:7–11
Finkel EJ, Eastwick PW, Karney BR, Reis HT, Sprecher S. 2012. Online dating: a critical analysis from the

perspective of psychological science. Psychol. Sci. Public Interest 13:3–66
Finkel EJ, Fitzsimons GM, vanDellen MR. 2016. Self-regulation as a transactive process: reconceptualizing

the unit of analysis for goal setting, pursuit, and outcomes. In Handbook of Self-Regulation, ed. KD Vohs,
RF Baumeister, pp. 264–82. New York: Guilford. 3rd ed.

Finkel EJ, Hui CM, Carswell KL, Larson GM. 2014a. The suffocation of marriage: climbing Mount Maslow
without enough oxygen. Psychol. Inq. 25:1–41

Finkel EJ, Larson GM, Carswell KL, Hui CM. 2014b. Marriage at the summit: response to the commentaries.
Psychol. Inq. 25:120–45

Finkel EJ, Simpson JA, eds. 2015. Relationship science. Curr. Opin. Psychol. 1(Spec. Issue). Amsterdam: Elsevier
Finkel EJ, Slotter EB, Luchies LB, Walton GM, Gross JJ. 2013. A brief intervention to promote conflict-

reappraisal preserves marital quality over time. Psychol. Sci. 24:1595–601
Fitzsimons GM, Finkel EJ. 2011. Outsourcing self-regulation. Psychol. Sci. 22:369–75
Fitzsimons GM, Finkel EJ, vanDellen MR. 2015. Transactive goal dynamics. Psychol. Rev. 122:648–73
Fitzsimons GM, Shah JY. 2008. How goal instrumentality shapes relationship evaluations. J. Personal. Soc.

Psychol. 95:319–37
Fletcher GJ, Kerr PS. 2010. Through the eyes of love: reality and illusion in intimate relationships. Psychol.

Bull. 136:627–58
Fletcher GJO, Simpson JA, Campbell L, Overall NC. 2015. Pair-bonding, romantic love, and evolution: the

curious case of Homo sapiens. Perspect. Psychol. Sci. 10:20–36
Fletcher GJO, Simpson JA, Thomas G. 2000a. Ideals, perceptions, and evaluations in early relationship

development. J. Personal. Soc. Psychol. 79:933–40
Fletcher GJO, Simpson JA, Thomas G. 2000b. The measurement of perceived relationship quality compo-

nents: a confirmatory factor analytic approach. Personal. Soc. Psychol. Bull. 26:340–54
Fraley RC. 2007. A connectionist approach to the organization and continuity of working models of attach-

ment. J. Personal. 75:1157–80
Gable SL, Gosnell CL, Maisel NC, Strachman A. 2012. Safely testing the alarm: close others’ responses to

personal positive events. J. Personal. Soc. Psychol. 103:963–81
Gable SL, Impett EA. 2012. Approach and avoidance motives and close relationships. Soc. Personal. Psychol.

Compass 6:95–108
Gable SL, Reis HT. 2001. Appetitive and aversive social interaction. In Close Romantic Relationships: Maintenance

and Enhancement, ed. JH Harvey, A Wenzel, pp. 169–94. Mahwah, NJ: Lawrence Erlbaum Assoc.
Gable SL, Reis HT. 2010. Good news! Capitalizing on positive events in an interpersonal context. Adv. Exp.

Soc. Psychol. 42:195–257
Gagné FM, Lydon JE. 2004. Bias and accuracy in close relationships: an integrative review. Personal. Soc.

Psychol. Rev. 8:322–38
Gangestad SW, Haselton MG, Buss DM. 2006. Evolutionary foundations of cultural variation: evoked culture

and mate preferences. Psychol. Inq. 17:75–95
Goldin C. 2014. A grand gender convergence: its last chapter. Am. Econ. Rev. 104:1091–119
Gottman JM. 1998. Psychology and the study of marital processes. Annu. Rev. Psychol. 49:169–97
Gottman JM, Levenson RW. 1999. Rebound from marital conflict and divorce prediction. Fam. Process.

38:287–92
Hazan C, Shaver P. 1987. Romantic love conceptualized as an attachment process. J. Personal. Soc. Psychol.

52:511–24
Hazan C, Shaver PR. 1994. Attachment as an organizational framework for research on close relationships.

Psychol. Inq. 5:1–22

www.annualreviews.org • Relationship Science 407

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Helm JL, Sbarra DA, Ferrer E. 2014. Coregulation of respiratory sinus arrhythmia in adult romantic partners.
Emotion 14:522–31

Hill R. 1949. Families under Stress: Adjustment to the Crises of War Separation and Reunion. New York: Harper
& Brothers

Holmes JG. 1981. The exchange process in close relationships: microbehavior and macromotives. In The
Justice Motive in Social Behavior, ed. MJ Lerner, SC Lerner, pp. 261–84. New York: Plenum

Holmes JG. 2002. Interpersonal expectations as the building blocks of social cognition: an interdependence
theory perspective. Pers. Relat. 9:1–26

Holmes JG, Rempel JK. 1989. Trust in close relationships. In Review of Personality and Social Psychology, Vol. 10,
Close Relationships, ed. C Hendrick, pp. 187–220. Thousand Oaks, CA: Sage

Holt-Lunstad J, Smith TB, Layton JB. 2010. Social relationships and mortality risk: a meta-analytic review.
PLOS Med. 7:e1000316

IJzerman H, Coan JA, Wagemans FM, Missler MA, Van Beest I, et al. 2015. A theory of social thermoregulation
in human primates. Front. Psychol. 6:464

Johnson DJ, Rusbult CE. 1989. Resisting temptation: devaluation of alternative partners as a means of main-
taining commitment in close relationships. J. Personal. Soc. Psychol. 57:967–80

Karney BR, Bradbury TN. 1995. The longitudinal course of marital quality and stability: a review of theory,
method, and research. Psychol. Bull. 118:3–34

Karney BR, Neff LA. 2015. Couples and stress: how demands outside a relationship affect intimacy within the
relationship. See Simpson & Campbell 2013, pp. 664–84

Karremans JC, Verwijmeren T. 2008. Mimicking attractive opposite-sex others: the role of romantic relation-
ship status. Personal. Soc. Psychol. Bull. 34:939–50

Kelley HH. 1979. Personal Relationships: Their Structures and Processes. Hillsdale, NJ: Erlbaum
Kelley HH, Berscheid E, Christensen A, Harvey JH, Huston TL, et al. 1983. Close Relationships. New York:

Freeman
Kelley HH, Holmes JG, Kerr NL, Reis HT, Rusbult CE, Van Lange PAM. 2003. An Atlas of Interpersonal

Situations. New York: Cambridge Univ. Press
Kelley HH, Thibaut JW. 1978. Interpersonal Relations: A Theory of Interdependence. New York: Wiley
Kenny DA, Kashy DA. 2011. Dyadic data analysis using multilevel modeling. In The Handbook of Advanced

Multilevel Analysis, ed. J Hox, JK Roberts, pp. 335–70. London: Taylor & Francis
Kim JS. 2013. The influence of local sex ratio on romantic relationship maintenance processes. PhD Thesis, Univ.

Minn., Minneapolis, MN
Knapp ML, Vangelisti AL, Caughlin JP. 2014. Interpersonal Communication and Human Relationships. Boston:

Allyn & Bacon. 7th ed.
Kurdek LA. 2004. Are gay and lesbian cohabiting couples really different from heterosexual married couples?

J. Marriage Fam. 66:880–900
Kurdek LA. 2005. What do we know about gay and lesbian couples? Curr. Dir. Psychol. Sci. 14:251–54
La Guardia JG, Ryan RM, Couchman CE, Deci EL. 2000. Within-person variation in security of attachment:

a self-determination theory perspective on attachment, need fulfillment, and well-being. Personal. Soc.
Psychol. 79:367–84

Lakey B, Orehek E. 2011. Relational regulation theory: a new approach to explain the link between perceived
social support and mental health. Psychol. Rev. 118:482–95

Laumann EO, Gagnon J, Michael R, Michaels S. 1994. The Social Organization of Sexuality: Sexual Practices in
the United States. Chicago: Univ. Chicago Press

Laurenceau J-P, Feldman Barrett LA, Pietromonaco PR. 1998. Intimacy as an interpersonal process: the
importance of self-disclosure and perceived partner responsiveness in interpersonal exchanges. J. Personal.
Soc. Psychol. 74:1238–51

Lavner JA, Bradbury TN, Karney BR. 2012. Incremental change or initial differences? Testing two models
of marital deterioration. J. Fam. Psychol. 26:606–16

Le B, Agnew CR. 2003. Commitment and its theorized determinants: a meta-analysis of the investment model.
Pers. Relat. 17:37–57

Le B, Dove NL, Agnew CR, Korn MS, Mutso AA. 2010. Predicting nonmarital romantic relationship disso-
lution: a meta-analytic synthesis. Pers. Relat. 17:377–90

408 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Lemay EP Jr., Dudley KL. 2011. Caution: fragile! Regulating the interpersonal security of chronically insecure
partners. J. Personal. Soc. Psychol. 100:681–702

Levinger G, Snoek JD. 1972. Attraction in Relationship: A New Look at Interpersonal Attraction. Morristown, NJ:
Gen. Learn. Press

Lewin K. 1936. Principles of Topological Psychology. New York: McGraw-Hill
Light AE, Fitzsimons GM. 2014. Contextualizing marriage as a means and a goal. Psychol. Inq. 25:88–94
Loving TJ, Gleason MEJ, Pope MT. 2009. Transition novelty moderates daters’ cortisol responses when

talking about marriage. Pers. Relat. 16:187–203
Luchies LB, Finkel EJ, McNulty JK, Kumashiro M. 2010. The doormat effect: when forgiving erodes self-

respect and self-concept clarity. J. Personal. Soc. Psychol. 98:734–49
Luchies LB, Wieselquist J, Rusbult CE, Kumashiro M, Eastwick PW. 2013. Trust and biased memory of

transgressions in romantic relationships. J. Personal. Soc. Psychol. 104:673–94
Lydon JE, Karremans JC. 2015. Relationship regulation in the face of eye candy: a motivated cognition

framework for understanding responses to attractive alternatives. Curr. Opin. Psychol. 1:76–80
Maner JK, Gailliot MT, Miller SL. 2009. The implicit cognition of relationship maintenance: inattention to

attractive alternatives. J. Exp. Soc. Psychol. 45:174–79
Maner JK, Rouby DA, Gonzaga GC. 2008. Automatic inattention to attractive alternatives: the evolved psy-

chology of relationship maintenance. Evol. Hum. Behav. 29:343–49
Marshall E, Simpson JA, Rholes WS. 2015. Personality, communication, and depressive symptoms across the

transition to parenthood: a dyadic longitudinal investigation. Eur. J. Personal. 29:216–34
Mashek DJ, Aron A, Boncimino M. 2003. Confusions of self with close others. Personal. Soc. Psychol. Bull.

29:382–92
McCubbin HI, Patterson JM. 1983. Family transitions: adaptation to stress. In Stress and the Family: Coping

with Normative Transitions, Vol. 1, ed. HI McCubbin, CR Figley, pp. 5–25. New York: Brunner-Mazel
McNulty JK. 2008. Forgiveness in marriage: putting the benefits into context. J. Fam. Psychol. 22:171–75
McNulty JK. 2013. Personality and relationships. See Simpson & Campbell 2013, pp. 535–52
McNulty JK. 2016a. Highlighting the contextual nature of interpersonal relationships. Adv. Exp. Soc. Psychol.

54:247–315
McNulty JK. 2016b. Should spouses be demanding less from marriage? A contextual perspective on the

implications of interpersonal standards. Personal. Soc. Psychol. Bull. 42:444–57
McNulty JK, Karney BR. 2004. Positive expectations in the early years of marriage: Should couples expect

the best or brace for the worst? J. Personal. Soc. Psychol. 86:729–43
McNulty JK, Olson MA, Meltzer AL, Shaffer MJ. 2013. Though they may be unaware, newlyweds implicitly

know whether their marriage will be satisfying. Science 342:1119–20
Mikulincer M, Shaver PR. 2007. Attachment in Adulthood: Structure, Dynamics, and Change. New York: Guilford

Press
Miller RS. 1997. Inattentive and contented: relationship commitment and attention to alternatives. J. Personal.

Soc. Psychol. 73:758–66
Miller SL, Maner JK. 2010. Evolution and relationship maintenance: Fertility cues lead committed men to

devalue relationship alternatives. J. Exp. Soc. Psychol. 46:1081–84
Murray SL, Gomillion S, Holmes JG, Harris B, Lamarche V. 2013. The dynamics of relationship promotion:

controlling the automatic inclination to trust. J. Personal. Soc. Psychol. 104:305–34
Murray SL, Griffin DW, Derrick JL, Harris B, Aloni M, Leder S. 2011. Tempting fate or inviting happiness?

Unrealistic idealization prevents the decline of marital satisfaction. Psychol. Sci. 22:619–26
Murray SL, Holmes J. 2009. The architecture of interdependent minds: a motivation-management theory of

mutual responsiveness. Psychol. Rev. 116:908–28
Murray SL, Holmes JG, Collins NL. 2006. Optimizing assurance: the risk regulation system in relationships.

Psychol. Bull. 132:641–66
Murray SL, Holmes JG, Griffin DW. 1996. The self-fulfilling nature of positive illusions in romantic rela-

tionships: Love is not blind, but prescient. J. Personal. Soc. Psychol. 71:1155–80
Neff LA, Broady EF. 2011. Stress resilience in early marriage: Can practice make perfect? J. Personal. Soc.

Psychol. 101:1050–67

www.annualreviews.org • Relationship Science 409

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Neff LA, Karney BR. 2003. The dynamic structure of relationship perceptions: differential importance as a
strategy of relationship maintenance. Personal. Soc. Psychol. Bull. 29:1433–46

Neff LA, Karney BR. 2009. Stress and reactivity to daily relationship experiences: how stress hinders adaptive
processes in marriage. J. Personal. Soc. Psychol. 97:435–50

Otto AK, Laurenceau JP, Siegel SD, Belcher AJ. 2015. Capitalizing on everyday positive events uniquely
predicts daily intimacy and well-being in couples coping with breast cancer. J. Personal. Soc. Psychol.
29:69–79

Overall NC, Fletcher GJO, Simpson JA, Sibley CG. 2009. Regulating partners in intimate relationships: the
costs and benefits of different communication strategies. J. Personal. Soc. Psychol. 96:620–39

Overall NC, McNulty JK. 2017. What type of communication during conflict is beneficial for intimate
relationships? Curr. Opin. Psychol. 13:1–5

Pierce T, Lydon JE. 2001. Global and specific relational models in the experience of social interactions.
J. Personal. Soc. Psychol. 80:613–31

Randall AK, Bodenmann G. 2009. The role of stress on close relationships and marital satisfaction. Clin.
Psychol. Rev. 29:105–15

Reis HT. 2007. Steps toward the ripening of relationship science. Pers. Relat. 14:1–23
Reis HT, Clark MS. 2013. Responsiveness. See Simpson & Campbell 2013, pp. 400–23
Reis HT, Shaver PR. 1988. Intimacy as an interpersonal process. In Handbook of Personal Relationships: Theory,

Research, and Interventions, ed. S Duck, pp. 367–89. Chichester: Wiley
Repetti RL. 1989. Effects of daily workload on subsequent behavior during marital interaction: the roles of

social withdrawal and spouse support. J. Personal. Soc. Psychol. 57:651–59
Rholes WS, Simpson JA, Kohn JL, Wilson CL, Martin AM, et al. 2011. Attachment orientations and depres-

sion: a longitudinal study of new parents. J. Personal. Soc. Psychol. 100:567–86
Richerson PJ, Boyd R. 2005. Not by Genes Alone: How Culture Transformed Human Evolution. Chicago: Univ.

Chicago Press
Robles TF, Slatcher RB, Trombello JM, McGinn MM. 2014. Marital quality and health: a meta-analytic

review. Psychol. Bull. 140:140–87
Rubin H, Campbell L. 2012. Day-to-day changes in intimacy predict heightened relationship passion, sexual

occurrence, and sexual satisfaction: a dyadic diary analysis. Soc. Psychol. Personal. Sci. 3:224–31
Rusbult CE. 1983. A longitudinal test of the investment model: the development (and deterioration) of satis-

faction and commitment in heterosexual involvements. J. Personal. Soc. Psychol. 45:101–17
Rusbult CE, Finkel EJ, Kumashiro M. 2009. The Michelangelo phenomenon. Curr. Dir. Psychol. Sci. 18:305–9
Rusbult CE, Olsen N, Davis JL, Hannon P. 2001. Commitment and relationship maintenance mechanisms.

In Close Romantic Relationships: Maintenance and Enhancement, ed. JH Harvey, A Wenzel, pp. 87–113.
Mahwah, NJ: Erlbaum

Rusbult CE, Van Lange P. 2003. Interdependence, interaction, and relationships. Annu. Rev. Psychol. 54:351–
75

Rusbult CE, Van Lange PA, Wildschut T, Yovetich NA, Verette J. 2000. Perceived superiority in close
relationships: why it exists and persists. J. Personal. Soc. Psychol. 79:521–45

Rusbult CE, Zembrodt IM, Gunn LK. 1982. Exit, voice, loyalty and neglect: responses to dissatisfaction in
romantic involvements. J. Personal. Soc. Psychol. 43:1230–42

Sbarra DA, Hazan C. 2008. Coregulation, dysregulation, self-regulation: an integrative analysis and empirical
agenda for understanding adult attachment, separation, loss, and recovery. Personal. Soc. Psychol. Rev.
12:141–67

Schaller M, Murray DR. 2008. Pathogens, personality, and culture: Disease prevalence predicts worldwide
variability in sociosexuality, extraversion, and openness to experience. J. Personal. Soc. Psychol. 95:212–21

Schmitt DP. 2014. On the proper functions of human mate preference adaptations: comment on Eastwick,
Luchies, Finkel, and Hunt 2014. Psychol. Bull. 140:666–72

Schmitt DP, Alcalay L, Allensworth M, Allik J, Ault L, et al. 2005. Sociosexuality from Argentina to Zimbabwe:
a 48-nation study of sex, culture, and the strategies of human mating. Behav. Brain Sci. 28:247–311

Sedikides C, Campbell WK, Reeder GD, Elliot AJ. 1998. The self-serving bias in relational context. J. Personal.
Soc. Psychol. 74(2):378–86

410 Finkel · Simpson · Eastwick

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68CH15-Finkel ARI 16 November 2016 8:6

Shallcross S, Howland M, Bemis J, Simpson JA, Frazier P. 2011. Not “capitalizing” on social capitalization
interactions: the role of attachment insecurity. J. Fam. Psychol. 25:77–85

Shallcross S, Simpson JA. 2012. Trust and responsiveness in strain-test situations: a dyadic perspective.
J. Personal. Soc. Psychol. 102:1031–44

Simon W, Gagnon JH. 2003. Sexual scripts: origins, influences and changes. Qual. Sociol. 26:491–97
Simpson JA. 2007. Psychological foundations of trust. Curr. Dir. Psychol. Sci. 16:264–68
Simpson JA, Belsky J. 2016. Attachment theory within a modern evolutionary framework. In The Handbook of

Attachment: Theory, Research, and Clinical Applications, ed. J Cassidy, PR Shaver, pp. 91–116. New York:
Guilford. 3rd ed.

Simpson JA, Campbell L. 2013. The Oxford Handbook of Close Relationships. New York: Oxford Univ. Press
Simpson JA, Fletcher GJO, Campbell L. 2001. The structure and function of ideal standards in close rela-

tionships. In Blackwell Handbook of Social Psychology: Interpersonal Processes, ed. GJO Fletcher, MS Clark,
pp. 86–106. Malden, MA: Blackwell Publishers

Simpson JA, Gangestad SW, Lerma M. 1990. Perception of physical attractiveness: mechanisms involved in
the maintenance of romantic relationships. J. Personal. Soc. Psychol. 59:1192–201

Simpson JA, Ickes W, Blackstone T. 1995. When the head protects the heart: empathic accuracy in dating
relationships. J. Personal. Soc. Psychol. 69:629–41

Simpson JA, Overall NC. 2014. Partner buffering of attachment insecurity. Curr. Dir. Psychol. Sci. 23:54–59
Simpson JA, Rholes WS. 2012. Adult attachment orientations, stress, and romantic relationships. Adv. Exp.

Soc. Psychol. 45:279–328
Simpson JA, Rholes WS, Nelligan JS. 1992. Support-seeking and support-giving within couples in an anxiety-

provoking situation: the role of attachment styles. J. Personal. Soc. Psychol. 62:434–46
Slotter EB, Gardner WL. 2009. Where do you end and I begin? Evidence for anticipatory, motivated self-other

integration between relationship partners. J. Personal. Soc. Psychol. 96:1137–51
Sprecher S. 2011. The influence of social networks on romantic relationships: through the lens of the social

network. Pers. Relat. 18:630–44
Sternberg RJ. 1986. A triangular theory of love. Psychol. Rev. 93:119–35
Szepsenwol O, Griskevicius V, Simpson JA, Young ES, Fleck C, Jones RE. 2016. The effect of predictable

early childhood environments on sociosexuality in early adulthood. Evol. Behav. Sci. In press
Thibaut JW, Kelley HH. 1959. The Social Psychology of Groups. New York: Wiley
Uchino BN, Bosch JA, Smith TW, Carlisle M, Birmingham W, et al. 2013. Relationships and cardiovascular

risk: perceived spousal ambivalence in specific relationship contexts and its links to inflammation. Health
Psychol. 32:1067–75

West TV, Kenny DA. 2011. The truth and bias model of judgment. Psychol. Rev. 118:357–78
Wieselquist J, Rusbult C, Foster C, Agnew C. 1999. Commitment, prorelationship behavior, and trust in close

relationships. J. Personal. Soc. Psychol. 77:942–66
Zeifman D, Hazan C. 2008. Pairbonds as attachments: reevaluating the evidence. In The Handbook of At-

tachment: Theory, Research, and Clinical Applications, ed. J Cassidy, PR Shaver, pp. 436–55. New York:
Guilford

www.annualreviews.org • Relationship Science 411

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68-FrontMatter ARI 7 November 2016 13:25

Annual Review of
Psychology

Volume 68, 2017

Contents
Eavesdropping on Memory

Elizabeth F. Loftus � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 1

Memory: Organization and Control
Howard Eichenbaum � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �19

Neural Mechanisms of Selective Visual Attention
Tirin Moore and Marc Zirnsak � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �47

Learning, Reward, and Decision Making
John P. O’Doherty, Jeffrey Cockburn, and Wolfgang M. Pauli � � � � � � � � � � � � � � � � � � � � � � � � � � � �73

Reinforcement Learning and Episodic Memory in Humans and Animals:
An Integrative Framework
Samuel J. Gershman and Nathaniel D. Daw � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 101

Social Learning and Culture in Child and Chimpanzee
Andrew Whiten � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 129

Survival of the Friendliest: Homo sapiens Evolved via Selection for
Prosociality
Brian Hare � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 155

Numerical Development
Robert S. Siegler and David W. Braithwaite � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 187

Gene × Environment Interactions: From Molecular Mechanisms to
Behavior
Thorhildur Halldorsdottir and Elisabeth B. Binder � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 215

The Structure of Social Cognition: In(ter)dependence of Sociocognitive
Processes
Francesca Happé, Jennifer L. Cook, and Geoffrey Bird � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 243

Toward a Social Psychophysics of Face Communication
Rachael E. Jack and Philippe G. Schyns � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 269

Social Motivation: Costs and Benefits of Selfishness and Otherishness
Jennifer Crocker, Amy Canevello, and Ashley A. Brown � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 299

Attitude Strength
Lauren C. Howe and Jon A. Krosnick � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 327

How Power Affects People: Activating, Wanting, and Goal Seeking
Ana Guinote � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 353

vi

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

PS68-FrontMatter ARI 7 November 2016 13:25

The Psychology of Close Relationships: Fourteen Core Principles
Eli J. Finkel, Jeffry A. Simpson, and Paul W. Eastwick � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 383

Moving Beyond Correlations in Assessing the Consequences of Poverty
Greg J. Duncan, Katherine Magnuson, and Elizabeth Votruba-Drzal � � � � � � � � � � � � � � � � � � 413

Culture Three Ways: Culture and Subcultures Within Countries
Daphna Oyserman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 435

Learning from Errors
Janet Metcalfe � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 465

Mindfulness Interventions
J. David Creswell � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 491

Hidden Wounds? Inflammatory Links Between Childhood Trauma and
Psychopathology
Andrea Danese and Jessie R. Baldwin � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 517

Adjusting to Chronic Health Conditions
Vicki S. Helgeson and Melissa Zajdel � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 545

Health Behavior Change: Moving from Observation to Intervention
Paschal Sheeran, William M.P. Klein, and Alexander J. Rothman � � � � � � � � � � � � � � � � � � � � � � 573

Experiments with More than One Random Factor: Designs, Analytic
Models, and Statistical Power
Charles M. Judd, Jacob Westfall, and David A. Kenny � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 601

Interactions with Robots: The Truths We Reveal About Ourselves
Elizabeth Broadbent � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 627

Indexes

Cumulative Index of Contributing Authors, Volumes 58–68 � � � � � � � � � � � � � � � � � � � � � � � � � � � 653

Cumulative Index of Article Titles, Volumes 58–68 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 658

Errata

An online log of corrections to Annual Review of Psychology articles may be found at
http://www.annualreviews.org/errata/psych

Contents vii

A
nn

u.
R

ev
. P

sy
ch

ol
. 2

01
7.

68
:3

83
-4

11
. D

ow
nl

oa
de

d
fr

om
w

w
w

.a
nn

ua
lr

ev
ie

w
s.

or
g

A
cc

es
s

pr
ov

id
ed

b
y

R
eg

is
U

ni
ve

rs
it

y
on

0
2/

15
/2

2.
F

or
p

er
so

na
l

us
e

on
ly

.

Psychological Bulletin
1995, Vol. 117, No. 3, 497-529

Copyright 1995 by the American Psychological Association, Inc.
0033-2909/95/S3.00

The Need to Belong: Desire for Interpersonal Attachments as a
Fundamental Human Motivation

Roy F. Baumeister
Case Western Reserve University

Mark R. Leary
Wake Forest University

A hypothesized need to form and maintain strong, stable interpersonal relationships is evaluated in
light of the empirical literature. The need is for frequent, nonaversive interactions within an ongoing
relational bond. Consistent with the belongingness hypothesis, people form social attachments
readily under most conditions and resist the dissolution of existing bonds. Belongingness appears to
have multiple and strong effects on emotional patterns and on cognitive processes. Lack of attach-
ments is linked to a variety of ill effects on health, adjustment, and well-being. Other evidence, such
as that concerning satiation, substitution, and behavioral consequences, is likewise consistent with
the hypothesized motivation. Several seeming counterexamples turned out not to disconfirm the
hypothesis. Existing evidence supports the hypothesis that the need to belong is a powerful, funda-
mental, and extremely pervasive motivation.

The purpose of this review is to develop and evaluate the hy-
pothesis that a need to belong is a fundamental human motiva-
tion and to propose that the need to belong can provide a point
of departure for understanding and integrating a great deal of
the existing literature regarding human interpersonal behavior.
More precisely, the belongingness hypothesis is that human be-
ings have a pervasive drive to form and maintain at least a min-
imum quantity of lasting, positive, and significant interpersonal
relationships. Satisfying this drive involves two criteria: First,
there is a need for frequent, affectively pleasant interactions
with a few other people, and, second, these interactions must
take place in the context of a temporally stable and enduring
framework of affective concern for each other’s welfare. Interac-
tions with a constantly changing sequence of partners will be
less satisfactory than repeated interactions with the same
person (s), and relatedness without frequent contact will also be
unsatisfactory. A lack of belongingness should constitute severe
deprivation and cause a variety of ill effects. Furthermore, a
great deal of human behavior, emotion, and thought is caused
by this fundamental interpersonal motive.

The hypothesis that people are motivated to form and maintain
interpersonal bonds is not new, of course. John Donne (1975) has
been widely quoted for the line “No [person] is an island.” In psy-
chology, the need for interpersonal contact was asserted in several
ways by Freud (e.g., 1930), although he tended to see the motive
as derived from the sex drive and from the filial bond. Maslow

Roy F. Baumeister, Department of Psychology, Case Western Reserve
University; Mark R. Leary, Department of Psychology, Wake Forest
University.

We thank Bob Hogan, Ned Jones, Richard Moreland, Dave Myers,
Len Newman, Paula Pietromonaco, Harry Reis, Dan Wegner, and Di-
anne Tice for comments on preliminary drafts.

Correspondence concerning this article should be addressed to Roy
F. Baumeister, Department of Psychology, Case Western Reserve Uni-
versity, 10900 Euclid Avenue, Cleveland, Ohio 44106-7123. Electronic
mail may be sent via Internet to [email protected]

(1968) ranked “love and belongingness needs” in the middle of his
motivational hierarchy; that is, belongingness needs do not emerge
until food, hunger, safety, and other basic needs are satisfied, but
they take precedence over esteem and self-actualization. Bowlby’s
(e.g., 1969,1973) attachment theory also posited the need to form
and maintain relationships. His early thinking followed the Freud-
ian pattern of deriving attachment needs from the relationship to
one’s mother; he regarded the adult’s need for attachment as an
effort to recapture the intimate contact that the individual had, as
an infant, with his or her mother.1 Horney (1945), Sullivan
(1953), Fromm (1955, 1956), de Rivera(1984), Hogan (1983),
Epstein (1992), Ryan (1991), Guisinger and Blatt (1994), and
others have made similar suggestions. The existence of a need to
belong is thus a familiar point of theory and speculation, although
not all theorists have anticipated our particular formulation of this
need as the combination of frequent interaction plus persistent
caring. Moreover, most theorists have neglected to provide system-
atic empirical evaluation of this hypothesis. For example, Mas-
low’s (1968) influential assertion of a belongingness need was ac-
companied by neither original data nor review of previous find-
ings. Thus, despite frequent, speculative assertions that people
need to belong, the belongingness hypothesis needs to be critically
evaluated in light of empirical evidence. A main goal of the present
article is to assemble a large body of empirical findings pertinent
to the belongingness hypothesis to evaluate how well the hypothe-
sis fits the data.

Another goal of this article is to demonstrate the broad appli-
cability of the need to belong for understanding human motiva-
tion and behavior. Even though many psychological theorists
have noted human affiliative tendencies in one form or another,
the field as a whole has neglected the broad applicability of this

1 His later thinking may, however, have moved beyond this view to
regard attachment needs as having a separate, even innate basis rather
than being derived from the contact with one’s mother; in this later
view, he treated the relationship to one’s mother as simply an influential
prototype of attachment.

497

498 ROY F. BAUMEISTER AND MARK R. LEARY

need to a wide range of behaviors. Thus, for example, the mo-
tive literature has been dominated by research on the respective
needs for power, achievement, intimacy, approval, and, to a
lesser extent, affiliation. But the need for power may well be
driven by the need to belong, as we suggest later. Likewise, peo-
ple prefer achievements that are validated, recognized, and val-
ued by other people over solitary achievements, so there may
be a substantial interpersonal component behind the need for
achievement. And the needs for approval and intimacy are un-
doubtedly linked to the fact that approval is a prerequisite for
forming and maintaining social bonds, and intimacy is a denn-
ing characteristic of close relationships. The need to belong
could thus be linked to all of them.

Furthermore, even a quick glance at research on social be-
havior from the perspective of the belongingness hypothesis
raises the possibility that much of what human beings do is done
in the service of belongingness. Thus, the belongingness hypoth-
esis might have considerable value for personality and social
psychology and even for psychology as a whole. As a broad in-
tegrative hypothesis, it might help rectify what some observers
have criticized as fragmentation and atomization in the concep-
tual underpinnings of the field (see Vallacher & Nowak, 1994;
West, Newsom, & Fenaughty, 1992).

At the interdisciplinary level, the belongingness hypothesis
might help psychology recover from the challenge posed by cul-
tural materialism. Cultural materialism (e.g., Harris, 1974,
1978, 1979) is based on the assumption that human culture is
shaped primarily by economic needs and opportunities, and so
historical, anthropological, sociological, and other cultural pat-
terns should mainly be analyzed with reference to economic
causes. In that framework, psychology is reduced to a vastly
subordinate role; psychological phenomena are regarded
merely as symptoms or coping mechanisms that follow from
economic realities. In contrast, the belongingness hypothesis
would suggest that human culture is at least partly adapted to
enable people to satisfy the psychological need to live together
(along with economic needs, to be sure), thereby assigning
some fundamental causal power to psychological forces. We
suggest that belongingness can be almost as compelling a need
as food and that human culture is significantly conditioned by
the pressure to provide belongingness.

Modern personality and social psychologists have shown a
pervasive reluctance to entertain sweeping generalizations and
broad hypotheses. This reluctance may well be a response to
speculative excesses of earlier generations of theorists, who sup-
posedly rushed to formulate broad theories from intuition and
impression. Today there may be a sense that it is more appro-
priate to await the passing of a substantial interval, until con-
siderable empirical work has been done. We propose that such
an interval has passed, however, making it possible to begin con-
sidering broad hypotheses in light of the evidence accumulated
through the last three decades. That is what we undertake here.

Conceptual Background

Fundamental Motivations: Metatheory

Before proceeding with our examination of the need to be-
long, we must consider briefly the metatheoretical requirements

of our hypothesis. That is, what criteria must be satisfied to con-
clude that the need to belong, or any other drive, is a fundamen-
tal human motivation? We suggest the following. A fundamen-
tal motivation should (a) produce effects readily under all but
adverse conditions, (b) have affective consequences, (c) direct
cognitive processing, (d) lead to ill effects (such as on health or
adjustment) when thwarted, (e) elicit goal-oriented behavior
designed to satisfy it (subject to motivational patterns such as
object substitutability and satiation), (f) be universal in the
sense of applying to all people, (g) not be derivative of other
motives, ( h ) affect a broad variety of behaviors, and ( i ) have
implications that go beyond immediate psychological function-
ing. We consider each of these criteria in turn.

The first criterion is that a fundamental motivation should
operate in a wide variety of settings: any motive that requires
highly specific or supportive circumstances to produce effects
cannot properly be called fundamental. Certain circumstances
may retard or prevent its operation, but in general the more
widely it can produce effects, the stronger its claim to being a
fundamental motivation.

The second and third criteria refer to emotional and cognitive
patterns. Cognitive and emotional responses reflect subjective
importance and concern, and a motivation that fails to guide
emotion and cognition (at least sometimes) can hardly be con-
sidered an important one. In addition, most motivational and
drive systems involve hedonic consequences that alert the indi-
vidual to undesired state changes that motivate behavior to re-
store the desired state and whose removal serves as negative re-
inforcement for goal attainment.

The fourth criterion is that failure to satisfy a fundamental
motivation should produce ill effects that go beyond temporary
affective distress. A motivation can be considered to be funda-
mental only if health, adjustment, or well-being requires that it
be satisfied. Also, motivations can be sorted into wants and
needs, the difference being in the scope of ill effects that follow
from nonsatisfaction: Unsatisfied needs should lead to pathol-
ogy (medical, psychological, or behavioral), unlike unsatisfied
wants. Thus, if belongingness is a need rather than simply a
want, then people who lack belongingness should exhibit patho-
logical consequences beyond mere temporary distress.

Substitution and satiation are two familiar hallmarks of mo-
tivation. If the need to belong is a fundamental need, then be-
longing to one group should satisfy it and hence obviate or re-
duce the need to belong to another group. People may be driven
to form social bonds until they have a certain number, whereaf-
ter the drive to form attachments would presumably subside.
Furthermore, attachment partners should be to some degree in-
terchangeable. Of course, this does not mean that a 20-year
spouse or friend can be simply replaced with a new acquain-
tance. In the long run, however, a new spouse or friend should
do as well as the previous one.

The sixth and seventh criteria involve universality and non-
derivativeness. Any motivation that is limited to certain human
beings or certain circumstances, or any motivation that is de-
rived from another motive, cannot be regarded as fundamental.
Universality can be indicated by transcending cultural bound-
aries. Establishing that a motive is not derivative is not easy,
although path-analytic models can suggest derivative patterns.
Satisfying the first criterion may also help satisfy the seventh,

THE NEED TO BELONG 499

because if the motivation operates in a broad variety of situa-
tions without requiring particular, favorable circumstances,
then it may be presumed to be fundamental. Meanwhile, if the
evidence contradicts evolutionary patterns or fails to indicate
physiological mechanisms, then the hypothesis of universality
or innateness would lose credibility.

The eighth criterion is the ability to affect a wide and diverse
assortment of behaviors. The more behaviors that appear to be
influenced by a particular motive, the stronger its case for being
one of the fundamental motives. Lastly, we suggest that a fun-
damental motive should have implications that go beyond psy-
chological functioning. If a motivation is truly fundamental, it
should influence a broad range of human activity, and hence it
should be capable of offering viable and consistent inter-
pretations of patterns observed in historical, economic, or so-
ciological studies.

Falsification is only one relevant approach to evaluating a
broad hypothesis about belongingness being a fundamental mo-
tivation. The belongingness hypothesis could indeed be falsified
if it were shown, for example, that many people can live happy,
healthy lives in social isolation or that many people show no
cognitive or emotional responses to looming significant changes
in their belongingness status. In addition to such criteria, how-
ever, hypotheses about fundamental motivations must be evalu-
ated in terms of their capacity to interpret and explain a wide
range of phenomena. Part of the value of such a theory is its
capacity to provide an integrative framework, and this value is
a direct function of the quantity and importance of the behavior
patterns that it can explain in a consistent, intelligible fashion.
We therefore pay close attention to the potential range of im-
plications of the belongingness hypothesis, in addition to exam-
ining how many falsification tests the hypothesis has managed
to survive.

The Need to Belong: Theory

In view of the metatheoretical requirements listed in the pre-
vious section, we propose that a need to belong, that is, a need
to form and maintain at least a minimum quantity of interper-
sonal relationships, is innately prepared (and hence nearly
universal) among human beings. Thus, unlike the Freudian
(1930) view that regarded sexuality and aggression as the major
driving psychological forces, and unlike the most ambitious be-
haviorist views that considered each newborn a tabula rasa, our
view depicts the human being as naturally driven toward estab-
lishing and sustaining belongingness. The need to belong should
therefore be found to some degree in all humans in all cultures,
although naturally one would expect there to be individual
differences in strength and intensity, as well as cultural and in-
dividual variations in how people express and satisfy the need.
But it should prove difficult or impossible for culture to eradi-
cate the need to belong (except perhaps for an occasional, seri-
ously warped individual).

The innate quality presumably has an evolutionary basis.
It seems clear that a desire to form and maintain social bonds
would have both survival and reproductive benefits (see Ains-
worth, 1989; Axelrod & Hamilton, 1981; Barash, 1977;
Bowlby, 1969; D. M. Buss, 1990, 1991; Hogan, Jones, &
Cheek, 1985; Moreland, 1987). Groups can share food, pro-

vide mates, and help care for offspring (including orphans).
Some survival tasks, such as hunting large animals or main-
taining defensive vigilance against predatory enemies, are
best accomplished by group cooperation. Children who de-
sired to stay together with adults (and who would resist being
left alone) would be more likely to survive until their repro-
ductive years than other children because they would be more
likely to receive care and food as well as protection. Cues that
connote possible harm, such as illness, danger, nightfall, and
disaster, seem to increase the need to be with others (see also
Rofe, 1984), which again underscores the protective value of
group membership. Adults who formed attachments would
be more likely to reproduce than those who failed to form
them, and long-term relationships would increase the
chances that the offspring would reach maturity and repro-
duce in turn (see also Shaver, Hazan, & Bradshaw, 1988).2

Competition for limited resources could also provide a pow-
erful stimulus to forming interpersonal connections. There are
several potential, although debatable, advantages to forming a
group under conditions of scarcity. For example, groups may
share resources and thus prevent any individual from starving
(although sharing deprives other group members of some of
their resources), and groups may appropriate resources from
nonmembers (although there is the problem of how to distrib-
ute them in the group). What appears less debatable is the se-
vere competitive disadvantage of the lone individual confront-
ing a group when both want the same resource. When other
people are in groups, it is vital to belong to a group oneself,
particularly a group of familiar, cooperative people who care
about one’s welfare. Thus, an inclination to form and sustain
social bonds would have important benefits of defending oneself
and protecting one’s resources against external threats.

The likely result of this evolutionary selection would be a set
of internal mechanisms that guide individual human beings
into social groups and lasting relationships. These mechanisms
would presumably include a tendency to orient toward other
members of the species, a tendency to experience affective dis-
tress when deprived of social contact or relationships, and a ten-
dency to feel pleasure or positive affect from social contact and
relatedness. These affective mechanisms would stimulate learn-
ing by making positive social contact reinforcing and social de-
privation punishing.

Our version of the belongingness hypothesis does not regard
the need as derived from a particular relationship or focused on
a particular individual. In this, it differs from the early, Freudian
version of Bowlby’s work, in which the relationship to the
mother was regarded as the cause of the desire for attachment.
Thus, Bowlby suggested that adult attachments to work organi-
zations, religious groups, or others are derived from the child’s
tie to mother and revolve around personal attachment to the
group leader or supervisor (Bowlby, 1969, p. 207). In contrast,

2 A possible sex difference could be suggested in the mode of express-
ing this need, however, in that men may be more oriented toward form-
ing relationships, whereas women may be more oriented toward main-
taining them. Men can reproduce many times by forming many brief
relationships, whereas women can reproduce only about once per year,
and so their most effective reproductive strategy would be to enable each
child to receive maximal care and protection (D. M. Buss, 1991).

500 ROY F. BAUMEISTER AND MARK R. LEARY

we propose that the need to belong can, in principle, be directed
toward any other human being, and the loss of relationship with
one person can to some extent be replaced by any other. The
main obstacle to such substitution is that formation of new re-
lationships takes time, such as in the gradual accumulation of
intimacy and shared experience (see Sternberg, 1986, on the
time course of intimacy). Social contact with a long-term inti-
mate would therefore provide some satisfactions, including a
sense of belonging, that would not be available in interactions
with strangers or new acquaintances.

The belongingness hypothesis can be distinguished from a hy-
pothesized need for mere social contact in terms of whether in-
teractions with strangers or with people one dislikes or hates
would satisfy the need. It can be distinguished from a hypothe-
sized need for positive, pleasant social contact in terms of
whether nonhostile interactions with strangers would satisfy it.
The need to belong entails that relationships are desired, so in-
teractions with strangers would mainly be appealing as possible
first steps toward long-term contacts (including practicing so-
cial skills or learning about one’s capacity to attract partners),
and interactions with disliked people would not satisfy it.

Additional differences between the belongingness hypothesis
and attachment theory could be suggested, although it may be a
matter of interpretation whether these are merely differences of
emphasis or fundamental theoretical differences. In our under-
standing, the (very real) strengths of attachment theory are two-
fold. First, attachment theory has emphasized the task of elab-
orating individual differences in attachment style (e.g., Hazan
&Shaver, 1994a, 1994b; Shaver etal., 1988), whereas we focus
on the commonality of the overarching need to belong. Second,
attachment theory has emphasized certain emotional needs and
satisfactions implicit in certain kinds of relationships, whereas
we regard it as at least plausible that the need to belong could
be satisfied in other ways. For example, one might imagine a
young fellow without any family or intimate relationships who
is nonetheless satisfied by being heavily involved in an ideologi-
cally radical political movement. There are undoubtedly strong
emotional mechanisms associated with belongingness, as we
show later, but these could be understood as mediating mecha-
nisms rather than as essential properties.

Asa fundamental motivation, the need to belong should stim-
ulate goal-directed activity designed to satisfy it. People should
show tendencies to seek out interpersonal contacts and cultivate
possible relationships, at least until they have reached a mini-
mum level of social contact and relatedness. Meanwhile, social
bonds should form easily, readily, and without requiring highly
particular or conducive settings. (Indeed, if social attachments
form through shared unpleasant experiences, contrary to what
simple association models might predict, this would be espe-
cially compelling support for the belongingness hypothesis.)
Cognitive activity should reflect a pervasive concern with form-
ing and maintaining relationships. Emotional reactions should
follow directly from outcomes that pertain to the need to be-
long. More precisely, positive affect should follow from forming
and solidifying social bonds, and negative affect should ensue
when relationships are broken, threatened, or refused.

If belongingness is indeed a fundamental need, then aver-
sive reactions to a loss of belongingnsss should go beyond
negative affect to include some types of pathology. People

who are socially deprived should exhibit a variety of ill
effects, such as signs of maladjustment or stress, behavioral or
psychological pathology, and possibly health problems. They
should also show an increase in goal-directed activity aimed
at forming relationships.

In addition, the belongingness hypothesis entails that people
should strive to achieve a certain minimum quantity and qual-
ity of social contacts but that once this level is surpassed, the
motivation should diminish. The need is presumably for a cer-
tain minimum number of bonds and quantity of interaction.
The formation of further social attachments beyond that mini-
mal level should be subject to diminishing returns; that is, peo-
ple should experience less satisfaction on formation of such ex-
tra relationships, as well as less distress on terminating them.
Satiation patterns should be evident, such that people who are
well enmeshed in social relationships would be less inclined to
seek and form additional bonds than would people who are so-
cially deprived. Relationships should substitute for each other,
to some extent, as would be indicated by effective replacement
of lost relationship partners and by a capacity for social related-
ness in one sphere to overcome potential ill effects of social de-
privation in another sphere (e.g., if strong family ties compen-
sate for aloneness at work).

We propose that the need to belong has two main features.
First, people need frequent personal contacts or interactions
with the other person. Ideally, these interactions would be affec-
tively positive or pleasant, but it is mainly important that the
majority be free from conflict and negative affect.

Second, people need to perceive that there is an interpersonal
bond or relationship marked by stability, affective concern, and
continuation into the foreseeable future. This aspect provides a
relational context to one’s interactions with the other person,
and so the perception of the bond is essential for satisfying the
need to belong. When compared with essentially identical in-
teractions with other people with whom one is not connected, a
strictly behavioral record might reveal nothing special or re-
warding about these interactions. Yet an interaction with a per-
son in the context of an ongoing relationship is subjectively
different from and often more rewarding than an interaction
with a stranger or casual acquaintance. To satisfy the need to
belong, the person must believe that the other cares about his or
her welfare and likes (or loves) him or her.

Ideally this concern would be mutual, so that the person has
reciprocal feelings about the other. M. S. Clark and her col-
leagues (e.g., Clark, 1984; Clark & Mills, 1979; Clark, Mills, &
Corcoran, 1989; Clark, Mills, & Powell, 1986) have shown that
a framework of mutual concern produces a relationship quali-
tatively different from one based on self-interested social ex-
change. Still, it is plausible that mutuality is merely desirable
rather than essential. The decisive aspect may be the perception
that one is the recipient of the other’s lasting concern.

Viewed in this way, the need to belong is something other than
a need for mere affiliation. Frequent contacts with nonsup-
portive, indifferent others can go only so far in promoting one’s
general well-being and would do little to satisfy the need to be-
long. Conversely, relationships characterized by strong feelings
of attachment, intimacy, or commitment but lacking regular
contact will also fail to satisfy the need. Simply knowing that
a bond exists may be emotionally reassuring, yet it would not

THE NEED TO BELONG 501

provide full belongingness if one does not interact with the other
person. Thus, we view the need to belong as something more
than either a need for affiliation or a need for intimate
attachment.

The notion that people need relationships characterized by
both regular contact and an ongoing bond has been anti-
cipated to some degree by Weiss (1973; see also Shaver &
Buhrmester, 1983), who suggested that feelings of loneliness
can be precipitated either by an insufficient amount of social
contact (social loneliness) or by a lack of meaningful, inti-
mate relatedness (emotional loneliness). Weiss’s distinction
has been criticized on conceptual and empirical grounds
(e.g., Paloutzian & Janigian, 1987; Perlman, 1987), and
efforts to operationalize and test the distinction have met with
mixed results (DiTommaso & Spinner, 1993; Saklofske &
Yackulic, 1989; Vaux, 1988). In our view, the difficulty with
this distinction arises from the assumption that people have a
need for mere social contact and a separate need for intimate
relationships. Rather, the need is for regular social contact
with those to whom one feels connected. From an evolution-
ary perspective, relationships characterized by both of these
features would have greater survival and reproductive value
than would relationships characterized by only one. Accord-
ingly, the need to belong should be marked by both aspects.

Review of Empirical Findings

We searched the empirical literature of social and personality
psychology for findings relevant to the belongingness hypothe-
sis. The following sections summarize the evidence we found
pertaining to the series of predictions about belongingness.

Forming Social Bonds

A first prediction of the belongingness hypothesis is that so-
cial bonds should form relatively easily, without requiring spe-
cially conducive circumstances. Such evidence not only would
attest to the presence and power of the need to belong but would
suggest that the need is not a derivative of other needs (insofar
as it is not limited to circumstances that meet other require-
ments or follow from other events).

There is abundant evidence that social bonds form easily. In-
deed, people in every society on earth belong to small primary
groups that involve face-to-face, personal interactions (Mann,
1980). The anthropologist Coon (1946) asserted that natural
groups are characteristic of all human beings. Societies differ in
the type, number, and permanence of the groups that people
join, but people of all cultures quite naturally form groups.

The classic Robbers Cave study conducted by Sherif, Harvey,
White, Hood, and Sherif (1961/1988) showed that when pre-
viously unacquainted boys were randomly assigned to newly
created groups, strong loyalty and group identification ties en-
sued rapidly. In fact, later in that study, the two strongly opposed
groups were recombined into a single group with cooperative
goals, and emotional and behavioral patterns quickly accom-
modated to the new group (although the prior antagonistic
identifications did hamper the process).

The tendency for laboratory or experimentally created
groups to quickly become cohesive has also been noted in the

minimal intergroup situation (Brewer, 1979). Tajfel and his
colleagues (Billig&Tajfel, 1973; Tajfel, 1970; Tajfel &Billig,
1974; Tajfel, Flament, Billig, & Bundy, 1971) showed that
assigning participants to categories on a seemingly arbitrary
basis was sufficient to cause them to allocate greater rewards
to in-group members than to out-group members. Indeed,
the original goal of Tajfel et al. ( 1 9 7 1 ) was not to study group
formation but to understand the causes of in-group favorit-
ism. To do this, they sought to set up an experimental group
that would be so trivial that no favoritism would be found,
intending then to add other variables progressively so as to
determine at what point favoritism would start. To their sur-
prise, however, in-group favoritism appeared at once, even in
the minimal and supposedly trivial situation (see also Turner,
1985).

This preferential treatment of in-group members does not
appear to be due to inferred self-interest or to issues of novelty
and uncertainty about the task (Brewer & Silver, 1978; Tajfel,
1970; Tajfel & Billig, 1974). Inferred similarity of self to in-
group members was a viable explanation for many of the early
findings, but Locksley, Ortiz, and Hepburn (1980) ruled this
out by showing that people show in-group favoritism even when
they have been assigned to groups by a random lottery. Thus,
patterns of in-group favoritism, such as sharing rewards and
categorizing others relative to the group, appeared quite readily,
even in the absence of experiences designed to bond people to
the group emotionally or materially.

Several other studies suggest how little it takes (other than
frequent contact) to create social attachments. Bowlby (1969)
noted that infants form attachments to caregivers very early in
life, long before babies are able to calculate benefits or even
speak. Festinger, Schachter, and Back (1950) found that mere
proximity was a potent factor in relationship formation; people
seemed to develop social bonds with each other simply because
they lived near each other. Nahemow and Lawton (1975) repli-
cated those findings and also showed that pairs of best friends
who differed by age or race were particularly likely to have lived
very close together, suggesting that extreme proximity may over-
come tendencies to bond with similar others. Wilder and
Thompson (1980) showed that people seem to form favorable
views toward whomever they spend time with, even if these oth-
ers are members of a previously disliked or stereotyped out-
group. In their study, intergroup biases decreased as contact
with members of the out-groups increased (and as in-group
contact decreased).

We noted that the formation of social attachments under ad-
verse circumstances would be especially compelling evidence
because it avoids the alternative explanations based on classical
conditioning (i.e., that positive associations breed attraction).
Latane, Eckman, and Joy (1966) found that participants who
experienced electric shock together tended to like each other
more than control participants who did not experience shock,
although the effect was significant only among firstborns. Ken-
rick and Johnson (1979) found that participants rated each
other more positively in the presence of aversive than nonaver-
sive noise. Elder and Clipp (1988) compared the persistence of
attachments among military veterans and found that the great-
est persistence occurred among groups that had undergone
heavy combat resulting in the deaths of some friends and com-

502 ROY F. BAUMEISTER AND MARK R. LEARY

rades. Although it would be rash to suggest that all shared neg-
ative experiences increase attraction, it does appear that posi-
tive bonding will occur even under adverse circumstances.

The development of interpersonal attraction under fearful
circumstances has been explained in terms of both
misattribution (i.e., people may misinterpret their anxious
arousal as attraction to another person) and reinforcement
theory (i.e., when the presence of some other person reduces
one’s distress, a positive emotional response becomes associ-
ated with that person; Kenrick & Cialdini, 1977). The mis-
attribution explanation is largely irrelevant to the belong-
ingness hypothesis, but the reinforcement explanation is ger-
mane. Specifically, although others may reduce one’s distress
through various routes (such as distraction, humor, or
reassurance), evidence suggests strongly that the mere pres-
ence of other people can be comforting (Schachter, 1959).
Such effects may well be conditioned through years of experi-
ence with supportive others, but they also may indicate that
threatening events stimulate the need to belong.

The fact that people sometimes form attachments with for-
mer rivals or opponents is itself a meaningful indicator of a gen-
eral inclination to form bonds. Cognitive consistency pressures
and affective memories would militate against forming positive
social bonds with people who have been rivals or opponents.
Yet, as we have already noted, the Robbers Cave study (Sherif
et al., 1961 /1988) showed that people could join and work to-
gether with others who had been bitterly opposed very recently,
and Wilder and Thompson (1980) showed that social contact
could overcome established intergroup prejudices and stereo-
types. Orbell, van de Kragt, and Dawes (1988) likewise showed
that impulses toward forming positive attachments could over-
come oppositional patterns. In their study using the prisoner’s
dilemma game, having a discussion period led to decreased
competition and increased cooperation, as a result of either the
formation of a group identity that joined the potential rivals
together or explicit agreements to cooperate. Thus, belong-
ingness motivations appear to be able to overcome some antag-
onistic, competitive, or divisive tendencies.

Similar shifts have been suggested by M. S. Clark (1984,
1986; Clark, Mills, & Powell, 1986; Clark, Ouellette, Powell, &
Milberg, 1987), who showed that people move toward a com-
munal orientation when there is a chance to form a relation-
ship. When participants were confronted with a person who
seemingly would not be amenable to relationship formation
(i.e., because she was already married), they interacted with
her on the basis of norms of equitable exchange and individual-
ity; when they believed she would be a possible relationship
partner, however, they interacted with her on a communal basis
(i.e., mutuality and sharing, without respect to individual eq-
uity concerns).

Critical assessment. The remarkable ease with which social
bonds form has been shown with experimental methods and
confirmed by other methods. The main limitation would be
that people do not always form relationships with all available
or proximal others, which could mean that satiation processes
limit the number of relationships people seek and which also
indicates that other factors and processes affect the formation
of relationships. Some patterns (e.g., in-group favoritism in

minimal groups) have been well replicated with careful efforts
to rule out alternative explanations.

Conclusion. In brief, people seem widely and strongly in-
clined to form social relationships quite easily in the absence of
any special set of eliciting circumstances or ulterior motives.
Friendships and group allegiance seem to arise spontaneously
and readily, without needing evidence of material advantage or
inferred similarity. Not only do relationships emerge quite nat-
urally, but people invest a great deal of time and effort in foster-
ing supportive relationships with others. External threat seems
to increase the tendency to form strong bonds.

Not Breaking Bonds

The belongingness hypothesis predicts that people should
generally be at least as reluctant to break social bonds as they
are eager to form them in the first place. A variety of patterns
supports the view that people try to preserve relationships and
avoid ending them. In fact, Hazan and Shaver (1994a, p. 14)
recently concluded that the tendency for human beings to re-
spond with distress and protest to the end of a relationship is
nearly universal, even across different cultures and across the
age span.

Some relationships are limited in time by external factors,
and so these are logically the first place to look for evidence that
people show distress and resistance to breaking bonds. En-
counter groups and training groups, for example, are often con-
vened with the explicit understanding that the meetings will stop
at a certain point in the future. Even so, it is a familiar observa-
tion in the empirical literature (e.g., Egan, 1970; Lacoursiere,
1980; Lieberman, Yalom, & Miles, 1973) that the members of
such groups resist the notion that the group will dissolve. Even
though the group’s purpose may have been fulfilled, the partici-
pants want to hold on to the social bonds and relationships they
have formed with each other. They promise individually and
sometimes collectively to stay in touch with each other, they
plan for reunions, and they take other steps to ensure a continu-
ity of future contacts. In actuality, only a small minority of these
envisioned reunions or contacts take place, and so the wide-
spread exercise of making them can be regarded as a symptom
of resistance to the threatened dissolution (Lacoursiere, 1980,
p.216).

Other relationships are limited in time by external transitions
such as graduating from college, moving to a different city, or
getting a new job. As such transitions approach, people com-
monly get together formally and informally and promise to re-
main in contact, to share meals or other social occasions to-
gether, to write and call each other, and to continue the relation-
ship in other ways. They also cry or show other signs of distress
over the impending separation (Bridges, 1980). These patterns
seem to occur even if the dissolving relationship (e.g., with
neighbors) had no important practical or instrumental func-
tion and there is no realistic likelihood of further contact.

More generally, many social institutions and behavior pat-
terns seem to serve a need to preserve at least the appearance of
social attachment in the absence of actual, continued interac-
tion. Reunions constitute an occasion for people to see former
acquaintances. The massive exchange of greeting cards during
the Christmas holiday season includes many cases in which the

THE NEED TO BELONG 503

card is the sole contact that two people have had during the
entire year, but people still resist dropping each other’s name
from the mailing list because to do so signifies a final dissolution
of the social bond. In fact, most people will send Christmas
cards to perfect strangers from whom they receive cards (Kunz
& Woolcott, 1976). People seem not to want to risk damaging
a relationship even if they do not know the identity of the other
person!

Likewise, social rituals involving greetings and farewells serve
to assure others of the continuation of one’s relationships with
them. Many greetings, particularly those directed at family
members and close friends, seem designed to indicate that one’s
relationship has remained intact since the last contact, and fare-
wells often include some hint that the relationship will be main-
tained until the people see one another again (Goffman, 1971).
The importance of such rituals in the maintenance of belong-
ingness is reflected in the distress people sometimes experience
when they feel that another’s greeting is inadequately warm or
that the other’s farewell expresses insufficient concern about the
impending separation.

In many cases, people seem reluctant to dissolve even bad or
destructive relationships. The apparent unwillingness of many
women to leave abusive, battering spouses or boyfriends (Roy,
1977; Strube, 1988) has prompted several generations of spec-
ulative explanations, ranging from masochistic or self-destruc-
tive liking for abuse to calculations of economic self-interest
that supposedly override considerations of physical harm. The
belongingness hypothesis offers yet one more potential perspec-
tive: The unwillingness to leave an abusive intimate partner is
another manifestation of the strength of the need to belong and
of the resulting reluctance to break social bonds. The fact that
people resist breaking off an attachment that causes pain attests
to how deeply rooted and powerful the need to belong is.

Moreover, when people do decide to break off an intimate
relationship, they typically experience considerable distress
over the dissolution (which we cover in more detail in the later
section on emotion). This is ironic: Although goal attainment
is usually marked by positive affect such as satisfaction and joy,
attaining the goal of getting a divorce is generally accompanied
by negative affect. To be sure, in some cases the distress over
divorce is accompanied by an admixture of positive affect, but
the negative affect nonetheless indicates the resistance to break-
ing the bond.

It is also relevant and noteworthy that the social bond often
continues despite the divorce. In her study on divorce, Vaughan
(1986) concluded that “in most cases [marital] relationships
don’t end. They change, but they don’t end” (p. 282). Weiss
(1979) also found that some form of (often ambivalent) attach-
ment persists after divorce. The persistence of intimate relation-
ships past the occasion of mutually agreed and formally institu-
tionalized dissolution may be yet another indication of people’s
reluctance to break social bonds.

Critical assessment. Because ethical and practical con-
straints prevent laboratory experimentation on the ending of
significant relationships, the evidence in this section was drawn
from observational studies and other methods, and so the hy-
pothesis of resistance to relationship dissolution is not as con-
clusively supported as might be desired. Alternative explana-
tions exist for some of the findings. For example, the persistence

of relatedness after divorce is partly due to ongoing practical
concerns, such as joint responsibility for child care; although
Vaughan (1986) was emphatic in asserting that such pragmatic
concerns fall far short of explaining the extent of continuing
attachments, she was vague about the evidence to back up her
assertion. Also, as we noted, the tendency for battered women
to return to their abusive partners has been explained in many
ways, and the hypothesized reluctance to break off a relation-
ship is only one of them.

On the positive side, however, the persistence of such bonds
has been observed by a variety of researchers. The fact that
these researchers are from different disciplines suggests that
these conclusions do not stem from a single methodological or
theoretical bias. More systematic research on possible bound-
ary and limiting conditions of the resistance to dissolve bonds
would be desirable.

Conclusion. Despite some methodological weaknesses and
ambiguities, the weight of the evidence does favor the conclu-
sion that people strongly and generally resist the dissolution of
relationships and social bonds. Moreover, this resistance ap-
pears to go well beyond rational considerations of practical or
material advantage.

Cognition

Intelligent thought is generally recognized as the most impor-
tant adaptive trait among human beings, and so it seems rea-
sonable to assume that issues of fundamental concern and im-
portance are likely to be the focus of cognitive activity. The be-
longingness hypothesis therefore would predict that people will
devote considerable cognitive processing to interpersonal in-
teractions and relationships.

Basic patterns of thought appear to reflect a fundamental
concern with social relationships. Sedikides, Olsen, and Reis
(1993) showed that relationships are natural categories; that is,
people spontaneously classify incoming information in terms
of social relationships. Participants stored information about
relationship partners together, and they did this more for strong,
close relationships (marriage) than for weak or distant ones
(e.g., acquaintanceship). Pryor and Ostrom (1981) showed
that people use the individual person as a cognitive unit of anal-
ysis for familiar people more than for unfamiliar people. These
researchers began by questioning the basic assumption that the
person is the fundamental unit of social perception. That is,
information is not necessarily or inherently processed and
stored in memory on a person-by-person basis, but it is, in fact,
processed and stored on such a basis when it pertains to signifi-
cant others. Ostrom, Carpenter, Sedikides, and Li (1993) pro-
vided evidence that information about out-group members
tends to be stored and organized on the basis of attribute cate-
gories (such as traits, preferences, and duties), whereas in-
group information is processed on the basis of person catego-
ries. Thus, social bonds create a pattern in cognitive processing
that gives priority to organizing information on the basis of the
person with whom one has some sort of connection.

Several studies have pursued the notion that people process
information about close relationship partners differently
from the way they process information about strangers or dis-
tant acquaintances. For example, research has shown that,

504 ROY F. BAUMEISTER AND MARK R. LEARY

when a group of people take turns reading words aloud, they
each have high recall for the words they personally speak but
have poor recall for the words preceding and following their
performance. Brenner ( 1 9 7 6 ) found that this next-in-line
effect occurs not only for one’s own performance but also for
words spoken by one’s dating partner (and the words imme-
diately preceding and following).

In a series of studies, Aron, Aron, Tudor, and Nelson (1991)
showed that close relationship partners, unlike strangers, have
cognitive effects similar to those of the self. Thus, when people
form an image of themselves or their mothers interacting with
some object, they have more difficulty recalling that object than
if they imagined a famous but personally unacquainted person
interacting with that same object. In another study, participants
had more difficulty in making me-not me judgments about
traits on which they differed from their spouse than in making
judgments about traits on which they resembled the spouse.
These results suggest that cognitive processes tend to blur the
boundaries between relationship partners and the self, in the
form of “including [the] other in the self” (p. 241). In short,
these studies confirm that information about relationship part-
ners is singled out for special processing, and they raise the pos-
sibility that the need to belong leads to a cognitive merging of
self with particular other people. Such patterns of subsuming
the individual in the interpersonal unit indicate the importance
of these relationships.

Many of the special biases that people exhibit for processing
information in ways that favor and flatter themselves are ex-
tended to partners in close relationships. Fincham, Beach, and
Baucom (1987) showed that self-serving biases that take credit
for success and refuse blame for failure operate just as
strongly—or even more strongly—when people interpret their
spouses’ outcomes as when they interpret their own outcomes.
That is, events are interpreted in a way that is maximally flat-
tering to the spouse, just as they are interpreted in ways that
enhance and protect the self. (These patterns are extended only
to partners in good, strong, happy relationships, however; high
marital distress is correlated with a breakdown in these partner-
serving attributions.)

Likewise, the “illusion of unique invulnerability” (Perloff
& Fetzer, 1986) turns out not to be as unique as first thought.
Although people are more extremely and unrealistically
optimistic about themselves than about some vague target
such as the average person, they are equally optimistic about
their closest friends and family members. That is, they think
that bad things are not as likely to happen either to them-
selves or to their close friends as to strangers or to a hypothet-
ical average person.3 Along the same lines, Brown (1986)
showed that people (particularly those with high self-esteem)
tend to extend self-serving biases to their friends. Specifically,
people rate both self and a same-sex friend more favorably
than they rate people in general.

Group memberships also appear to exert important influ-
ences on cognitive patterns. People expect more favorable and
fewer objectionable actions by their in-group than by out-group
members, and these expectations bias information processing
and memory, leading people to forget the bad things (relative to
good things) that their fellow in-group members do (Howard &
Rothbart, 1980). People also make group-serving or “sociocen-

tric” attributions for the performance of the groups to which
they belong. Members of a successful group may make group-
serving attributions that put the entire group in a good light,
whereas, after failure, group members may join together in ab-
solving one another of responsibility (Forsyth & Schlenker,
1977; Leary & Forsyth, 1987; Zander, 1971).

Linville and Jones (1980) showed that people tend to process
information about out-group members in extreme, black-and-
white, simplistic, polarized ways, whereas similar information
about members of their own group is processed in a more com-
plex fashion. Thus, the mere existence of a social bond leads to
more complex (and sometimes more biased) information
processing.

Of broader interest is evidence that belongingness can affect
how people process information about nearly all categories of
stimuli in the social world. Wegner (1986) noted the irony that
traditional theories of the “group mind” tended to assume that
all members would essentially think the same thing, because
much more far-reaching advantages could be realized through
a group mind if each member was responsible for different in-
formation, thereby enabling the group to process considerably
more information than any one person could. Wegner went on
to propose that transactive memory processes operate in close
relationships and groups by assigning each person a significant
category of expertise, with the result that each person becomes
expert in one or a few areas and simply consults others when
alternative areas come up. An empirical study conducted by
Wegner, Erber, and Raymond (1991) supported the transactive
memory hypothesis by showing that partners in close relation-
ships apparently have established procedures for determining
which person should remember which information. Partici-
pants were people in dating couples who were paired either with
their partner or with a stranger. The preexisting couples showed
better memory for experimental stimuli than the impromptu
assigned couples, except when the researchers assigned people
at random to be the expert responsible for various categories
of stimuli. In this latter condition, apparently, the assignment
disrupted the couples’ preexisting system and hence impaired
the processing of information.

Another broad and very basic issue is how often interpersonal
belongingness is used as an interpretive category. C. A. Ander-
son (e.g., 1991) sought to establish the fundamental dimensions
people use for making attributions about the causes of events.
His study coded participants’ attributional activity along 13 di-
mensions, including all of the ones featured in the major attri-

3 Perloff and Fetzer (1986) favored an interpretation for their results
in terms of the vagueness of the comparison target over the motivational
explanation that people want to regard their closest relationship part-
ners as equally invulnerable (equal to themselves). Their discrimina-
tion between the two hypotheses rested on the “one of your friends”
condition in their second study: They found that the “closest friend”
was seen as being highly invulnerable, whereas when participants chose
one of their other friends, this person was seen as more vulnerable. Their
findings suggested that participants in that condition selected a friend
who seemed most likely to have the problem asked about, so it is diffi-
cult to evaluate the motivational hypothesis. Thus, the interpretation
emphasized here is consistent with all of Perloff and Fetzer’s findings, as
they acknowledged, even though their own interpretations tended to
favor explanation in terms of vague versus specific targets.

THE NEED TO BELONG 505

butional theories (e.g., locus, stability, globality, and
controllability). To his surprise, however, the strongest single
dimension was what he called interpersonalness, which was de-
nned as the degree to which the causes of the focal event re-
flected on the relationship between the individual attributor and
other people (e.g., doing something because one is married).
Thus, although interpersonalness was not a central concern of
his investigation (because the major attribution theories had
largely ignored it), it emerged as a major dimension in the way
people normally think about and interpret the causes of events.
The unexpected emergence of interpersonalness as a powerful
fundamental dimension of causal attribution is consistent with
the view that belongingness is one of humanity’s basic concerns.

Thus far we have provided evidence that interpersonal rela-
tionships are centrally important in the way people think. Ad-
ditional predictions about cognitive activity can be derived from
the belongingness hypothesis. Although the evidence is consis-
tent with these predictions, it tends to be subject to alternative
explanations based on short-term, pragmatic concerns, so it is
less compelling for present purposes. We include brief coverage
for the sake of thoroughness.

Clearly, one would predict, on the basis of a need to belong,
that people should tend to think particularly about actual and
potential relationship partners more than about other people.
This would be reflected in increased cognitive processing
caused by the expectation of future or further interactions, be-
cause these conditions hold the possibility of forming a relation-
ship. Devine, Sedikides, and Fuhrman (1989) confronted par-
ticipants with advance information about various stimulus per-
sons and found that this information received more thorough
and detailed processing when it pertained to a future interaction
partner. Monson, Keel, Stephens, and Genung (1982) found
that people made more extreme—and more valid—trait attri-
butions from identical information when it pertained to a fu-
ture interaction partner than when it pertained to someone with
whom no interaction was anticipated. Erber and Fiske (1984)
showed that interpersonal dependency (outcome dependency)
overcame the usual tendency to ignore information that runs
counter to expectations. When participants were outcome de-
pendent on the confederate, they paid extra attention to incon-
sistent information about the confederate and seemed to think
more in terms of dispositional attributions about the partner.
Thus, belonging to another person changes the way one pro-
cesses information about that person.

Some of these interaction effects could be interpreted as
guided by short-term concerns. Still, the prospect of forming a
relationship with a recently met person appears to be sufficient
to alter the way people process the interaction. M. S. Clark
(1984) showed that people keep track of information differently
when the interaction partner is a potential relationship partner.
Furthermore, recent work by Tice, Butler, Muraven, and Still-
well (1994) showed that when people were interacting with
friends as opposed to strangers, they changed the way they pre-
sented information about themselves (i.e., they became more
modest). Moreover, the way they encoded and recalled the in-
teraction depended on the relationship: Memory was best if one
had been modest with friends or self-enhancing with strangers,
and otherwise it was impaired.

Critical assessment. The evidence that interpersonal con-

cerns affect cognitive processing is methodologically strong and
extensive. A broad variety of experimental procedures has been
involved in demonstrating such effects. For present purposes,
the main critique would be that some of the studies have not
been directly concerned with close relationships. Some have
shown that the expectation of interaction with a stranger or new
acquaintance is sufficient to alter cognitive processing. Al-
though it is reasonable to infer that people regard meeting new
people as the first step in possible relationship formation
(perhaps especially among the young adult populations who
constitute most of the experimental samples), this inference re-
quires further validation before one can have full confidence in
interpreting those findings as evidence for the need to belong,
because short-term concerns of practical or material advantage
may also play a role in some such situations.

Nonetheless, many of the findings reviewed in this section do
pertain to close relationship partners, and there is evidence that
information pertaining to interaction partners is processed
differently depending on its relevance to lasting relationships. It
is thus quite clear that relatedness affects cognitive processing;
only the extent of that influence and some of its processes are
still open to debate.

Conclusion. Concern with belongingness appears to be a
powerful factor shaping human thought. People interpret situa-
tions and events with regard to their implications for relation-
ships, and they think more thoroughly about relationship (and
interaction) partners than about other people. Moreover, the
special patterns of processing information about the self are
sometimes used for information about relationship partners as
well. Thus, both actual and potential bonds exert substantial
effects on how people think.

Emotion

The main emotional implication of the belongingness hy-
pothesis is that real, potential, or imagined changes in one’s be-
longingness status will produce emotional responses, with posi-
tive affect linked to increases in belongingness and negative
affect linked to decreases in it. Also, stable or chronic conditions
of high belongingness should produce a general abundance of
positive affect, whereas chronic deprivation should produce a
tendency toward abundant negative affect.

Positive affect. In general, the formation of social bonds is
associated with positive emotions. Perhaps the prototype of re-
lationship formation is the experience of falling in love, which
is typically marked by periods of intense bliss and joy, at least
if the love is mutual (e.g., Steinberg, 1986). When love arises
without belongingness, as in unrequited love, the result is typi-
cally distress and disappointment (Baumeister & Wotman,
1992). Belongingness is thus crucial if love is to produce bliss.

Likewise, occasions such as new employment, childbirth, fra-
ternity or sorority pledging, and religious conversion, all of
which are based on the entry into new relationships and the
formation of new social bonds, are typically marked by positive
emotions and celebrated as joyous. Childbirth is especially sig-
nificant in this regard because the data show that parenthood
reduces happiness and increases stress, strain, and marital dis-
satisfaction (e.g., S. A. Anderson, Russell, & Schumm, 1983;
Campbell, Converse, & Rodgers, 1976; Glenn & McLanahan,

506 ROY F. BAUMEISTER AND MARK R. LEARY

1982; for reviews, see Baumeister, 1991; Bernard, 1982; Camp-
bell, 1981; Spanier & Lewis, 1980), yet people nonetheless re-
tain a positive image of it, celebrate it, and feel positive about
it, both in advance and in retrospect. It is plausible that the
formation of the new social bond is directly responsible for the
joy and positive feelings, whereas the negative aspects and feel-
ings associated with parenthood arise indirectly from the has-
sles, conflicts, and stresses that accompany the social bond.

If the formation of bonds is one occasion for joy, a second
occasion comes when the bond is formalized into a more recog-
nizably permanent status. A wedding, for example, does not
create a new relationship, at least in modern Western cultures,
because the bride and groom typically have known each other
intimately for some time. The wedding does, however, signify an
increase in commitment to maintaining the relationship per-
manently, and the joyful celebration of the wedding can be re-
garded as an affective consequence of solidifying the social
bond. It is noteworthy that many traditional wedding vows in-
clude an actuarially implausible pledge that the marriage will
never end (“till death do us part”). In essence, such vows are
an institutionalized mechanism for committing people to meet
their spouse’s belongingness needs.

Although we have emphasized the view of affect as a result
of attachment, positive affect may in turn help solidify social
attachment. Probably the most influential view of this sort was
developed by Shaver et al. (1988), who portrayed romantic love
as a kind of glue designed by nature to solidify the attachment
between two adults whose interaction is likely to lead to parent-
ing. In their view, love elaborates on sexual attraction in a way
that will hold the couple together when their sexual intercourse
leads to reproduction. Along the same lines, various studies
have found that positive affective experiences increase attrac-
tion and solidify social bonds (L. A. Clark & Watson, 1988;
Gouaux, 1971; May & Hamilton, 1980; Veitch & Griffitt,
1976). Moreland (1987) concluded that the development of
shared emotions is one of the principal causes of the formation
of small groups.

More generally, happiness in life is strongly correlated with
having some close personal relationships. Research suggests
that it does not seem to make a great deal of difference what sort
of relationship one has, but the absence of close social bonds is
strongly linked to unhappiness, depression, and other woes
(e.g.,Argyle, 1987; Freedman, 1978; Myers, 1992). People with
high levels of intimacy motivation tend to enjoy higher levels
of happiness and subjective well-being (McAdams & Bryant,
1987), which is likely a result of their tendency to form and
maintain a rich network of friendships and other social bonds
(McAdams, 1985). Having some intimate bond appears to be
important and perhaps even necessary for happiness. Social iso-
lation is practically incompatible with high levels of happiness.

Negative affect. Threats to social attachments, especially
the dissolution of social bonds, are a primary source of negative
affect. People feel anxious at the prospect of losing important
relationships, feel depressed or grief stricken when their con-
nections with certain other people are severed, and feel lonely
when they lack important relationships (Leary, 1990; Leary &
Downs, in press; Tambor & Leary, 1993).

Anxiety is often regarded as the extreme or prototype of neg-
ative affect, and it is clearly linked to damaged, lost, or threat-

ened social bonds. In fact, social exclusion may well be the most
common and important cause of anxiety (Baumeister & Tice,
1990). Horney (1945) identified the source of “basic anxiety”
as the feeling of “being isolated and helpless in a potentially
hostile world” (p. 41); of course, that formula mixes two
different sources, insofar as isolation is a function of the belong-
ingness need, whereas helplessness is a frustration of control
(which is probably another fundamental motivation). Anxiety
and general distress seem to be a natural consequence of being
separated from important others. Children as young as 1 year
old show extreme distress—separation anxiety—on being sep-
arated from their mothers (Bowlby, 1973), and adults show
similar reactions when they must leave loved ones for an ex-
tended period of time. Furthermore, people’s memories of past
rejections are tainted with anxiety (Tambor & Leary, 1993),
and even just imagining social rejection increases physiological
arousal (Craighead, Kimball, & Rehak, 1979).

Consistent with the social exclusion theory of anxiety,
Barden, Garber, Leiman, Ford, and Masters (1985) found that
anxiety ensues if people are excluded from social groups, but
experiences of social inclusion appear to counteract the effects
of exclusion and remove the anxiety. Mathes, Adams, and Da-
vies (1985) predicted that a threat to self-esteem would mediate
the link between jealousy and anxiety, but their results did not
support their hypothesis. Instead, they found that the loss of
relationship led directly to anxiety.

Like anxiety, depression may be precipitated by a variety of
events, but failing to feel accepted or included is certainly one
of them. Both general depression and social depression (i.e.,
dysphoria about the nature of one’s social relationships) are
inversely related to the degree to which one feels included and
accepted by others (Tambor & Leary, 1993). Hoyle and Craw-
ford (in press) found that both depression and anxiety were sig-
nificantly correlated (negatively) with students’ sense of belong-
ing to their university.

Jealousy is another negative affective state that is a com-
mon response to threats to one’s relationships. Pines and Ar-
onson (1983) reported that, in a series of surveys, some ex-
perience of jealousy was essentially universal, in the sense
that everyone experiences it sooner or later. Moreover, more
than half of their respondents described themselves as being
“a jealous person” and correctly estimated that slightly more
than half of the other participants would respond in that
same way; however, they also said that the true incidence of
jealous people was even higher, because some jealous people
deny their jealousy. Pines and Aronson emphasized that
“feeling excluded” is a major cause of jealousy.

Regarding jealousy, perhaps the most relevant finding for our
purposes was that of Reiss( 1986), who concluded that jealousy
is cross-culturally universal. Reiss carefully investigated the ex-
travagant claims made by some observers and anthropologists
that, in certain cultures, people are able to exchange sexual
partners and intimate partners without any possessiveness or
jealousy, and in every case the claim turned out to be unwar-
ranted. Cultures may indeed vary as to which particular actions
or signs of affection are regarded as justifying jealous reactions,
and they may differ in how people express their jealousy, but
sexual jealousy is found in all cultures.

Loneliness reflects “an individual’s subjective perception of

THE NEED TO BELONG 507

deficiencies in his or her social relationships” (Russell, Cu-
trona, Rose, & Yurko, 1984, p. 1313). In other words, people
feel lonely when their belongingness needs are being insuffi-
ciently met. Moreover, it appears that belongingness, rather
than mere social contact, is the crucial factor. Mere social con-
tact does not, by itself, buffer people against loneliness. Lonely
and nonlonely people do not differ markedly in the amount of
time they spend with other people. However, lonely people
spend less time with friends and family—those who are most
likely to fulfill their needs to belong—than nonlonely people
(Jones, 1981). Furthermore, loneliness is much more strongly
related to one’s sense of social isolation than to objective in-
dexes of one’s social network, such as one’s sheer number of
friends (Williams & Solano, 1983). In one study, the correla-
tion between self-reported loneliness and the degree to which
people felt included and accepted by others was found to be
-.71 (Spivey, 1990). Generally, loneliness seems to be a matter
more of a lack of intimate connections than of a lack of social
contact (Reis, 1990; Wheeler, Reis, & Nezlek, 1983).

Yet another highly aversive emotional state is guilt. Despite a
long tradition of analyzing guilt in terms of self-evaluation accord-
ing to abstract moral standards, recent work has increasingly em-
phasized the interpersonal structure of guilt (Baumeister, Stillwell,
& Heatherton, 1994; Cunningham, Steinberg, & Grev, 1980;
Jones & Kugler, in press; Jones, Kugler, & Adams, 1995; Miceli,
1992; Tangney, 1992). Empirical studies of how people induce
guilt in others have found that such inductions are almost entirely
confined to close interpersonal relationships and that a major rea-
son for inducing guilt is to cause one’s partner to exert himself or
herself more to maintain the interpersonal relationship (e.g., by
spending more time with or paying more attention to oneself;
Baumeister, Stillwell, & Heatherton, in press; Vangelisti, Daly, &
Rudnick, 1991). Many episodes of guilt can thus be understood as
responses to disturbances or threats to interpersonal attachments.

Two specific events that thwart people’s need to belong are
divorce and death. Divorce is not generally recognized as an
occasion for joyful celebration, even if the divorce was desired
more fervently than the wedding had been. Divorce produces
varied forms of distress, including anger, depression, desolation,
and loneliness, in nearly everyone. Weiss (1979) concluded that
some “emotional upset. . . appears to be a nearly inevitable
accompaniment to marital separation” (p. 210) and is found
“even though [the] marriage had become unhappy” (p. 202).
In contrast, Spanier and Casto (1979) and Goode (1956) did
find a minority of participants who reported relatively little dis-
tress in response to divorce. Spanier and Casto (1979) thought
that one possible explanation for the discrepancy was that their
single interview technique (in contrast to Weiss’s multiple
sessions) was less sensitive to some deep or occasional feelings.
Consistent with this, they concluded that certain forms of dis-
tress, such as regret, yearning, and bitterness, “actually may in-
crease over time” (p. 226). Price and McKenry (1988) sug-
gested another reason that one-time measures may fail to find
universal distress after divorce: Many couples may have passed
through the most distressing phase before the researchers col-
lect their data.

Spanier and Casto (1979) listed the emotional turmoil after
divorce as mixed from among

feelings about the (former) spouse, such as love, hate, bitterness,
guilt, anger, envy, concern, and attachment; feelings about the mar-
riage, such as regret, disappointment, bitterness, sadness, and fail-
ure; and more general feelings, such as failure, depression, eupho-
ria, relief, guilt, lowered self-esteem, and lowered self-confidence.
( p . 2 1 3 )

Price and McKenry (1988) summarized the common emo-
tional reactions to divorce as including “extreme stress, includ-
ing feelings of rejection, depression, hostility, bitterness, loneli-
ness, ambivalence, guilt, failure, confusion, disorganization,
and sometimes relief” (p. 42). It is clear that plenty of negative
affect accompanies divorce.

Perhaps the strongest emotional reactions human beings ex-
perience involve death, both the death of oneself and the death
of other people. The death of a spouse, child, or close friend
ranks among the most stressful events that people experience
(T. H. Holmes & Rahe, 1967; Weiss, 1979). Grief often takes
the form of an especially severe depression. Some conceptual-
izations of grief portray it not as a reaction to the loss of the
person but as a reaction to the loss of a linkage with another
person (Lofland, 1982). It is interesting that people even grieve
deeply over the death of spouses with whom they had had trou-
bled marriages. As Weiss (1979) observed, “Apart from minor
variations,. . . nearly disabling grief was the rule, even among
individuals who could say about the preceding marriage, as one
widow did, ‘Ours wasn’t the best marriage in the world'” (p.
202).

Anxiety about death, whether of oneself or others, can be re-
garded as stemming (at least in part) from a threat to belong-
ingness (Baumeister & Tice, 1990). As Lofland (1982) pointed
out, when people die, relationships end. Along these lines,
Conte, Weiner, and Plutchik (1982) linked death anxiety to fear
of loneliness (see also Mijuskovic, 1980). In a study of death
anxieties, Bednarski and Leary (1994) found that a primary
basis of people’s fears about death involved concerns with being
separated from friends and family. These interpersonal con-
cerns appeared to be a more important source of death anxiety
than fears about no longer existing or uncertainty about what
happens after death. This link between death anxiety and sepa-
ration anxiety may explain why most positive depictions of life
after death have emphasized togetherness with family and loved
ones, with a broad community of like-minded believers, with a
loving deity, or with all of the above (e.g., Baumeister, 1991). If
death anxiety is rooted in threats to belongingness and social
inclusion, then fears of death can best be soothed by emphasiz-
ing that death will involve a continuation or even an improve-
ment in one’s belongingness status.

Indirect effects. Although we have emphasized emotional
consequences of changes in belongingness, there may also be
indirect ways in which belongingness affects emotion. As shown
earlier with cognitive processes, emotional processes may
change when the situation involves a close friend or intimate
partner. Tesser (1991) has reviewed a number of such effects.
The main implication is that emotional responses to the relative
outcomes of self and other depend heavily on whether the other
person is a close relationship partner such as a good friend.
When the performance involves a domain that is important to
the self, it is upsetting to be outperformed by another person,
and the emotional distress is magnified if the other person is a

508 ROY F. BAUMEISTER AND MARK R. LEARY

close friend (see also Tesser, Millar, & Moore, 1988). In con-
trast, if the performance involves some ability that is not im-
portant to self-definition, then superior performances by
friends (but not strangers) produce positive affect. One key
difference is what Tesser (1991) called the reflection process:
The positive achievements of one’s relationship partners reflect
favorably on the self (as long as they do not make the self look
bad by comparison in some important way). Similar achieve-
ments by strangers do not reflect on the self, of course, and so
they do not produce positive affect. Meanwhile, it appears that
the positive accomplishments of close others in domains rele-
vant to one’s own identity have a special capacity to generate
distress by threatening one’s cherished views of one’s own im-
portant abilities. Thus, the existence of a close relationship with
another person changes the way one responds emotionally to
that person’s performance outcomes in complex but predict-
able ways.

Critical assessment. The evidence reviewed in this section
was drawn from sociology, anthropology, and several subfields
of psychology, and it is based on a variety of methods including
surveys, observational studies, cross-cultural comparisons, au-
tobiographical narratives, and experiments. Although several
of these methods are generally regarded as less conclusive than
experimentation, the consistency of the conclusion across
multiple methodologies is itself a source of confidence. Thus,
for example, one could dispute Pines and Aronson’s (1983) de-
termination about the pervasiveness of jealousy by noting that
their sample was possibly skewed to include a high proportion
of people who were interested in jealousy, but the very high
(indeed, universal) incidence of jealousy across different cul-
tures, as attested by Reiss’s (1986) review, makes it seem un-
likely that Pines and Aronson were wrong in concluding that
jealousy is very common.

Probably the greatest ambiguity in this section’s evidence at-
taches to the discussion of death. To be sure, it is implausible to
dispute that emotional distress very typically attends the death
of a loved one or relationship partner. Still, there are alternate
explanations for this distress that could possibly dispense with
the need to belong. A partner’s death may have effects on mate-
rial and pragmatic concerns (e.g., loss of income), may create
distressing uncertainty about one’s own future, may affect the
self-concept, and may activate worries about one’s own death.

Conclusion. Many of the strongest emotions people experi-
ence, both positive and negative, are linked to belongingness.
Evidence suggests a general conclusion that being accepted, in-
cluded, or welcomed leads to a variety of positive emotions
(e.g., happiness, elation, contentment, and calm), whereas be-
ing rejected, excluded, or ignored leads to potent negative feel-
ings (e.g., anxiety, depression, grief, jealousy, and loneliness).
The near universality of distress associated with divorce and be-
reavement is consistent with the belongingness hypothesis; in-
deed, there is no firm evidence in those literatures that signifi-
cant social bonds can ever be broken without suffering or dis-
tress, even though (as noted) not every recently divorced or
bereaved person will necessarily be suffering acutely when the
interviewer happens to call.

Although the evidence was not equally abundant or equally
strong for all emotions, the consistency across multiple emo-
tions was impressive. It seems quite safe to conclude that both

positive and negative emotional reactions are pervasively linked
to relationship status. The existence of an interpersonal bond
changes the way one responds emotionally to the performances
and actions of a relationship partner and indeed intensifies
many emotional reactions. Moreover, actual or possible changes
in belongingness status constitute an important cause of emo-
tions. The evidence is sufficiently broad and consistent to sug-
gest that one of the basic functions of emotion is to regulate
behavior so as to form and maintain social bonds.

Consequences of Deprivation

The general argument is that deprivation of belongingness
should lead to a variety of affiliative behaviors and cause various
undesirable effects, including decrements in health, happiness,
and adjustment. We have already documented (in the preceding
section) that loss of social bonds causes emotional distress,
which is sufficient to show that belongingness is something peo-
ple want. To regard it as a need, however, it is necessary to show
effects that go beyond mere frustration and emotional distress.

Considerable research shows that people who do not have ad-
equate supportive relationships experience greater stress than
those who do. In part, this is because having other people avail-
able for support and assistance can enhance coping and provide
a buffer against stress. However, evidence suggests that simply
being part of a supportive social network reduces stress, even
if other people do not provide explicit emotional or practical
assistance (Cohen & Wills, 1985). Although this finding has
been interpreted in terms of the stress-reducing effects of social
support, an equally plausible explanation is that the deprivation
of the need to belong is inherently stressful.

Direct evidence that deprivation of belongingness is maladap-
tive was provided by DeLongis, Folkman, and Lazarus (1988).
They found that happily married couples were less likely to ex-
perience psychological and somatic health problems, both on
and after stressful days, than other participants. Medical re-
search has suggested that these beneficial effects extend beyond
mere health complaints. Lynch (1979) summarized the evi-
dence from many studies by stating that “U.S. mortality rates
for all causes of death . . . are consistently higher for divorced,
single, and widowed individuals” than for married individuals
(p. 38). Lynch’s own data showed the greater incidence of fatal
heart attacks among unattached individuals than among mar-
ried people, but he noted that similar effects can be found for
tuberculosis, cancer, and many other illnesses, as well as overall
patterns. Of course, there are multiple possible explanations for
such an effect that might have nothing to do with belongingness,
but efforts to control for these variables have often found a per-
sistent, independent, robust effect of social relations. Goodwin,
Hunt, Key, and Samet (1987) found that married participants
survived cancer better than single ones even after the timing of
diagnosis, likelihood of receiving treatment, and cigarette
smoking had been controlled, and they cited other evidence that
the effect remains after family income has been controlled.

Indeed, being deprived of belongingness may have direct
effects on the immune system. Kiecolt-Glaser, Garner, et al.
(1984) found that loneliness was associated with a decrease in
immunocompetence, specifically in natural killer cell activity,
and this effect was independent of changes in perceived distress.

THE NEED TO BELONG 509

Kiecolt-Glaser, Ricker, et al. (1984) replicated this effect and
also found elevated urinary cortisol levels among lonely partic-
ipants. Kiecolt-Glaser et al. (1987) found poorer immune func-
tion on several measures among women suffering from marital
disruption, including divorce, separation, and unhappy
marriage.

The effects of belongingness on mental illness parallel those
on physical illness. Rejected children have a higher incidence of
psychopathology than other children (Bhatti, Derezotes, Kim,
& Specht, 1989; Hamachek, 1992). Children who grow up
without receiving adequate attention from caregivers show
emotional and behavioral pathologies, as demonstrated experi-
mentally by Harlow, Harlow, and Suomi (1971) with animals
and as corroborated by observations of human children by
Bowlby(1969,1973;seealsoRutter, 1979).4

Marital status also has strong correlations with mental ill-
ness. Bloom, White, and Asher (1979) reviewed the literature
and concluded that, in all studies, mental hospital admission
rates are highest among divorced and separated people, inter-
mediate among never-married people, and lowest among mar-
ried people. In fact, as measured by admissions to mental
hospitals,5 mental illness is at least 3 and possibly up to 22 times
higher among divorced people than among married people.

Even problems that might at first seem unrelated to social
interaction and relationships are sometimes found to have so-
cial deprivation or failed belongingness as an underlying cause.
Problems with attachment have been identified as a major fac-
tor in eating disorders. Sours (1974), for example, noted that
patients with eating disorders tended to have been (as children)
overly sensitive to separation from their mothers. Armstrong
and Roth (1989) found that women with eating disorders had
significantly more intense and severe separation and attach-
ment difficulties than a normal comparison group.

Combat-related stress is also moderated by belongingness.
Veterans who perceive that they have a high degree of social
support are significantly less likely to experience post-traumatic
stress disorder than those who have lower perceived support
(Hobfall & London, 1986; Solomon, Waysman, & Mikulincer,
1990). In fact, the authors of one study concluded that loneli-
ness “is the most direct antecedent of psychopathology and so-
cial dysfunction” in combat stress reactions (Solomon et al.,
1990, p. 468).

Crime may also be affected by belongingness. Sampson and
Laub (1993) showed that having a good marriage and a stable
job each had a strong negative effect on adult crime, consistent
with other evidence. Other evidence suggests that social bonds
to other criminals or to criminal groups may foster crime. Re-
cent news coverage of gangs has repeatedly suggested that a need
to belong attracts unattached young people to join violent
gangs, which tend to serve as a surrogate “family” (Olmos,
1994; cf. Jankowski, 1991). Sampson and Laub likewise found
that having social relationships with delinquent peers was one
of the strongest independent predictors of juvenile delinquency,
consistent with plenty of previous evidence. They did, however,
caution that this well-established link is based on largely corre-
lational data and that ambiguities about the direction of causal-
ity remain to be addressed. Still, for present purposes, the link
is important evidence that belongingness needs are important
among deviants, regardless of whether the link arises because

having delinquent peers causes delinquent activity or because
delinquent activity leads to bonding with delinquent peers.

Meanwhile, in laboratory experimentation, Geis and Moon
(1981) sought to involve participants in lying, cheating, and
stealing at the behest of an assigned group partner (a
confederate). They found that 67% of a sample of college stu-
dents acquiesced in an act of cheating and in a monetary theft
by their partner and that they actively lied to conceal the theft.
Thus, it appears that even recently formed group bonds may be
strong enough to overcome some salient prohibitions of tradi-
tional morality. (It is noteworthy that the group loyalty in that
study may have been intensified by the presence of a hostile rival
group.) More extreme versions of the phenomenon of going
along with objectionable actions by fellow group members be-
cause of loyalty have been commonly observed as central fac-
tors in group violence, such as spontaneous atrocities commit-
ted by the Ku Klux Klan (Wade, 1987), Nazi police guards
(Browning, 1992), and others (Staub, 1989; see also Groth,
1979, on gang rape).

The relevance of belongingness to suicide was suggested
nearly a century ago by Durkheim (1897/1963). His seminal
work proposed that suicide could be explained as a result of a
failure of social integration. People who are well integrated into
society by multiple and strong relationships are unlikely to
commit suicide, whereas unintegrated people are much more
likely to kill themselves. Durkheim’s hypothesis has held up far
better than most social science hypotheses over the decades, and
the evidence continues to show that a lack of social integration
increases the likelihood of suicide (Trout, 1980). For example,
single, divorced, and widowed people are more likely to commit
suicide than married people (e.g., Rothberg & Jones, 1987).
Those who are unemployed have a higher suicide rate than
those who are employed. People who belong to subcultural
groups that are shrinking have increased suicide rates. People
who work in occupations that are shrinking are also more likely
than others to commit suicide. Indeed, the main criticism that
can be leveled against Durkheim’s hypothesis is that it is incom-
plete in the sense that it does not explain everything about sui-
cide (e.g., Baumeister, 1990; Douglas, 1967), but it is correct as
far as it goes. For present purposes, the important point is that
strong social ties are associated with a lower risk of suicide,
probably because such ties help restrain people from killing
themselves.

Social support research is relevant to the belongingness hy-
pothesis because social support is based on relationships and

4 Several studies have shown that physically unattractive people have
a higher incidence of psychopathology than attractive people (e.g., Ba-
rocas & Vance, 1974; Cash, 1985; Farina, Burns, Austad, Bugglin, &
Fischer, 1986; O’Grady, 1989). One reason may be that they tack be-
longingness, because society tends to reject unattractive individuals
(Berscheid & Walster, 1974).

5 Admittedly, hospital admissions is an imprecise measure. One
might object that married people can stay out of institutions because
they have someone at home to take care of them. On the other hand,
many people are admitted to such institutions at the behest of family
members, and so one could argue that the true difference is even larger.
Given the size and consistency of the effect, it seems reasonable to con-
clude that marital status is related to mental illness, although further
and methodologically better evidence is needed.

510 ROY F. BAUMEISTER AND MARK R. LEARY

positive interactions with others, and so any benefits of such
support would constitute further confirmation of the belong-
ingness hypothesis. The benefits of social support appear to be
well established. Thus, for example, Cohen, Sherrod, and Clark
(1986) showed that the availability of social support—which
can be restated as the existence of social bonds—buffers people
against the ill effects of stress. Cutrona (1989) showed that so-
cial support reduced depression during pregnancy and postpar-
tum depression among adolescent girls. Responding to method-
ological criticisms that had attacked social support research as
merely self-report bias, Cutrona’s study included ratings of each
girl’s support network by an adult informant who knew the girls
well, and these external informants’ ratings predicted health
outcomes (in some cases, even better than the girls’ own ratings
of their support). Thus, the benefits of belongingness in coping
with major life stress appear to go beyond mere self-report bias.

Older adults who have a close, intimate friend (i.e., a
“confidant”) maintain higher morale in the face of life stresses
such as retirement and spousal death than individuals who lack
such a relationship. For example, Lowenthal and Haven (1968)
found that widows who have a confidant have been found to
be only slightly more depressed than married women, whereas
those without a confidant have been found to be much more
dysphoric. These researchers also found that the majority of
older adults who recently lost a confidant were depressed, but
the majority who currently had a confidant were satisfied.

Rook (1987b) distinguished between social support and
companionship. Social support was in this case rather narrowly
interpreted in terms of direct help, whereas companionship
meant the expressive aspects of social interaction. Both were
found to be important and beneficial, but companionship may
be the more important of the two, especially for psychological
well-being, social satisfaction, and coping with minor stress.
These data are particularly important for the relevance of social
support research to the belongingness hypothesis because one
could conceivably argue that belongingness per se is irrelevant
and that the practical, material help that people derive from
their social networks is solely responsible for the benefits of so-
cial support. Rook’s data suggested, on the contrary, that the
practical help is secondary (except in extreme circumstances
in which major assistance is needed), whereas belongingness is
highly beneficial by itself.

Perhaps most generally, general well-being and happiness in
life depend on having some close social ties. Social isolation is
strongly related to various patterns of unhappiness (for reviews,
see Argyle, 1987; Baumeister, 1991; Freedman, 1978; Myers,
1992). Indeed, Baumeister (1991) noted that it is about the
only objective factor that shows a substantial correlation with
subjective well-being. Happiness also appears to be fairly stable
across time and circumstance (e.g., Costa, McCrae, & Zonder-
man, 1987), leading many to conclude that it is linked to per-
sonality factors. The broad trait of extraversion appears to be
strongly related to happiness and positive affectivity (see Costa
& McCrae, 1980, 1984), and extraversion encompasses several
factors, such as sociability, gregariousness, warmth, and social
involvement, that seem likely to enhance the tendency to form
and maintain social ties. Moreover, belongingness appears to
be sufficient to overcome the relative deficit in happiness that
introverts suffer. Hotard, McFatter, McWhirter, and Stegall

(1989) found that introverts who have a good network of social
relationships are just as happy as extraverts. Thus, introverts’
deficit in happiness may be a result of their experiencing less
belongingness.

Further support for the importance of belongingness to psy-
chological well-being is provided by the fact that the psychother-
apeutic process is facilitated by close personal bonds. Numer-
ous therapeutic orientations stress the importance of the rela-
tionship between the therapist and the client. Rogers (1959),
for example, urged psychotherapists to display a willingness to
accept and support the client regardless of his or her behavior or
contribution to the relationship. Such “unconditional positive
regard” is perhaps the ultimate way to fulfill another person’s
belongingness needs. From the standpoint of the belongingness
hypothesis, however, the essential ingredient in client-centered
therapy is not unconditional positive regard (i.e., appraisal) but
unconditional social acceptance (i.e., belongingness).6

Furthermore, some have suggested that one goal of psycho-
therapy should be to enhance clients’ ability to elicit social sup-
port in their everyday lives (Brehm, 1987). To the extent that
people who have strong connections with others are happier,
healthier, and better able to cope with the stresses of everyday
life, most clients would presumably benefit from enhancing
their belongingness.

The psychotherapeutic usefulness of belonging can also be
seen in the effectiveness of group therapy. As Lewin ( 1 9 5 1 )
flatly stated, “It is easier to change individuals formed into a
group than to change them separately” (p. 228). In part, the
effectiveness of group therapy seems to depend on engendering
a sense of belongingness, as some authors have asserted (Larkin,
1972; Yalom, 1985). Forsyth (1991), in his review of research
on group therapy, observed that therapeutic groups provide the
member “with a sense of belonging, protection from harm, and
acceptance” (p. 675).

People differ, of course, in the degree to which they believe
that their belongingness needs are being met irrespective of the
extensiveness of their social networks or the strength of social
support they receive. Lakey and Cassady (1990) provided data
suggesting that perceived social support operates much like a
cognitive schema. People have relatively stable, organized be-
liefs about the extent and quality of their interpersonal relation-
ships. These belief systems lead to biased interpretation of so-
cial interactions, as well as to a biased recall of past interper-
sonal events. As a result, some people have a predisposition to
perceive others as unsupportive, leading them to experience be-
longingness deprivation even when others are in fact being
supportive.

Critical assessment. The diversity of methodologies and the
multiplicity of disciplines that have furnished the evidence re-
viewed in this section make it highly implausible to suggest that
all such evidence can be explained away as the result of con-
founds or artifacts. At worst, some of the findings have alternate
explanations. Not all studies have maintained careful distinc-
tions between the pragmatic benefits of certain relationships
and the direct benefits of belongingness. The fact that happily

6 The two overlap in many ways, of course. Cutrona (1986) has noted
that esteem support is an important element of social support, particu-
larly for helping people avoid depressive reactions to stressful events.

THE NEED TO BELONG 511

married people commit fewer crimes than other adults, for ex-
ample, might be partly (or even wholly) due to the material
benefits of being married. Even so, researchers who have main-
tained such distinctions (such as several of the social support
researchers) have found pragmatic benefits to be a secondary
factor. Belongingness thus has important and direct benefits.

A more serious limitation is that several of the findings are
correlational. The higher rates of mental and physical illness
among loners could reflect a tendency for people to reject
deviants as potential relationship partners. By the same to-
ken, the higher levels of life satisfaction found among happily
married people could be partly due to a tendency for chroni-
cally unhappy people to be rejected as marriage partners.
Still, those studies that have provided evidence about the di-
rection of causality have consistently identified belongingness
as the causal factor.

Conclusion. Deprivation of stable, good relationships has
been linked to a large array of aversive and pathological conse-
quences. People who lack belongingness suffer higher levels of
mental and physical illness and are relatively highly prone to a
broad range of behavioral problems, ranging from traffic acci-
dents to criminality to suicide. Some of these findings may be
subject to alternative explanations, and for some the direction
of causality has not been established; however, the weight of ev-
idence suggests that lack of belongingness is a primary cause of
multiple and diverse problems. It therefore seems appropriate
to regard belongingness as a need rather than simply a want.

Partial Deprivation: Relatedness Without Interaction

We have proposed that the need to belong has two aspects:
People require frequent interactions with the same person, and
people want a stable, enduring context of concern and caring.
This section examines some instances in which people have the
first of these while being deprived of the second, and the next
section examines the opposite case. These cases are important
for establishing whether the need to belong does indeed involve
both aspects. To confirm this version of the belongingness hy-
pothesis, a rather precise pattern of comparisons is needed, one
based on the assumption that satisfying only one of the compo-
nents should bring only partial satisfaction. People with only
one of the two components should presumably be slightly better
off than people who have neither, but they should be worse off
than people who have both.

One example of relatedness without interaction involves peo-
ple in prison. Many prisoners have families or loved ones on
the outside, but interactions with them are severely restricted.
Although systematic, quantitative data are scarce, works on
prison life appear to be in emphatic agreement that prisoners
treasure and cling to these ties yet suffer greatly over the lack of
interaction (Baunach, 1985; Isenberg, 1991; Toch, 1977). At
least some efforts at prison reform and the cultivation of al-
ternatives to standard imprisonment emphasize that increasing
contact with family members is beneficial to the prisoner
(Scudder, 1952). Toch (1977) documented the extensive an-
guish suffered by prisoners over lack of contact with family and
romantic partners, although he noted that the perceived threat
of losing the bond was often a source of suffering, which suggests

that the deprivation of interaction is not fully responsible for
the distress.

One group that might be relatively immune to this fear would
be imprisoned mothers, insofar as mother-child bonds cannot
be broken through divorce or other mechanisms (unlike ro-
mantic ties and friendships). Baunach (1985) and Giallom-
bardo (1966) both reported that imprisoned mothers lamented
the loss of interaction with their children and used every avail-
able means to maximize contact. They noted that these efforts
were especially impressive in that both the prison institution
and the collective wisdom of the prison subculture stress the
need to suspend all emotional interest in events outside the
prison, because such concerns produce frustration and helpless-
ness. Thus, these women’s efforts to maintain such ties are op-
posed by pervasive situational pressures, but they remain strong
anyway, suggesting that the bonds continue to offer satisfaction
of some powerful need despite the effort and frustration in-
volved in maintaining them.

Noncustodial divorced parents represent another group re-
strained from interacting with their children. Wilbur and Wil-
bur (1988) observed that most such parents refused to accept
that the bond to their children was severed or even damaged,
even when their lawyers advised them to abandon efforts to con-
tinue the relationship. Thus, the bond is apparently very impor-
tant to these parents even if it is mainly associated with frustra-
tion, aggravation, and disappointment. Meanwhile, the lack of
interaction and contact with the children was often very upset-
ting to these noncustodial parents, and indeed most of the di-
lemmas that Wilbur and Wilbur associated with noncustodial
parenthood revolved around a lack of contact and interaction.

Children of divorce are often in a similar position of losing
interaction access to the noncustodial parent. R. Rosen (1979)
found that children who had free, unlimited access to interact
with the noncustodial parent were least likely to perceive the
divorce as traumatic, although that finding was based on corre-
lational data and both variables (access and trauma) could have
been confounded with how well the parents got along with each
other after the divorce. Rosen also found that most children ex-
pressed a clear preference for such free access to the noncusto-
dial parent, and a large minority indicated that they had less
contact than they wanted to have with that parent. Thus, even
if the relational bond continues to exist, many children seem to
suffer from the reduction in interaction. A later study by Drill
(1987) concluded strongly that most children want to maintain
the bond despite the reduced interaction. Drill observed that
children of divorce were most prone to depression if they per-
ceived the noncustodial parent as being lost, in the sense of hav-
ing the bond severed. Fortunately, most children apparently
perceive the bond to remain in existence, which presumably
accounted for Drill’s finding that children of divorced parents
were no more likely overall than children of non-divorced par-
ents to be depressed.

Weiss (1973) reported that housewives who had recently
moved to the Boston area often reported loneliness despite hav-
ing a strong marital bond. They were lonely because they were
deprived of interactions most of the time: They had no local
friends, and their husbands were away all day and preoccupied
with their new jobs. Hoyle and Crawford (in press) found that
students’ sense of belonging to their university involved more

512 ROY F. BAUMEISTER AND MARK R. LEARY

than mere identification with the institution; it also had a strong
component of behavioral involvement. This sense of belonging
was heavily correlated (.65) with their involvement in univer-
sity activities, suggesting that daily interactions are an impor-
tant part of belongingness. Although alternative explanations
for these findings cannot be ruled out, the findings are consis-
tent with the general pattern that a bond alone is not enough to
satisfy the need to belong.

Long-distance relationships and commuter marriages offer
another set of circumstances in which people have an interper-
sonal bond but are relatively deprived of interaction. Gerstel
and Gross (1982) observed that people cling to these relation-
ships, which suggests that they are positively valued and provide
some rewards, but also find them stressful, consistent with the
view that relationship without interaction is less than fully sat-
isfactory. Similarly intermediate results were reported by Go-
vaerts and Dixon (1988): Commuters did not show any sig-
nificant drop in overall marital satisfaction, but they did express
dissatisfaction with time spent together and affectional commu-
nication. Gerstel and Gross found that the stressful aspect of
commuter marriage was significantly reduced by regular week-
end visits; thus, the opportunity for regular and fairly frequent
interactions was very beneficial (see also Holt & Stone, 1988).
They also found that couples who had been married longer and
therefore had a greater sense of stability suffered less from the
stress of separation than other commuter couples, presumably
because they could remain more secure and confident that the
attachment to the spouse would survive. Thus, these people
have a solid bond but still express a strong need for interactions.

In a later work, Gerstel and Gross (1984) reported that com-
muter couples valued the bond but suffered over the loss of in-
teraction. Couples seemed to find it ironic that small talk over
trivial matters would turn out to be something they missed, but
as Gerstel and Gross noted, these seemingly insubstantial in-
teractions are believed to be an important aspect without which
the marital bond is not fully satisfactory or fulfilling. Frequent
(long-distance) telephone conversations were a common but
not fully satisfactory solution to the deprivation of interaction.
Respondents in that study noted that telephone conversations
seemed adequate for sharing information and discussing prac-
tical affairs but were frequently deficient for producing pleasant
social interactions or enjoying one another’s company. This sug-
gests that regular interactions do have something to offer that is
not contained in merely knowing that the social bond exists and
exchanging information. In addition, loss of shared leisure ac-
tivities was a common complaint.7

Winfield’s (1985) study of commuter marriages confirmed
many of Gerstel and Gross’s (1982, 1984) conclusions. In ad-
dition, Winfield found a surprisingly low rate of sexual infidelity
(see also Gerstel & Gross, 1984) and concluded that married
people who live apart are, ironically, only about half as likely to
be unfaithful as married people who live together (despite the
presumably much greater opportunity and temptation). She
cited commitment to the relationship as an important reason
for this increased fidelity, and so it reflects on how these people
value the social bond. Still, it was clear that many couples suffer
and feel deprived because of the lack of interaction, and Win-
field observed that loneliness was a frequent problem. A similar
point was made by Bunker, Zubek, Vanderslice, and Rice

(1992), who found that commuting spouses were less satisfied
with their marital relationship, family life, and overall quality
of life than were spouses who lived together. For present
purposes, the implication is that the bond to an absent spouse
appears to furnish some positive benefits and satisfactions, but
people still suffer over the lack of contact. The evidence from
commuter marriages thus appears to confirm the importance of
two separate components of belongingness, namely the secure
confidence in an enduring bond of mutual caring and the regu-
lar experience of pleasant, affectively positive interactions.

Similar findings have emerged from studies of the spouses of
military personnel. Several articles on the wives of submariners
have shown that these women suffer anxiety, depression, and
physical illness during the long absences of their husbands (K.
Beckman, Marsella, & Finney, 1979; A. I. Snyder, 1978; see
Harrison & Connors, 1984, for a review). Pearlman (1970) ob-
served that each departure typically involved a crisis.

Critical assessment. Evidence from multiple fields and
seemingly quite different populations points to the same con-
clusion about the need for interactions. All of the studies can
be criticized on methodological grounds, however. The prison
samples may be atypical and pathological. Commuters may be
atypical because of having chosen to live apart (although the
fact that they still suffer from the deprivation despite this choice
would seemingly strengthen rather than weaken the argument
that frequent interactions are needed). The observations about
children of divorce seem less tainted by such concerns, but it
may be difficult to disentangle the multiple causes of distress.
The spouses of military personnel may be most representative
of the population at large. A further problem is that most of
these studies have used samples of convenience rather than sys-
tematically created ones. For prisoners in particular, and in
some of the studies of other groups, the data are largely obser-
vational and impressionistic, and it would be much better to
have quantified comparisons with well-chosen control groups.

Alternative explanations also plague the prison studies. Pris-
oners derive practical benefits from maintaining contact with
people outside the prison who can bring them material goods
and do them favors (Isenberg, 1991). As already noted, some of
the concern about lack of contact with loved ones may reflect a
fear of losing the bond, so it is not safe to regard prisoners as a
pure example of people who have a stable bond but lack interac-
tions. To some extent, this problem can be rectified by consid-
ering mothers, who should be less worried about being aban-
doned by their children; in some cases, however, they too fear
that the child will bond with someone else and become es-
tranged from them (Baunach, 1985), so it may be appropriate
to regard this fear as merely reduced, not eliminated, among
them. Baunach (1985) also noted that it is impossible to rule

7 One might wonder whether sexual deprivation was responsible for
the problems reported in commuter marriages. However, studies of
these couples indicate that most see each other a couple of days each
week, which in principle would be sufficient for the approximately
weekly sexual intercourse that is the norm among married couples. Ger-
stel and Gross (1984) found that most of these couples had had sex only
on weekends even when they lived together, so there was little decline in
sexual frequency as a result of commuting; most couples reported that
their sex lives were basically the same after they started commuting.

THE NEED TO BELONG 513

out the alternative explanation that some imprisoned mothers’
displays of concern for their children are feigned ploys to im-
press the parole board.

The growing literatures on commuter marriages and filial at-
tachments to divorced, noncustodial parents are less subject to
alternative explanations than the prison studies, but they too
are far from controlled, prospective studies that conclusively
demonstrate causal effects. There have been attempts to study
direct effects of frequency of interaction independent of prag-
matic and other benefits. Most of the findings are still correla-
tional, but on a priori grounds it seems implausible to suggest
the reverse causal hypothesis (e.g., that unhappiness over lack
of interaction causes people to spend less time together).

Despite these concerns, the convergence across different
groups and methods is encouraging. At present, it seems appro-
priate to accept the converging conclusions from these studies,
at least until contrary evidence is found.

Conclusion. Broad and consistent but methodologically
weak evidence supports the conclusion that having a relation-
ship without frequent interactions offers only partial, incom-
plete satisfaction of the need to belong. Researchers have stud-
ied several different circumstances in which people find them-
selves having relationships without interactions, and in each
case the same conclusion has emerged: People with such bonds
do seem to treat them as desirable and valuable (consistent with
the view that they do offer some rewards) but suffer over the lack
of direct contact with the other person.

Partial Deprivation: Interaction Without a Bond of
Caring

Interaction without an ongoing bond of caring should also
be only partly satisfactory. Two predictions can be made.
First, insofar as the need to belong requires that some interac-
tions reflect a relationship context, it can be predicted that
interactions with changing series of partners should be less
than satisfying. Second, if the interactions are supposed to
reflect the context of positive emotional concern, then people
should not be satisfied by interactions within the context of
an ongoing relationship or social bond that is not marked by
positive caring. We look for evidence for the specifically mu-
tual nature of the bond.

Need for r-elatedness. Can people be satisfied by frequent
interactions without stable relationships? Weiss (1973) ob-
served that “loneliness is not simply a desire for company, any
company; rather it yields only to very specific forms of relation-
ship” (p. 13). Wheeler et al. (1983) showed that loneliness is
largely independent of one’s amount of social contact, thereby
confirming Weiss’s observation. In the next section of this arti-
cle, we cover several studies showing that people seem to prefer
a few close friendships over a high number of transient or super-
ficial encounters and that evidence could be taken to indicate
that the relationship bond is essential to full satisfaction.

One possible population of people who have many interac-
tions without the bond of mutual caring would consist of pros-
titutes, who may have a high frequency of physically intimate
interactions with partners with whom there is no ongoing bond.
Sure enough, prostitutes often describe their occupation as hav-
ing the benefits of meeting interesting people and not being as

boring as other jobs (e.g., McLeod, 1982, p. 31). If intimate
interactions were sufficient to satisfy social needs without any
lasting bond, prostitutes might be very happy and well adjusted.
On the contrary, however, it appears that prostitutes are far
from satisfied by these interactions and instead seek and culti-
vate lasting bonds with others. The desire for bonds of mutual
caring is apparently often responsible for irrational, even self-
destructive attachments to procurers and other men (Adler,
1980; McLeod, 1982; Symanski, 1980). Also, many prostitutes
are single mothers, and the bond with the child is very impor-
tant (McLeod, 1982). Several signs indicate that prostitutes do
like to cultivate long-term relationships with clients, as evi-
denced by some brothel rules designed to prevent the formation
of such attachments (Symanski, 1980). Indeed, Symanski
(1980) calculated that prostitutes would maximize their finan-
cial earnings by working in brothels and serving the most cus-
tomers, yet many specifically objected to the procedures involv-
ing many brief contacts and sought to work in other settings
where they could have more time with each client and cultivate
repeat customers. These observations must be regarded as ten-
tative, however, because the studies lack methodological rigor.

Bond of caring. The next issue is whether all relationship
bonds can satisfy the need to belong. It appears that only bonds
marked by positive concern and caring offer satisfaction. Even
if a person has both an enduring bond and frequent interactions,
he or she may feel that the need to belong is not fully satisfied.
We turn now to relevant evidence involving cases in which the
person is firmly linked to others but has unpleasant or unsatis-
fying interactions with them.

Earlier, we listed a series of apparent benefits of social bonds
for health, adjustment, happiness, and general welfare. There is
an important qualification, however. In many cases, it is not
the mere fact of having an interpersonal attachment, but rather
having an attachment that brings positive interactions, that is
decisive. Relationships marked by conflictual interactions are
much less beneficial and sometimes harmful. DeLongis et al.
(1988) found that happily married people were much healthier
than were people in unsupportive social relationships. Thus, it
is not the mere fact of marriage, but rather having a supportive
marital relationship, that provides health benefits, and people
who are deprived of such a satisfying relationship are more vul-
nerable. Coyne and DeLongis (1986) reviewed evidence and
concluded that bad marriages may be worse than being alone in
terms of effects on happiness and health. Kiecolt-Glaser et al.
(1987) found decrements in immune function among unhap-
pily married women and among women who were separated
from their husbands while remaining emotionally attached to
them. Myers’s (1992) review of the literature on happiness con-
cluded that whereas good marriages provide a powerful boost
to happiness, bad marriages lead to extreme unhappiness. Like-
wise, research on social participation by Reis, Wheeler, Kernis,
Spiegel, and Nezlek (1985) found that the quality rather than
the quantity of social interactions predicted health. Specifically,
participants (particularly women) who had better quality in-
teractions (defined in terms of intimacy, pleasantness, satisfac-
tion, mutual disclosure, initiation, and influence) fared better
on a variety of measures of physical and mental health.

Although the lack of a good marital relationship appears to
be detrimental to mental health, the existence of a bad marital

514 ROY F. BAUMEISTER AND MARK R. LEARY

relationship is arguably worse. Having a spouse or close partner
may preclude the person from seeking other, more satisfying and
beneficial relationships, and the pervasive and salient conflic-
tual interactions may intensify the person’s feeling of not be-
longing. Thus, to complement the standard finding that good
social support is beneficial for mental health, Vinokur and van
Ryn ( 1 9 9 3 ) showedthat social undermining (i.e., conflict, crit-
icism, making life difficult, and inducing feelings of being
unwanted) in close relationships has a strongly negative effect
on mental health. Indeed, in their sample of unemployed peo-
ple, the effect of social undermining was stronger than the effect
of social support. Carnelley, Pietromonaco, and Jaffe (1994)
confirmed the link between problematic relationships to par-
ents and subsequent depression, and they also found that the
current romantic involvements of depressed adults tended to be
characterized by fearful avoidance and anxious ambivalence.
Although their results are correlational, they are quite consis-
tent with the view that problems and deficiencies in close rela-
tionships contribute to depression (with attachment style as a
mediating variable).

We also mentioned Sampson and Laub’s (1993) finding that
linked marriage and job involvement to reduction in criminal
activity. These reductions in crime were limited to people who
had good, stable, happy marriages and who were employed in
steady jobs. (The marital and job effects were independent.) In
contrast, the mere fact of being married, or the level of one’s
income, had no relation to crime. Thus, being well integrated
into good relationships, rather than merely having a social at-
tachment, reduces criminality.

Also relevant are studies on how a good marital relationship
affects offspring; indeed, for evolutionary analyses, these inves-
tigations may be especially important. Several reviews have con-
cluded that conflict between parents leads to aggressive, antiso-
cial behavior (such as juvenile delinquency) and perhaps other
behavior problems in children (Belsky, 1981; Emery, 1982;
Rutter & Garmezy, 1983). Indeed, Emery (1982) concluded
that parental conflict, rather than separation, is the main factor
responsible for the bad effects of divorce on children, because
the problems covary much more closely with conflict (in either
intact or separated parents) than with separation. Recent work
indicates that a good marital relationship tends to cause greater
warmth toward the children, which in turn reduces angry and
defiant misbehavior on the part of the children (N. B. Miller,
Cowan, Cowan, Hetherington, & Clingempeel, 1993; see also
Belsky, 1979).

Mutuality. The last issue concerns how important it is that
caring, concern, and affection be mutual and reciprocal. One
can well understand why people are better off to interact with
partners who care about them, because the partners might pro-
vide more material rewards and other benefits. But is there any
value to caring about the other person, as opposed to being
merely cared about?

In the first place, it does appear that mutuality is the norm.
M. S. Clark et al. (1987) showed that the desire to receive help
from others was correlated with the desire to give help and re-
spond to others’ needs. This suggests that the desire for commu-
nal relationships is based partly on the appeal of a framework
in which people have mutual concern for each other’s welfare.
The alternative explanation for Clark et al.’s findings would be

based on social exchange theory, which would propose that peo-
ple might prefer to be involved in one-way relationships, so that
they would receive the benefits of the other person’s care but not
incur the costs of having to care for the other person. Perhaps
mutuality is the norm only because people cannot find others
who will care for them without getting anything in return. The
evidence runs contrary to this view, however, despite its eco-
nomic and utilitarian logic. Hays (1985) examined relationship
satisfaction as a function of the costs and benefits to the individ-
ual. From a behavioristic standpoint, he predicted that satisfac-
tion would be predicted by an index of the rewards minus the
costs, which is precisely what economic rationality would favor.
Contrary to that prediction, however, Hays found that satisfac-
tion was much better predicted by an index of rewards plus
costs. In other words, people preferred relationships in which
both parties gave and received care.

Mutuality seems to improve and strengthen the relation-
ship. Rusbult, Verette, and Drigotus (1994) found that mu-
tuality of commitment predicted good marital adjustment.
This effect was independent of the actual level of commit-
ment, which shows that mutuality per se is indeed beneficial.
The other side of this was demonstrated by Hill, Rubin, and
Peplau (1976), who showed that unequal involvement was a
strong predictor of romantic breakup. Moreover, it was not
simply the case that the less involved partner was more likely
to break off the attachment, because in many cases the more
involved person initiated the breakup. Only when both part-
ners reported that both were equally involved was the couple
likely to still be together 2 years later.

If mutuality is good for relationships, it is also good for indi-
viduals, as indicated by recent findings from studies of unre-
quited love (Baumeister & Wotman, 1992; Baumeister, Wot-
man, & Stillwell, 1993). These studies compared people who
received love without giving it and people who gave love without
receiving it. To the researchers’ surprise, both groups tended to
describe the experience as aversive. Apparently, love is highly
satisfying and desirable only if it is mutual.

The parent-child bond is inevitably asymmetrical, insofar as
the child cannot provide the parent with the nurturant care and
concern that the parent must provide the child. If there are any
exceptions to the principle that mutuality is optimal, they
would presumably be found among parents. The difficulty, of
course, is determining what is the appropriate comparison. One
strategy would be to compare mutual and nonmutual two-per-
son families, that is, compare families consisting only of two
adults (i.e., childless marriages) and families consisting only of
a parent and child (i.e., single parents). Research has abun-
dantly shown that those two types of families differ dramatically
in terms of happiness (of the adult): The childless spouses are
happier than average, and the single parents are less happy than
average. In other words, if an adult woman is to have only one
other person in her family, she will be happier if this person is a
husband rather than a child (e.g., S. A. Anderson et al., 1983;
Baumeister, 1991; Bernard, 1982; Campbell, 1981; Campbell
etal., 1976).

One reason for the importance of mutuality may be trust.
J. G. Holmes and Rempel (1989) reviewed evidence that trust
is often a crucial and influential feature of good, beneficial, and
satisfying relationships and concluded that trust depends

THE NEED TO BELONG 515

heavily on mutuality, especially the mutual recognition of re-
ciprocal concern and attachment. Dissimilar feelings and un-
equal involvement prevent the growth of trust and thereby
thwart or weaken relationships.

Critical assessment. The evidence in this section was un-
even in quality and quantity. We have found no methodolog-
ically rigorous evidence indicating that frequent interactions
without an ongoing relationship offer partial or intermediate
satisfaction of the need to belong. The prostitution studies
were merely correlational and impressionistic, and, even if
they had been based on systematic samples of prostitutes
(which they were not), one would be reluctant to generalize
from prostitutes to the rest of the population.

In contrast, the evidence is stronger with regard to the
inadequacy of negative or conflictual interactions to provide
satisfaction. Although much of this evidence is correlational,
there is some time-sequence evidence suggesting that un-
happy marriages and other problematic relationships lead to
distress and illness.

The evidence for mutuality is scattered and fragmentary, al-
though it is consistent. Most of it is somewhat indirect. Further
research is needed to provide direct evidence about the impor-
tance of mutuality, particularly whether one’s own caring for
the partner is important for satisfying one’s own need to belong.

Conclusion. First, there is some evidence that interactions
with a changing series of partners, without any ongoing rela-
tionship bond, fail to satisfy people, but this evidence is sug-
gestive rather than conclusive. Second, several studies have
indicated that problematic or unhappy marriages fail to pro-
duce the benefits normally linked to belongingness and, in
fact, may make things worse. Thus, the mere fact of a social
bond is not enough to protect people from these problems
and pathologies. Rather, it appears that people require their
primary social bonds to be characterized by affectively posi-
tive interactions that signify the other’s affectionate concern.
Third, there are several indications that people prefer rela-
tionships marked by mutual, reciprocal concern, but stronger
and more direct evidence is needed. It is also plausible that
mutuality is merely a preference rather than a need.

Satiation and Substitution

The belongingness hypothesis holds that individuals need a
certain amount of social relatedness. Social relationships and
partners should therefore be to some extent interchangeable.
Moreover, people who have sufficient social bonds to satisfy the
need to belong should be less interested in forming additional
relationships than people who do not already have sufficient
bonds. These corollaries of the belongingness hypothesis can be
expressed in terms of satiation and substitution. Satiation refers
to the diminished motivation that ensues when the need to be-
long is already well satisfied, and substitution refers to the re-
placeability of one social bond with another. Satiation and sub-
stitution are not unrelated, of course, because both invoke the
basic assumption that people need a certain quantity of belong-
ingness, and attachments or interactions beyond that minimum
should be subject to a pattern of diminishing returns.

Satiation implies a diminishing returns principle in the pur-
suit of new relationships and partners. Even in people-rich en-

vironments such as colleges, people appear to restrict their so-
cial lives to some extent. Studies show that the vast majority of
the average student’s meaningful interactions are with the same
six people (Wheeler & Nezlek, 1977). Reis (1990) surveyed
students about their interpersonal goals, and although “having
lots of friends” received one of the lowest ratings, most of the
top-rated items referred to intimate sharing with a few close,
caring friends. Caldwell and Peplau (1982) found that a strong
majority of both men and women expressed a clear preference
for a few close friendships over a large number of good but less
intimate friendships. Thus, people appear to devote their time
and efforts toward deepening a limited number of relationships
rather than toward meeting ever new people or cultivating a
wider range of acquaintanceships. Consistent with the satiation
hypothesis, people seem to believe that, in terms of friendships,
quality (closeness) is far more important than quantity.

Audy (1980) suggested that this satiation is more or less es-
sential if a species is to survive. Organisms evolve a “limited
requirement for the frequency of social transactions and a cor-
responding optimum group size” that permit a maximum of
social gratification balanced by socially induced frustration
(pp. 123-124). As he noted, there is evidence that people have
evolved “a physiological structure and basic mental require-
ments suited to a particular group size that corresponds to
[their] need for a certain level of social transactions” (p. 124).

Satiation patterns, in the form of diminishing effects of social
approval as reinforcement, have also been investigated in the
context of learning theory. Gewirtz and Baer (1958) replicated
the standard pattern that children’s task performance would
improve in response to verbal approval reinforcers such as
praise and other approving remarks; moreover, the reinforce-
ment effect was intensified if the children had first been deprived
of social approval by being kept in isolation for a brief period.
In another condition, however, the children were first given an
interview in which they received praise and admiration for
whatever they said about themselves. After this interview, the
standard praise and approval remarks failed to elicit improve-
ments in task performance, which suggests that these partici-
pants had been satiated with approval and were unaffected by
further doses. Eisenberger (1970) reviewed the subsequent stud-
ies on the same topic and found that the initial results were well
replicated. Moreover, these effects were not a result of sensory
deprivation, and they also failed to alter the subsequent respon-
siveness to nonsocial rewards. Eisenberger concluded that social
deprivation and social (approval) satiation effects operated by
altering short-term motivation for obtaining approval. Al-
though these studies were generally conducted with children
and did not involve lasting relationships, they do indicate that
the motive to gain social approval is susceptible to satiation.

Social interaction patterns that accompany the formation of
an intimate romantic relationship are especially relevant, be-
cause both satiation and substitution are implicit. Milardo,
Johnson, and Huston (1983) found that as an intimate relation-
ship develops, people reduce the amount of time they spend in-
teracting with other people, including old friends. Thus, the ro-
mantic relationship appears to supplant the others and satisfy
the belongingness need previously satisfied by the other
friendships.

The belongingness hypothesis is, of course, not limited to the

516 ROY F. BAUMEISTER AND MARK R. LEARY

mere existence of some formal attachment but also depends on
the quality of the relationship. Consistent with this, Berman’s
(1988) research on attachment to ex-spouses found that the
positive qualities of the relationship were important determi-
nants of the distress over the loss, as indicated by the finding
that people who had more favorable memories of the marriage
also had more distress after it ended. For present purposes, the
important implication is that if bad marriages fail to satisfy the
need to belong, then, as a result, they should stimulate a search
for new attachments. Vaughan (1986) observed that when mar-
riages begin to develop significant problems that will eventually
lead to their dissolution, the individuals often begin to seek out
new friendships and relationships. Along the same lines, Law-
son’s (1988) research on adultery found evidence that substitu-
tion can be an important factor, particularly for women. She
found that the reason most commonly cited by women for en-
gaging in extramarital sex was the husband’s failure to satisfy
the wife’s intimacy needs. (For husbands, other factors such as
sexual novelty and variety were influential, but these factors are
irrelevant to the belongingness hypothesis.) The implication is
that when the marital relationship satisfies the need to belong,
women are unlikely to seek extramarital relationships, but
when the marriage is not satisfactory, extramarital substitutes
may be sought. This conclusion supports both the satiation and
substitution hypotheses.

Spanier and Casto (1979) found that most people relied
heavily on (and benefited from) social support from friends and
family during divorce. When friends and family were not sup-
portive, however, “this lack of support seemed to increase the
overall difficulties in adjusting to the separation, especially the
emotional adjustments” (p. 217). Spanier and Casto also noted
that a failure to make new friends made the adjustment worse.
In a direct test of the hypothesis that more social interaction
will lead to less adjustment problems after divorce, they found
a strong relationship between social activity and adjustment
problems. They also found that forming new heterosexual or
romantic relationships eased the transition of divorce and led to
far fewer difficulties of adjustment. When new relationships fail
to form, the emotional distress associated with the divorce and
the ex-spouse may actually increase rather than decrease over
time (Spanier & Casto, 1979, p. 226), which again implies that
substitution is an effective way of recovering from relationship
dissolution. A very different source of evidence for the same
conclusion is Bowlby’s( 1969,1973) observation that children’s
anxiety and distress over separation from the mother seemed to
be greatly reduced if the children were accompanied by a famil-
iar other person at that time.

Populations of old people offer a useful way to examine
progressive social deprivation, because in many cases old
people have retired from work, are losing spouses to death,
and cease to make new intimate friendships (e.g., Kaufman,
1986). Like Kaufman, L. J. Beckman (1981) found that old
women’s relationships to adult children had become increas-
ingly important to their lives. The happiness of old women
with children was unrelated to the amount of social interac-
tions with other people; among childless old widows, however,
happiness in life was significantly correlated with the quality
and quantity of social interaction with other people. Thus,
the rewards of social interaction with children appear to be

“exchangeable and interchangeable” (L. J. Beckman, 1981,
p. 1085) with the rewards of interacting with other people.
Similarly, older adults who have a close friend are no more
likely to become depressed if, for one reason or another, the
amount of social interaction they have with other people de-
creases. In contrast, older adults without a confidant who de-
crease their interactions with others are at a very high risk for
depression (Lowenthal & Haven, 1968). These results sup-
port the view that people need some social attachments to be
happy and that these attachments are to some extent inter-
changeable. In particular, close relations with nonrelatives
can apparently substitute for relationships with offspring, at
least in terms of preventing any significant loss of happiness.

L. J. Beckman ( 1 9 8 1 ) also obtained findings relevant to
the satiation hypothesis. She found that the total amount of
social interaction with others was a significant predictor of
happiness among childless women but not among old women
who did have children, and she suggested that restriction of
range may account for this differential predictability. Spe-
cifically, according to Beckman, most old women with chil-
dren do have at least a certain minimal level of social interac-
tion, provided by the children, and so these women hardly
ever fall into the category of extreme loneliness and social
deprivation. Although Beckman repeatedly found that, iron-
ically, interactions with nonoffspring had a bigger impact on
happiness than interactions with children, having children
visit occasionally seemed to be enough to satisfy the need to
belong sufficiently to prevent the most severe problems of de-
privation. Above that minimum, further quantity of social
interaction did not appear to have an effect.

Substitutability patterns were suggested in a very different
way by Rusbult, Zembrodt, and Gunn (1982; see also Rusbult,
1980). They suggested that people remain in their close rela-
tionships for several reasons, and one important factor is the
availability of desirable alternative potential partners. In other
words, people are more likely to leave an intimate relationship
if they have some prospect of forming another intimate rela-
tionship with someone else soon.

Divorced people are at risk for a multitude of bad outcomes,
including illness, homicide, suicide, criminality, and accidental
injury or death (Bloom et al., 1979). One explanation is that
the divorced population represents a self-selected group of
pathologically inclined individuals, but this dispositional argu-
ment is weakened by the finding that remarriage tends to reduce
or eliminate the elevation in risk. The trauma of divorce itself
may be partly responsible for the increase in vulnerability, but
although the risks are highest immediately after divorce, they
do not fully subside until remarriage. The fact that remarriage
appears to eliminate many negative consequences of divorce
can indeed be explained in several ways, but it is at least strongly
consistent with the hypothesis that the new marriage substitutes
for the old one.

Divorce may often be voluntary, but imprisonment is not,
and prisoners suffer deprivation of contact with relationship
partners outside the prison. Men’s prisons are physically dan-
gerous, and both the gang bonding and the cultivation of social
isolation commonly found among male prisoners may reflect
more a concern with physical safety than anything else. In con-
trast, female prisons are far less dangerous to inmates, and so it

THE NEED TO BELONG 517

is relevant to examine how women prisoners adapt to the depri-
vation of contact with outside partners. Several studies have
found that female prisoners commonly form substitute families
based on imaginary kinship ties with other prisoners (Burkhart,
1973; Giallombardo, 1966; Toch, 1975). Some pseudomarital
bonds appear to involve voluntary homosexuality, which is typ-
ically renounced on leaving the prison (Burkhart, 1973). In
some cases, these pseudofamilies become quite extensive and
complex, with many women playing parts of parent, child, hus-
band, wife, grandparent, and other roles for each other. These
patterns appear to be simply an adaptation to prison life based
on the desire to experience romantic and family-style belong-
ingness during the time one is deprived of contact with the usual
relationship partners.

As we noted earlier, when groups break apart or people move
away from relationship partners, there is often initially strong
resistance to dissolving the relationship, but this resistance
tends to diminish over time (e.g., Lacoursiere, 1980; Lieberman
et al., 1973). These efforts to maintain the bond may be driven
by the absence of social ties in the new environment, and as
people gradually form new attachments they lose the need to
sustain the old ones. If this is correct, then future research
should find that people’s efforts to sustain friendships across
long distances are inversely proportional to their opportunities
to develop new friendships. For example, when people move
overseas, where cultural differences may hinder the develop-
ment of new intimate friendships, they should be more likely to
stay in touch with old friends than when moving to another
place in the same country.

Not all relationships are interchangeable, of course. Close re-
lationships based on romantic love may offer a variety of satis-
factions that are not easily obtained through nonromantic, non-
sexual friendships. Ruehlman and Wolchik (1988) found that
there were indeed particular benefits connected with the rela-
tionship to the most significant other in a person’s life. More
precisely, they found that once the social support and hindrance
provided by the most significant other person in someone’s life
were taken into account, there were no additional significant
effects of the support and hindrance provided by other people.
This pattern of results suggests that people need at least one
particularly strong, close attachment and that once they have
that, further attachments are subject to some principle of di-
minishing returns. A similar point was made by Coyne and
DeLongis (1986), who concluded, from a review of the social
support literature, that the harmful effects of a bad marriage
are not offset by having other good relationships; thus, again,
the special importance of the marital bond was confirmed.
Likewise, although women prisoners adapt to prison by sus-
pending their emotional attachments to most outsiders and
forming substitute family relationships with other prisoners,
they do exert themselves extensively to maintain the bond with
their real children who remain outside the prison (Baunach,
1985). And of course, as Kazan and Shaver (1994a) have
pointed out, although children could conceivably affiliate with
any available person, they nearly always focus on one particular
person, and their need to interact (as evidenced, in part, by dis-
tress over separation) becomes mainly focused on that person.

Critical assessment. There is an assortment of evidence
consistent with the hypotheses of satiation and substitution, but

the evidence is neither systematic enough nor unambiguous
enough to regard those hypotheses as strongly supported. Thus,
the fact that forming a close romantic attachment leads to with-
drawal from other friendships could be partly due to having a
limited amount of time to spend socializing rather than to any
reduction in need for the other friendships. Likewise, the culti-
vation of external friendships when a marriage goes bad could
be due to a need to discuss the marital problems with a sympa-
thetic outsider rather than a quest to find a new social bond that
could furnish what the marriage no longer provides. Although
the diversity of spheres yielding consistent findings encourages
one to expect that further evidence will continue to fit the satia-
tion and substitution hypotheses, more systematic work is
needed to rule out alternative explanations.

Also, there appear to be limits on the extent to which rela-
tionships can be substituted. A close romantic attachment to a
partner, with sexual attraction, appears to have special benefits
that cannot be compensated by other relationships. Still, when
such a relationship ends, forming a new one appears to be
sufficient to bring the person back to an equally high level of
adaptation and happiness, which suggests that, in the final anal-
ysis, a new spouse may be an effective substitute for a previous
one.

Conclusion. People’s interaction patterns and surveys of
preferences suggest that people seek a limited number of rela-
tionships, consistent with the view that the need to belong is
subject to satiation and diminishing returns. The first few close
social bonds appear to be the most important, beyond which
additional ones furnish ever lesser benefits. When people lose
such bonds or find their particular partners inadequate, they
can often derive similar benefits from others, suggesting that
partners can be substituted to some extent. There are certain
kinds of relationships that cannot effectively be replaced with
other kinds of relationships, although finding a new relationship
of the same type appears, in many cases, to be viable and
effective. These conclusions are tentative, however, and further,
more systematic work is desirable.

Innateness, Universality, and Evolutionary Perspectives

We proposed that a fundamental need would presumably be
innate, which would entail that it is found in all human beings
and is not derivative of other motives. This will, of course, be
quite difficult to verify, because empirical criteria for testing
such a hypothesis are not widely recognized. One approach,
however, would be to examine how well the empirical evidence
conforms to evolutionary arguments. If evolution has instilled
the motivation, then it is presumably universal among human
beings and will be present in each person without needing to be
derived from other motives.

Barchas (1986) has asserted that “over the course of evolu-
tion, the small group became the basic survival strategy devel-
oped by the human species” (p. 212). He went on to suggest
that the brain and small groups evolved and adapted together,
with multiple interrelationships. The evidence reviewed by Bar-
chas remains preliminary, but it does seem that any link be-
tween brain structures and small-group formation would
strengthen the case for an innate motivation.

Although the psychobiological systems involved in social at-

518 ROY F. BAUMEISTER AND MARK R. LEARY

tachment are not yet well understood, early evidence implicates
the brain opioid system. According to Panksepp, Siviy, and Nor-
mansell (1985), both the tendency to form social bonds and
the emotional effects of social loss (e.g., sadness or grief) are
mediated by opioids. The formation and validation of relation-
ships apparently stimulate opioid production, whereas the dis-
solution of relationships impedes it. As Panksepp et al. put it,
“social affect and social bonding are in some fundamental neu-
rochemical sense opioid addictions” (p. 25). Thus, in their
view, the tendency to seek social connections with others is
based not only on the secondary reinforcements that other peo-
ple provide but on psychophysiological mechanisms as well.

Multiple evolutionary reasons could be suggested for the
readiness to form groups easily. Groups can share labor, re-
sources, and information; diffuse risk; and cooperate to over-
come stress or threat (Hogan et al., 1985). Defense against rival
groups would also be a significant factor: If other people form
into groups, lone individuals would be at a competitive disad-
vantage in many situations, and so evolution may have selected
for people who would form groups defensively. Hence, the evo-
lutionary argument would fit any evidence that group forma-
tion or cohesion patterns are increased by external threat.

It has long been noted that external threats increase group
cohesion, and some writers have treated this as axiomatic. Stein
(1976) reviewed these views in light of the evidence and found
that a broad variety of methods have yielded generally consis-
tent findings; that is, external threats do increase cohesion most
of the time. There are some circumstances in which groups dis-
integrate under threat, especially if the threat pertains only to
some members of the group or if group members must compete
against each other to survive the threat (e.g., if there are too
few lifeboats). Staw, Sandelands, and Dutton (1981) also found
evidence that group cohesion is sometimes weakened in the af-
termath of a threat, especially if the group has failed to defeat
the threat and the group members blame each other. Apart from
these circumscribed exceptions, however, it is safe to conclude
that external threats do generally increase group cohesion.

A remarkable demonstration of the power of external threat
to forge lasting bonds was provided by Elder and Clipp’s (1988)
study of World War II veterans’ groups. In Elder and Clipp’s
results, the effects of maximum threat were discernible 40 years
later. That is, four decades after the war, the most enduring and
strongest ties were found among veterans who had experienced
heavy combat together and had suffered the deaths of some close
comrades. Units that had experienced combat without fatalities
were less close 40 years later, but they retained stronger ties than
the units that had not been in combat together. In other words,
the sharing of military experience provided some lasting bonds,
these bonds were intensified by shared experience of combat,
and they were especially strong if it had been heavy combat that
had killed some members of the group. It seems clear that there
would be survival benefits to a pattern in which the death of a
group member strengthened the ties among the survivors, espe-
cially in the face of external danger.

The group formation effects in the Robbers Cave study
(described earlier; Sherif et al., 1961 /1988) accelerated rapidly
after the mutual discovery of the existence of the two rival
groups; that is, the implicit threat posed by the opposing group
seemed to motivate each boy to cling to his own group more

strongly. Similar processes have been observed in terrorist
groups, which mainly become cohesive in the face of external
threat and danger. During periods when the conflict with out-
siders lapses, terrorist groups experience internal dissent and
conflict and may fall apart (see McCauley & Segal, 1987).

Compelling evidence in favor of emphasizing the competitive
disadvantage motive for affiliating was provided by Hoyle,
Pinkley, and Insko (1989). These researchers noted the irony
that encounters between individuals are generally pleasant and
supportive, whereas encounters between groups are frequently
unpleasant and confrontational, and their first study confirmed
these general expectations and stereotypes empirically. In their
second study, they sought to determine the decisive factor by
comparing interactions between persons, between groups, and
between one person and one group. To their surprise, they found
that participants’ expectations about the interaction were deter-
mined mainly by the other party rather than by participants’
own belongingness status. When participants expected to in-
teract with a group, they expected an abrasive interaction; when
they expected to interact with an individual, they anticipated
a pleasant, agreeable interaction. Identical effects were found
regardless of whether the participant expected to be alone or to
be part of a group. Thus, apparently, the presence of an out-
group causes people to anticipate conflict and problematic in-
teractions. Such an expectation could well elicit a motivation to
form a group to protect oneself.

A similar conclusion was suggested by Lauderdale, Smith-
Cunnien, Parker, and Inverarity (1984). Following Schachter’s
( 1 9 5 1 ) studies on group rejection of deviants, they found that
increasing an external threat led to increased rejection. The im-
plication was that groups become increasingly oriented toward
solidarity when confronted with an external threat.

Apart from threat, the possibility of gaining resources also
seems to trigger group cohesion, even when it is functionally
irrelevant. Rabbie and Horwitz (1969) assigned participants
randomly to two groups. The random assignment alone yielded
no effects of group cohesion on their measures of in-group pref-
erence, but they did find significant effects after a manipulation
in which one group was given a prize (transistor radio) based
on a coin flip. The rewarded group and the deprived group both
showed increased in-group preference. The prize was logically
irrelevant to subsequent group activities and preferences. The
implication is apparently that the combination of limited re-
sources and multiple groups triggers an in-group preference re-
sponse that has no apparent practical or rational basis, which is
consistent with the view that it is a deeply rooted and possibly
innate tendency rather than a strategic or rational choice.

Critical assessment. The evidence linking external threat to
increased group cohesion is convincing but does not prove an
evolutionary hypothesis of innateness or universality. The evi-
dence for brain mechanisms is likewise supportive but inade-
quate to prove innateness. The evidence in this section is per-
haps best described by stating that the evolutionary hypothesis
nicely survived several tests that could have contradicted it.

Conclusion. Several patterns seem consistent with evolu-
tionary reasoning. It remains plausible (but unproven) that the
need to belong is part of the human biological inheritance. If so,
the case for universality and nonderivativeness would be strong.

THE NEED TO BELONG 519

At present, it seems fair to accept these hypotheses as tentative
working assumptions while waiting for further evidence.

Apparent Counterexamples

Although the evidence presented thus far has been largely
supportive of the belongingness hypothesis, one might object
that our literature search has been structured in ways that pre-
disposed it toward just such favorable indications. It is therefore
desirable to examine behavioral patterns that would seemingly
constitute boundary conditions or counterexamples to the need
to belong. This section briefly considers several.

Refusal to help or cooperate. People generally show a sig-
nificant willingness to help others, but often there may be self-
interested motives lurking behind the apparent altruism (e.g.,
Cialdini, Darby, & Vincent, 1973; Manucia, Baumann, & Cial-
dini, 1984). To be sure, in many cases people appear to put self-
interest ahead of the welfare of others, leading them to disdain
opportunities for helping others or cooperating. Entering into
the long-running debate about the possibility and reality of
truly altruistic behavior is beyond the scope of this article; our
goal is merely to ask whether such cases do indeed contradict
the belongingness hypothesis. In particular, it is necessary to ask
whether these nonhelpful, noncooperative behavior patterns are
reduced or eliminated by belongingness.

One of social psychology’s best-known findings concerns the
unhelpfulness of multiple bystanders at an emergency site. As
Darley and Latane’s (1968) study and many subsequent inves-
tigations (see Latane & Nida, 1981) showed, people tend not to
come to the aid of an emergency victim when many other people
are also present. Among the reasons for the bystander effect are
the sense that it is not one’s own responsibility to help and the
fear that helping may bring negative consequences to the self.
Various findings suggest, however, that belongingness can over-
come the nonresponsiveness of bystanders. The bystander effect
is apparently robust among strangers (e.g., Darley & Latane,
1968), but in cohesive groups, the opposite pattern is found,
namely that larger groups produce more helping (Rutkowski,
Cruder, & Romer, 1983). Even the mere anticipation of future
interaction among group members is enough to eliminate the
bystander effect, making group members quite willing and likely
to come to each other’s aid (Gottlieb & Carver, 1980).

Social loafing is another pattern in which people put self-in-
terest ahead of cooperative concern for others (e.g., Latane, Wil-
liams, & Harkins, 1979). In social loafing, people reduce their
efforts when submerged in the group, thereby gaining benefits
of the group success without having to exert themselves maxi-
mally. Group membership can foster a sense of duty or obliga-
tion that can effectively override tendencies to engage in social
loafing, however. Harkins and Petty (1982) showed that if peo-
ple believe that they can make a unique contribution to the
group, they do not engage in social loafing, even if individual
contributions to the group will not be identified (and thus even
if they will not receive credit for their contribution; see also
Hardy & Grace, 1991).

The prisoner’s dilemma game has been widely used to exam-
ine how people choose between a self-interested, individualistic
(competitive) response and a cooperative response that can po-
tentially maximize everyone’s collective outcomes at the cost or

risk of individual vulnerability to loss. Once again, the presence
or apparent possibility of social attachments seems to shift peo-
ple away from the self-oriented mode toward a more coopera-
tive, collectively beneficial mode of response. The expectation
of future interaction increases helpful cooperation in the pris-
oner’s dilemma game, although this effect appears to obtain
mainly among high self-monitors (Danheiser & Graziano,
1982). The opportunity to meet and talk with strangers appears
to be sufficient to alter responses to a subsequent prisoner’s di-
lemma game in favor of increased cooperation and decreased
exploitation-defensiveness (Orbell et al., 1988).

Lastly, the commons dilemma (in which people deplete
renewable resources for short-term individual gain) is an-
other pattern in which people typically seek personal advan-
tage at the expense of long-range collective welfare. The com-
mons dilemma also can be reduced or overcome by belong-
ingness, however. Kramer and Brewer (1984) showed that
when belongingness is stimulated by making the group iden-
tity salient, people are more likely to restrain their self-inter-
ested tendencies and instead cooperate with others for the
greater good of the group.

More generally, helping appears to be increased by the exis-
tence of social bonds. Schoenrade, Batson, Brandt, and Loud
(1986) found that the existence of a social relationship in-
creases the motivation for helping. In the absence of a relation-
ship, people help only for egoistic reasons (i.e., self-interest);
when a relationship exists, however, people will help for rela-
tively selfless, altruistic reasons (see also Toi & Batson, 1982).
Even among strangers, familiarity leads to increased helping, as
does a sense of interpersonal dependency (Pearce, 1980). The
fact that belongingness can overcome self-interested patterns is
shown by evidence that people prefer reciprocity in social ex-
change to the extent that even overbenefited individuals some-
times feel uncomfortable and distressed even though material
self-interest is maximally served under conditions of being
overbenefited (Rook, 1987a). The concern with equity and
with aiding others is further indicated by the occasionally nega-
tive responses of would-be helpers to having their helpful efforts
spurned by the intended recipients (e.g., S. Rosen, Mickler, &
Collins, 1987).

Throughout this article, we have suggested mat the need to
belong may be biologically prepared. Evidence with animal spe-
cies is therefore relevant here. Masserman, Wechkin, and Terris
(1964) taught rhesus monkeys to pull a chain for food and then,
in one condition, added the contingency that pulling the chain
would cause a shock to be delivered to another monkey. Most
monkeys refrained from pulling the chain under those condi-
tions, even to the extent of starving themselves for several days
rather than cause another monkey to be shocked. These pat-
terns were particularly strong when the 2 animals had pre-
viously been cage mates and thus may be presumed to have
formed some sort of bond; when the animals were strangers to
each other, less than a third showed this form of altruistic, self-
sacrificing behavior.

Nonreciprocation of love. Although mutual love provides
strong satisfactions and hedonic benefits, there are many cases
in which people do not reciprocate another’s affection and ro-
mantic interest. Such refusals to form a social bond might be
taken as evidence against the belongingness hypothesis.

520 ROY F. BAUMEISTER AND MARK R. LEARY

On closer examination, however, inspection of patterns of un-
requited love does not provide a serious challenge to the belong-
ingness hypothesis, for several reasons. First, most people do
want to form a close romantic relationship, and their refusals
are typically based on either already having such a relationship
with another partner (consistent with the satiation hypothesis)
or perceiving the aspiring partner as unsuitable for some reason,
such as unattractiveness or incompatibility. Moreover, in many
cases, rejectors experience considerable distress such as guilt
and empathic pain when rejecting another’s offer of love. This
distress is consistent with the view that rejecting social attach-
ment goes against some deeply rooted aspect of human nature,
even when the person is quite certain that he or she does not
want this particular attachment (Baumeister & Wotman, 1992;
Baumeisteretal, 1993).

Shyness. Shy behavior patterns may seem antisocial insofar
as the shy person sometimes avoids social encounters, with-
draws from ongoing interactions, and acts in other ways that
reduce the chances of forming relationships (Leary, 1983). In
fact, however, shy people are strongly motivated to form rela-
tionships, and shyness may reflect an interpersonal strategy that
partially protects the individual against rejection.

When people do not believe that they will be regarded in ways
that will result in social acceptance, they may avoid absolute
rejection by disaffiliating. Although reticence and withdrawal
are unlikely to make particularly good impressions or to bring
hearty acceptance from others, they reduce the risk of saying or
doing something that others might regard negatively. When one
fears rejection, the best tactic may seem to be to participate as
little as possible, thereby giving others few reasons to reject one
(Shepperd&Arkin, 1990).

At the same time that they pull back, however, shy people
engage in behaviors that have been characterized as “innocu-
ously sociable” (Leary, 1983). They smile more (even though
they feel anxious rather than happy), nod their heads more in
agreement, ask more questions, and use more verbal reinforcers
when others are speaking. These behaviors may reflect last-re-
sort tactics to maintain a minimum degree of interpersonal con-
nection in otherwise difficult or threatening encounters (Leary,
Knight, & Johnson, 1987).

General Discussion

We have considered a broad assortment of evidence pertain-
ing to the hypothesis that the desire for interpersonal attach-
ments—the need to belong—is a fundamental human motiva-
tion. Most of the metatheoretical requirements we outlined for
evaluating such a hypothesis appear to be satisfied, although
some issues remain. We begin by reviewing the major
conclusions.

Again and again, we found evidence of a basic desire to form
social attachments. People form social bonds readily, even un-
der seemingly adverse conditions. People who have anything in
common, who share common (even unpleasant) experiences,
or who simply are exposed to each other frequently tend to form
friendships or other attachments. Moreover, people resist losing
attachments and breaking social bonds, even if there is no ma-
terial or pragmatic reason to maintain the bond and even if
maintaining it would be difficult.

Abundant evidence also attests that the need to belong shapes
emotion and cognition. Forming or solidifying social attach-
ments generally produces positive emotion, whereas real, imag-
ined, or even potential threats to social bonds generate a variety
of unpleasant emotional states. In short, change in belong-
ingness is a strong and pervasive cause of emotion in ways that
support the hypothesis of a need to belong. It is also evident that
people think a great deal about belongingness. They devote a
disproportionate amount of cognitive processing to actual or
possible relationship partners and interaction partners, and
they reserve particular, more extensive, and more favorable pat-
terns of information processing for people with whom they
share social bonds.

Deficits in belongingness apparently lead to a variety of ill
effects, consistent with the view that belongingness is a need
(as opposed to merely a want). Both psychological and physical
health problems are more common among people who lack so-
cial attachments. Behavioral pathologies, ranging from eating
disorders to suicide, are more common among people who are
unattached. Although most of these findings are correlational
and many alternative explanations can be suggested, recent
efforts have begun controlling for these other factors, and the
pure, primary effects of belongingness appear to remain strong.
It appears, then, that belongingness is not only pleasant but also
apparently very beneficial to the individual in multiple ways.

We proposed two aspects of the need to belong, and both ap-
pear to be important. That is, people seem to need frequent,
affectively pleasant or positive interactions with the same indi-
viduals, and they need these interactions to occur in a frame-
work of long-term, stable caring and concern. People who can
satisfy one component but not the other tend to be less satisfied
and less well off then people who can satisfy both, but they do
seem to derive some benefits from satisfying the one component
(as opposed to satisfying neither). More and better evidence is
needed on this point, however; most evidence pertains to people
who have the bond and lack interactions, rather than the re-
verse. Also, it is unclear whether the interactions must be pleas-
ant or can be satisfactory if they are merely neutral. The evi-
dence suggests merely that aversive or conflictual interactions
fail to satisfy the need. Some evidence suggests that a framework
of mutual, reciprocal concern is best, but the effects and impor-
tance of mutuality need further investigation.

The need to belong also appears to conform to motivational
patterns of satiation and substitution. People need a few close
relationships, and forming additional bonds beyond those few
has less and less impact. Having two as opposed to no close
relationships may make a world of difference to the person’s
health and happiness; having eight as opposed to six may have
very little consequence. When a social bond is broken, people
appear to recover best if they form a new one, although each
individual life tends to involve some particularly special rela-
tionships (such as filial or marital bonds) that are not easily
replaced. People without intimate partners engage in a variety
of activities to find partners, but people who have partners al-
ready are much less active at seeking additional relationships,
consistent with the satiation hypothesis.

We reviewed evidence that the need to belong affects a broad
variety of behaviors; indeed, the range is sufficiently broad as to
render less plausible any notion that the need to belong is a

THE NEED TO BELONG 521

product of certain other factors or motives. We also noted that
evidence about belongingness seems to implicate some brain
mechanisms and to conform to patterns that evolutionary the-
ory would suggest, both of which seem consistent with the ar-
gument that the need is innate in humans. Still, the nonderiva-
tive hypothesis is probably the least well supported aspect of our
theory, not because of any clear evidence deriving the need to
belong from other motives but simply perhaps because it is rel-
atively difficult to collect compelling data to show that a motive
is not derivative. The issue of which motives derive from which
others appears to be an important challenge for future motiva-
tion research.

We also considered several counterexamples that at least su-
perficially suggested tendencies to reject social attachment. On
close inspection, these patterns did not stand up as counterex-
amples, and indeed there was generally strong evidence of a pos-
itive need to belong that increased the subjective difficulty of
rejecting or avoiding attachment.

We conclude, then, that the present state of the empirical ev-
idence is sufficient to confirm the belongingness hypothesis. The
need to belong can be considered a fundamental human
motivation.

Implications for Psychological Theory

If the belongingness hypothesis is indeed correct, then it
seems plausible that many psychological phenomena may be
affected by this motivation. Clearly, patterns of group behavior
and close relationships can be understood as serving the need to
belong. It is thus not necessary to derive all group and intimate
affiliation patterns from other motives, such as the fact that
groups may confer pragmatic benefits or bolster self-esteem (see
also Turner, 1985). People may simply desire to belong to
groups. Patterns of self-presentation (Baumeister, 1982; Leary,
1994;Leary&Kowalski, 1990; Schlenker, 1980), interpersonal
redress and excuse making (R. S. Miller & Leary, 1992;
Schlenker, 1980; C. R. Snyder, Higgins, & Stucky, 1983), and
group conformity (Moreland & Levine, 1989) may all be seen
in the context of enhancing one’s chances of inclusion in groups
and relationships. Also, it may be no accident that people seem
most likely to be prejudiced against members of groups to
which they have little or no opportunity to belong. Thus, the
most common and widespread bases of prejudice are race, gen-
der, and national origin. People bolster their own in-group at
the expense of out-groups from which they are excluded (e.g.,
Meindl & Lerner, 1984).

Although antisocial behavior might, at first glance, be re-
garded as another potential counterexample for the belong-
ingness hypothesis (because antisocial behavior makes enemies
and alienates other people), it is readily apparent that belong-
ingness has close ties to it. Members of some groups are pres-
sured to commit violent acts, ranging from vandalism to mur-
der, to be accepted by and to demonstrate commitment to the
group (e.g., Breitman, 1991; Freud, 1913/1956; Hogan &
Jones, 1983; Rosenberg, 1991; Sarbin, 1982;Staub, 1989;Toch,
1992). It seems likely that aggression as well may have some
belongingness as a prerequisite, because aggression risks
alienating other people and so only people with firm attach-
ments can safely engage in aggressive behavior.

The centrality of belongingness to human psychological func-
tioning also has implications for the treatment of emotional and
behavioral problems. From our standpoint, a great deal of peo-
ple’s psychological difficulties reflects emotional and behavioral
reactions to perceived threats to social bonds. As has been
shown, many of the emotional problems for which people seek
professional help (anxiety, depression, grief, loneliness, rela-
tionship problems, and the like) result from people’s failure to
meet their belongingness needs. Furthermore, a great deal of
neurotic, maladaptive, and destructive behavior seems to reflect
either desperate attempts to establish or maintain relationships
with other people or sheer frustration and purposelessness when
one’s need to belong goes unmet.

Implications for Other Fields

We suggested at the outset that the belongingness hypothesis
ought to have implications that go beyond immediate psycho-
logical functioning and, indeed, that it might prove useful as an
explanatory construct for the phenomena studied by research-
ers in other fields. We now consider briefly some nonpsycholog-
ical applications of the need to belong.

Contrary to cultural materialism, we have proposed that
many aspects of human culture are directly and functionally
linked to enabling people to satisfy the psychological need to
belong. If this is correct, then some historical and sociological
changes in the structures of society should be linked to changes
in the bases for belongingness. For example, membership in
many organizations (including corporate employment) has
largely ceased to depend on family connections the way it once
did, with corresponding changes in the definition and power bal-
ance in families, the educational system (which provides
credentials), and other placement systems (e.g., Burgess &
Locke, 1945; Pass, 1977).

By the same token, the decline of arranged marriages and the
increasing availability of divorce have made romantic attach-
ment more dependent on individual attractiveness and other
traits. Concern over the self as an instrument for attracting oth-
ers and maintaining attachments should therefore increase.
Thus, becoming old or fat would be less threatening if divorce
were impossible or if marriages were arranged. With the in-
creasing threat, social structures should emerge, for example,
to help people look young or lose weight. Also, sexuality has a
frequently changing relationship to social inclusion, and vari-
ous eras have included or excluded people on the basis of sexual
chastity, skill, appeal, and perceived healthiness.

A general pattern may well be that cultures use social inclu-
sion to reward, and exclusion to punish, their members as a way
of enforcing their values. As is well known, many early civiliza-
tions equated exile with death, which seems to suggest that life
is desirable only within the network of close relationships to
which the person belongs. Modern civilizations tend to use
prison to punish people, which again invokes the principle that
depriving people of contact with relationship partners is highly
aversive; solitary confinement is generally recognized as the
most severe and aversive form of imprisonment. On the positive
side, the evolution of modern society has seen an increasingly
broad and fundamental quest for fame. Braudy’s (1986) history
of fame characterizes the desire for fame as based on a “dream

522 ROY F. BAUMEISTER AND MARK R. LEARY

of acceptance” that holds the (often illusory) promise that once
a person achieves fame, he or she will be embraced and sought
by others for the rest of his or her life. Fame may well be thus
another instance of the use of social inclusion as a reward. A. H.
Buss (1983) has pointed out that both the presence of others
and the attention of others are important social rewards, and
the deprivation of such contact has often been used as powerful
social punishment.

Turning to political science, a well-known article by
Morgenthau (1962) argued that the pursuit of power can be
understood as the counterpart to the pursuit of love in that
both involve an attempt to escape from loneliness. In Mor-
genthau’s analysis, the human condition suffers from the
threat of isolation, and by breaking down the barriers be-
tween one another people hope to achieve a sense of together-
ness. The main difference between love and power is that love
aspires to a mutual dissolving of personal boundaries, leading
to an egalitarian merging into a new whole, whereas power
seeks a unilateral overcoming of boundaries, by which the
will of the more powerful person becomes the will of both.
Morgenthau noted that the pursuit of power often fails to
overcome loneliness, so that, ironically, the most powerful in-
dividuals end up feeling still isolated and lonely (hence, the
tendency for rulers to demand that their subjects love them
too). For present purposes, the main point is that the need to
belong may be regarded as a major source of the desire for
power.

The role of belongingness is also apparent in religion. Al-
though ideological belief and acceptance of metaphysical doc-
trines are often regarded as the essence of religious participa-
tion, Stark and Bainbridge (1985) reviewed considerable evi-
dence suggesting that the need to belong may be a more
compelling factor than the need to believe. They noted that
movement into and out of religious groups (including cults,
sects, and mainstream denominations) depends much more
heavily on social ties than on ideological belief. Indeed, many
people do not fully grasp or understand the theological belief
structure of their own religion (e.g., the subtle differences be-
tween the many Protestant denominations), but they are well
aware of what sort of people in their community belong to
which religion. Cults mainly attract people who are socially iso-
lated or lonely, and these individuals are often attracted partic-
ularly by the promise of becoming part of a community or gain-
ing a sense of belonging. Those who form social attachments to
other members of the cult tend to remain, whereas those who
do not form social bonds tend to leave soon. By the same token,
Kirkpatrick and Shaver (1992) have shown multiple links be-
tween religious beliefs and adult attachment styles or relation-
ship patterns.

Thus far we have focused on the broad need itself, but some
specific patterns regarding human sociality may also have im-
plications for other fields. For example, we have noted that an
interesting psychological issue involves the factors that deter-
mine whether previously opposed groups do or do not integrate
into larger wholes (cf. Sherifetal., 1961/1988), such that indi-
viduals redefine their allegiance so as to belong to the new
group. Such reidentifications have been important throughout
history. The Wars of the Roses were finally decided by the battle
of Bosworth, in which Richard III was killed, thereby enabling

Henry Tudor to establish the dynasty that ruled Elizabethan
England; that battle (like others in the conflict) turned on the
dubious loyalty and betrayal of several major groups that were
incompletely merged into their respective sides (Ross, 1976).
The Zulu empire in South Africa was formed by incorporating
many other groups that had been rivals of and neighbors to the
original Zulus, and those new identifications persist even today,
long past the fall of that empire (Morris, 1965). Meanwhile,
however, Angola, Rwanda, and Uganda have suffered repeated
bouts of cruel violence and civil war between former rival
groups that failed to integrate and identify with the national
unity. Nor is this problem unique to Africa; the former Yugo-
slavia provided a vivid example of bitter factional violence re-
emerging after decades of seemingly peaceful coexistence, and
the same goes for Sri Lanka. In the United States, melting pot
ideology has recently gone out of fashion as the nation has be-
gun to accept the problematic reality of multiple, separate en-
claves defined by racial and ethnic backgrounds. In short, it ap-
pears that asking people to redefine their belongingness to ac-
commodate new realities is difficult and only sometimes
successful.

These applications are not intended as exhaustive, nor even
as the most compelling or important. They are merely intended
as an indication that the need to belong could be used as an
interpretive construct outside of psychological laboratories.

Concluding Remarks

At present, it seems fair to conclude that human beings are
fundamentally and pervasively motivated by a need to belong,
that is, by a strong desire to form and maintain enduring inter-
personal attachments. People seek frequent, affectively positive
interactions within the context of long-term, caring relation-
ships. As a speculative point of theory or impressionistic obser-
vation, the need to belong is not a new idea; indeed, we noted a
variety of previous psychological theorists who have proposed it
in one form or another. What is new, however, is the existence of
a large body of empirical evidence with which to evaluate that
hypothesis.

If psychology has erred with regard to the need to belong, in
our view, the error has not been to deny the existence of such a
motive so much as to underappreciate it. This review has shown
multiple links between the need to belong and cognitive pro-
cesses, emotional patterns, behavioral responses, and health and
well-being. The desire for interpersonal attachment may well be
one of the most far-reaching and integrative constructs cur-
rently available to understand human nature.

References

Adler, P. (1980). On becoming a prostitute. In E. Muga (Ed.), Studies
in prostitution (pp. 22-26). Nairobi, Kenya:’ Kenya Literature
Bureau.

Ainsworth, M. D. (1989). Attachments beyond infancy. American Psy-
chologist, 44, 709-716.

Anderson, C. A. (1991). How people think about causes: Examination
of the typical phenomenal organization of attributions for success and
failure. Social Cognition, 9, 295-329.

Anderson, S. A., Russell, C. S., & Schumm, W. R. (1983). Perceived

THE NEED TO BELONG 523

marital quality and family life-cycle categories: A further analysis.
Journal of Marriage and the Family, 45, 127-139.

Argyle, M. (1987). The psychology of happiness. London: Methuen.
Armstrong, J. G., & Roth, D. M. (1989). Attachment and separation

in eating disorders: A preliminary investigation. International Jour-
nal of Eating Disorders, 8, 141-155.

Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relation-
ships as including other in the self. Journal of Personality and Social
Psychology. 60, 241-253.

Audy, J. R. (1980). Man the lonely animal: Biological roots of loneli-
ness. In J. Hartog, J. R. Audy, & Y. A. Cohen (Eds.), The anatomy of
loneliness (pp. 111-128). New York: International Universities Press.

Axelrod, R., & Hamilton, W. D. (1981). The evolution of cooperation.
Science. 211, 1390-1396.

Barash. D. P. (1977). Sociobiology and behavior. New York: Elsevier.
Barchas, P. (1986). A sociophysiological orientation to small groups. In

E. Lawler (Ed.), Advances in group processes (Vol. 3, pp. 209-246).
Greenwich, CT: JAI Press.

Barden. R. C, Garber, J., Leiman, B., Ford, M. E., & Masters, J. C.
( 1 9 8 5 ) . Factors governing the effective remediation of negative affect
and its cognitive and behavioral consequences. Journal of Personality
and Social Psychology, 49, 1040-1053.

Barocas, R., & Vance, F. L. (1974). Physical appearance and personal
adjustment counseling. Journal of Counseling Psychology, 21, 96-
100.

Baumeister, R. F. (1982). A self-presentational view of social phenom-
ena. Ps,vchological Bulletin, 91, 3-26.

Baumeister, R. F. (1990). Suicide as escape from self. PsychologicalRe-
view, 97, 90-113.

Baumeister, R. F. ( 1 9 9 1 ) . Meanings of life. New York: Guilford Press.
Baumeister, R. F, Stillwell, A. M., & Heatherton, T. F. (1994). Guilt:

An interpersonal approach. Psychological Bulletin, 115, 243-267.
Baumeister, R. F., Stillwell, A. M., & Heatherton, T. F. (in press). Per-

sonal narratives about guilt: Role in action control and interpersonal
relationships. Basic and Applied Social Psychology.

Baumeister, R. F, & Tice, D. M. (1990). Anxiety and social exclusion.
Journal oj’Social and Clinical Psychology, 9, 165-195.

Baumeister, R. F, & Wotman, S. R. (1992). Breaking hearts: The two
sides of unrequited lave. New York: Guilford Press.

Baumeister, R. F, Wotman, S. R., & Stillwell, A. M. (1993). Unre-
quited love: On heartbreak, anger, guilt, scriptlessness, and humilia-
tion. Journal of Personality and Social Psychology, 64, 377-394.

Baunach, P. J. (1985). Mothers in prison. New Brunswick, NJ: Transac-
tion Books.

Beckman, K.., Marsella, A. J., & Finney, R. (1979). Depression in the
wives of nuclear submarine personnel. American Journal of Psychia-
try, 136, 524-526.

Beckman, L. J. ( 1 9 8 1 ) . Effects of social interaction and children’s rela-
tive inputs on older women’s psychological well-being. Journal of Per-
sonality and Social Psychology, 41, 1075-1086.

Bednarski, R., & Leary, M. R. (1994). Self-esteem and fear of death.
Unpublished manuscript, Wake Forest University, Winston-Salem,
NC.

Belsky, J. (1979). The interrelation of parental and spousal behavior
during infancy in traditional nuclear families: An exploratory analy-
sis. Journal of Marriage and the Family, 41, 749-755.

Belsky, J. ( 1 9 8 1 ) . Early human experience: A family perspective. De-
velopmental Psychology, 17, 3-23.

Berman, W. H. (1988). The role of attachment in the post-divorce ex-
perience. Journal of Personality and Social Psychology, 54, 496-503.

Bernard, J. (1982). The future of marriage. New Haven, CT: Yale Uni-
versity Press.

Berscheid, E., & Walster, E. (1974). Physical attractiveness. In L. Ber-

kowitz (Ed.), Advances in experimental social psychology (Vol. 7, pp.
158-215). New York: Academic Press.

Bhatti, B., Derezotes, D., Kim, S., & Specht, H. (1989). The association
between child maltreatment and self-esteem. In A. M. Mecca, N. J.
Smelser, & J. Vasconcellos (Eds.), The social importance of self-es-
teem (pp. 24-71). Berkeley: University of California Press.

Billig, M., & Tajfel, H. (1973). Social categorization and similarity in
intergroup behavior. European Journal of Social Psychology, 3, 27-
51.

Bloom, B. L., White, S. W, & Asher, S. J. (1979). Marital disruption as
a stressful life event. In G. Levinger & O. C. Moles (Eds.), Divorce
and separation: Context, causes, and consequences (pp. 184-200).
New York: Basic Books.

Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York:
Basic Books.

Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation anxiety and
anger. New York: Basic Books.

Braudy, L. (1986). The frenzy of renown: Fame and its history. New
York: Oxford University Press.

Brehm, S. S. (1987). Social support and clinical practice. In J. E. Mad-
dux, C. D. Stoltenberg, & R. Rosenwein (Eds.), Social processes in
clinical and counseling psychology (pp. 26-38). New York: Springer-
Verlag.

Breitman, R. (1991). The architect of genocide: Himmler and the final
solution. Hanover, NH: Brandeis University Press.

Brenner, M. W. (1976). Memory and interpersonal relations. Unpub-
lished doctoral dissertation, University of Michigan, Ann Arbor.

Brewer, M. B. (1979). Ingroup bias in the minimal intergroup situation:
A cognitive-motivational analysis. Psychological Bulletin, 86, 307-
324.

Brewer, M. B., & Silver, M. (1978). Ingroup bias as a function of task
characteristics. European Journal of Social Psychology, 8, 393-400.

Bridges, W. (1980). Transitions: Making sense of life’s changes. Read-
ing, MA: Addison-Wesley.

Brown, J. D. (1986). Evaluations of self and others: Self-enhancement
biases in social judgments. Social Cognition, 4, 353-376.

Browning, C. R. (1992). Ordinary men: Reserve Police Battalion 101
and the final solution in Poland. New York: Harper Collins.

Bunker, B. B., Zubek, J. M., Vanderslice, V. J., & Rice, R. W. (1992).
Quality of life in dual-career families: Commuting versus single-resi-
dence couples. Journal of Marriage and the Family, 54, 399-407.

Burgess, E. W., & Locke, H. J. (1945). The family: From institution to
companionship. New York: American Book.

Burkhart, K. (1973). Women in prison. Garden City, NY: Doubleday.
Buss, A. H. (1983). Social rewards and personality. Journal of Person-

ality and Social Psychology, 44, 553-563.
Buss, D. M. (1990). The evolution of anxiety and social exclusion.

Journal of Social and Clinical Psychology, 9, 196-210.
Buss, D. M. (1991). Evolutionary personality psychology. Annual Re-

view of Psychology, 42, 459-491.
Caldwell, M. A., & Peplau, L. A. (1982). Sex differences in same-sex

friendship. Sex Roles, 8, 721-732.
Campbell, A. (1981). The sense of well-being in America. New York:

McGraw-Hill.
Campbell, A., Converse, P. E., & Rodgers, W. L. (1976). The quality of

American life: Perceptions, evaluations, and satisfactions. New York:
Russell Sage Foundation.

Carnelley, K. B., Pietromonaco, P. R., & Jaffe, K. (1994). Depression,
working models of others, and relationship functioning. Journal of
Personality and Social Psychology, 66, 127-140.

Cash, T. F. (1985). Physical appearance and mental health. In J. A.
Graham&A. KJigman(Eds.), The psychology of cosmetic treatments
(pp. 196-216). New York: Praeger.

Cialdini, R. B., Darby, B., & Vincent, J. (1973). Transgression and al-

524 ROY F. BAUMEISTER AND MARK R. LEARY

truism: A case for hedonism. Journal of Experimental Social Psy-
chology, 9, 502-516.

Clark, L. A., & Watson, D. (1988). Mood and the mundane: Relations
between daily life events and self-reported mood. Journal of Person-
ality and Social Psychology, 54, 296-308.

Clark, M. S. (1984). Record keeping in two types of relationships.
Journal of Personality and Social Psychology, 47, 549-557.

Clark, M. S. (1986). Evidence for the effectiveness of manipulations of
communal and exchange relationships. Personality and Social Psy-
chology Bulletin, 12, 414-425.

Clark, M. S., & Mills, J. (1979). Interpersonal attraction in exchange
and communal relationships. Journal of Personality and Social Psy-
chology^?, 12-24.

Clark, M. S., Mills, J., & Corcoran, D. M. (1989). Keeping track of
needs and inputs of friends and strangers. Personality and Social Psy-
chology Bulletin, 15, 533-542.

Clark, M. S., Mills, J., & Powell, M. C. (1986). Keeping track of needs
in communal and exchange relationships. Journal of Personality and
Social Psychology, 51, 333-338.

Clark, M. S., Ouellette, R., Powell, M. C., & Milberg, S. (1987). Recip-
ient’s mood, relationship type, and helping. Journal of Personality
and Social Psychology, 53, 94-103.

Cohen, S., Sherrod, D. R., & Clark, M. S. (1986). Social skills and
the stress-protective role of social support. Journal of Personality and
Social Psychology, 50, 963-973.

Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buff-
ering hypothesis. Psychological Bulletin, 98, 310-357.

Conte, H. R., Weiner, M. B., & Plutchik, R. (1982). Measuring death
anxiety: Conceptual, psychometric, and factor-analytic aspects.
Journal of Personality and Social Psychology, 43, 775-785.

Coon, C. S. (1946). The universality of natural groupings in human
societies. Journal of Educational Sociology, 20, 163-168.

Costa, P. T., & McCrae, R. R. (1980). Influence of extraversion and
neuroticism on subjective well-being: Happy and unhappy people.
Journal of Personality and Social Psychology, 38, 668-678.

Costa, P. T., & McCrae, R. R. (1984). Personality as a lifelong determi-
nant of wellbeing. In C. Z. Malatesta & C. E. Izard (Eds.), Emotion
in adult development (pp. 141-157). Beverly Hills, CA: Sage.

Costa, P. T., McCrae, R. R., & Zonderman, A. B. (1987). Environmen-
tal and dispositional influences on well-being: Longitudinal follow-
up of an American national sample. British Journal of Psychology,
78, 299-306.

Coyne, J. C., & DeLongis, A. (1986). Going beyond social support: The
role of social relationships in adaptation. Journal of Consulting and
Clinical Psychology, 54,454-460.

Craighead, W. E., Kimball, W. H., & Rehak, R. S. (1979). Mood
changes, physiological responses, and self-statements during social
rejection imagery. Journal of Consulting and Clinical Psychology, 47,
385-396.

Cunningham, M. R., Steinberg, J., & Grev, R. (1980). Wanting to and
having to help: Separate motivations for positive mood and guilt-in-
duced helping. Journal of Personality and Social Psychology, 38,181-
192.

Cutrona, C. E. (1986). Behavioral manifestations of social support: A
microanalytic investigation. Journal of Personality and Social Psy-
chology, 51, 201-208.

Cutrona, C. E. (1989). Ratings of social support by adolescents and
adult informants: Degree of correspondence and prediction of de-
pressive symptoms. Journal of Personality and Social Psychology, 57,
723-730.

Danheiser, P. R., & Graziano, W. G. (1982). Self-monitoring and coop-
eration as a self-presentational strategy. Journal of Personality and
Social Psychology, 42,497-505.

Darley, J. M., & Latane, B. (1968). Bystander intervention in emergen-

cies: Diffusion of responsibility. Journal of Personality and Social
Psychology, 8, 377-383.

DeLongis, A., Folkman, S., & Lazarus, R. S. (1988). The impact of
daily stress on health and mood: Psychological and social resources
as mediators. Journal of Personality and Social Psychology, 54, 486-
495.

de Rivera, J. (1984). The structure of emotional relationships. In
P. Shaver (Ed.), Review of personality and social psychology, Vol. 5:
Emotions, relationships, and health (pp. 116-145). Beverly Hills,
CA: Sage.

Devine, P. G., Sedikides, C., & Fuhrman, R. W. (1989). Goals in social
information processing: The case of anticipated interaction. Journal
of Personality and Social Psychology, 56, 680-690.

DiTommaso, E., & Spinner, B. (1993). The development and initial
validation of the social and emotional loneliness scale for adults. Per-
sonality and Individual Differences, 14, 127-134.

Donne, J. (1975). Devotions upon emergent occasions. Montreal, Can-
ada: McGill Queens University Press.

Douglas, J. D. (1967). The social meanings of suicide. Princeton, NJ:
Princeton University Press.

Drill, R. L. (1987). Young adult children of divorced parents: Depres-
sion and the perception of loss. Journal of Divorce, 10, 169-187.

Durkheim, E. (1963). Suicide. New York: Free Press. (Original work
published 1897)

Egan, G. (1970). Encounter: Group processes for interpersonal growth.
Monterey, CA: Brooks/Cole.

Eisenberger, R. (1970). Is there a deprivation-satiation function for so-
cial approval? Psychological Bulletin, 74, 255-275.

Elder, G. H., & Clipp, E. C. (1988). Wartime losses and social bonding:
Influence across 40 years in men’s lives. Psychiatry, 51, 177-198.

Emery, R. E. (1982). Interparental conflict and the children of divorce.
Psychological Bulletin, 92, 310-330.

Epstein, S. (1992). The cognitive self, the psychoanalytic self, and the
forgotten selves. Psychological Inquiry, 3, 34-37.

Erber, F., & Fiske, S. T. (1984). Outcome dependency and attention to
inconsistent information. Journal of Personality and Social Psychol-
ogy, 47, 709-726.

Farina, A., Burns, G. L., Austad, C., Bugglin, C. S., & Fischer, E. H.
(1986). The role of physical attractiveness in the readjustment of
discharged psychiatric patients. Journal of Abnormal Psychology, 86,
510-517.

Fass, P. (1977). The damned and the beautiful: American youth in the
1920s. Oxford, England: Oxford University Press.

Festinger, L., Schachter, S., & Back, K. (1950). Social pressures in in-
formal groups: A study of a housing community. Palo Alto, CA: Stan-
ford University Press.

Fincham, F. D., Beach, S. R., & Baucom, D. H. (1987). Attribution
processes in distressed and nondistressed couples: 4. Self-partner at-
tribution differences. Journal of Personality and Social Psychology,
52, 739-748.

Forsyth, D. R. (1991). Change in therapeutic groups. In C. R. Snyder
& D. R. Forsyth (Eds.), Handbook of social and clinical psychology
(pp. 664-680). New York: Pergamon Press.

Forsyth, D. R., & Schlenker, B. R. (1977). Attributing the causes of
group performance: Effects of performance quality, task importance,
and future testing. Journal of Personality, 45, 220-236.

Freedman, J. (191 S ) . Happy people: What happiness is, who has it, and
why. New Ybrk: Harcourt Brace Jovanovich.

Freud, S. (1930). Civilization and its discontents (J. Riviere, Trans.).
London: Hogarth Press.

Freud, S. (1956). Totem und Tabu [ Totem and taboo ]. Frankfurt, Ger-
many: Fischer Biicherei. (Original work published 1913)

Fromm, E. (1955). The sane society. New York: Holt, Rinehart &
Winston.

THE NEED TO BELONG 525

Fromm, E. (1956). The art of loving. New York: Harper & Brothers.
Geis, F. L., & Moon, T. H. (1981)..Machiavellianism and deception.

Journal of Personality and Social Psychology, 41, 766-775.
Gerstel, N., & Gross, H. (1982). Commuter marriages: A review. Mar-

riage and Family Review, 5, 71-93.
Gerstel, N., & Gross, H. (1984). Commuter marriage: A study of work

and family. New York: Guilford Press.
Gewirtz, J. L., & Baer, D. M. (1958). Deprivation and satiation of social

reinforcers as drive conditions. Journal of Abnormal and Social Psy-
chology, 57, 165-172.

Giallombardo, R. (1966). Society of women: A study of a women’s
prison. New brk: Wiley.

Glenn, N. D., & McLanahan, S. (1982). Children and marital happi-
ness: A further specification of the relationship. Journal of Marriage
and the Family, 44, 63-72.

Goffman, E. (1971). Relations in public. New York: Harper Colophon.
Goode, W. J. (1956). After divorce. New York: Free Press.
Goodwin, J. S., Hunt, W. C, Key, C. R., & Samet, J. M. (1987). The

effect of marital status on stage, treatment, and survival of cancer
patients. Journal of the American Medical Association, 258, 3125-
3130.

Gottlieb, J., & Carver, C. (1980). Anticipation of future interaction and
the bystander effect. Journal of Experimental Social Psychology, 16,
253-260.

Gouaux, C. ( 1 9 7 1 ) . Induced affective states and interpersonal attrac-
tion. Journal of Personality and Social Psychology, 20, 37-43.

Govaerts, K., & Dixon, D. N. (1988). . . . Until careers do us part:
Vocational and marital satisfaction in the dual-career commuter mar-
riage. International Journal for the Advancement of Counselling, 11,
265-281.

Groth, A. N. (1979). Men who rape. New York: Plenum Press.
Guisinger, S., & Blatt, S. J. (1994). Individuality and relatedness: Evo-

lution of a fundamental dialectic. American Psychologist, 49, 104-
1 1 1 .

Hamachek, D. (1992). Encounters with the self(4lh ed.). San Diego,
CA: Harcourt Brace Jovanovich.

Hardy, C. J., & Grace, R. K. (1991). The effects of task structure and
teammate competence on social loafing. Journal of Sport and Exer-
cise Psychology, 13, 372-381.

Harkins, S. G., & Petty, R. E. (1982). Effects of task difficulty and task
uniqueness on social loafing. Journal of Personality and Social Psy-
chology, 43, 1214-1229.

Harlow, H. F., Harlow, M. K.., & Suomi, S. J. (1971). From thought to
therapy: Lessons from a primate laboratory. American Scientist, 59,
538-549.

Harris, M. (1974). Cows, pigs, wars, and witches: The riddles of culture.
New York: Random House.

Harris, M. (1978). Cannibals and kings: The origins of cultures. New
York: Random House.

Harris, M. (1979). Cultural materialism: The struggle for a science of
culture. New York: Random House.

Harrison, A. A., & Connors, M. M. (1984). Groups in exotic environ-
ments. In L. Berkowitz (Ed.), Advances in experimental social psy-
chology (Vol. 18, pp. 49-87). New York: Academic Press.

Hays, R. B. (1985). A longitudinal study of friendship development.
Journal of Personality and Social Psychology, 48, 909-924.

Hazan, C., & Shaver, P. R. (1994a). Attachment as an organizational
framework for research on close relationships. Psychological Inquiry,
5, 1-22.

Hazan, C., & Shaver, P. R. (1994b). Deeper into.attachment theory.
Psychological Inquiry, 5, 68-79.

Hill, C. T, Rubin, Z., & Peplau, L. A. (1976), Breakups before mar-
riage: The end of 103 affairs. Journal of Social Issues, 32, 147-168.

Hobfall, S. E., & London, P. (1986). The relationship of self-concept

and social support to emotional distress among women during the
war. Journal of Social and Clinical Psychology, 4, 189-203.

Hogan, R. (1983). A socioanalytic theory of personality. In M. Page &
R. Dienstbier (Eds.), Nebraska Symposium on Motivation, 1982 (pp.
55-89). Lincoln: University of Nebraska Press.

Hogan, R., & Jones, W. H. (1983). A role theoretical model of criminal
conduct. In W. S. Laufer & J. M. Days (Eds.), Personality theory,
moral development, and criminal behavior (pp. 3-21). Lexington,
MA: Lexington Books.

Hogan, R., Jones, W. H., & Cheek, J. M. (1985). Socioanalytic theory:
An alternative to armadillo psychology. In B. R. Schlenker (Ed.), The
self and social life (pp. 175-198). New York: McGraw-Hill.

Holmes, J. G., & Rempel, J. K. (1989). Trust in close relationships. In
M. Clark (Ed.), Close relationships: Review of personality and social
psychology (Vol. 10, pp. 187-220). Newbury Park, CA: Sage.

Holmes, T. H., & Rahe, R. H. (1967). The social readjustment rating
scale. Journal of Psychosomatic Research, 11, 213-218.

Holt, P. A., & Stone, G. L. (1988). Needs, coping strategies, and coping
outcomes associated with long-distance relationships. Journal of Col-
lege Student Development, 29, 136-141.

Horney, K. (1945). Our inner conflicts: A constructive theory of neuro-
sis. New York: Norton.

Hotard, S. R., McFatter, R. M., McWhirter, R. M., & Stegall, M. E.
(1989). Interactive effects of extraversion, neuroticism, and social
relationships on subjective well-being. Journal of Personality and So-
cial Psychology, 5 7, 321 -331.

Howard, J. W., & Rothbart, M. (1980). Social categorization and mem-
ory for in-group and out-group behavior. Journal of Personality and
Social Psychology, 38, 301-310.

Hoyle, R. H., & Crawford, A. M. ( i n press). Use of individual-level data
to investigate group phenomena: Issues and strategies. Small Group
Research.

Hoyle, R. H., Pinkley, R. L., & Insko, C. A. (1989). Perceptions of
social behavior: Evidence of differing expectations for interpersonal
and intergroup interaction. Personality and Social Psychology Bulle-
tin, 15, 365-376.

Isenberg, S. (1991). Women who love men who kill. New York: Simon &
Schuster.

Jankowski, M. S. (1991). Islands in the street: Gangs and American
urban society. Berkeley: University of California Press.

Jones, W. H. (1981). Loneliness and social contact. Journal of Social
Psychology, 113, 295-296.

Jones, W. H., & Kugler, K. (in press). Interpersonal correlates of the
guilt inventory. Journal of Personality Assessment.

Jones, W. H., Kugler, K., & Adams, P. (1995). You always hurt the one
you love: Guilt and transgressions against relationship partners. In K.
Fischer & J. Tangner (Eds.), Self-conscious emotion (pp. 301-321).
New York: Guilford Press.

Kaufman,S. R.( 1986). The ageless self: Sources of meaning in late life.
New York: Meridian.

Kenrick, D. T, & Cialdini, R. B. ( 1 9 7 7 ) . Romantic attraction: Misat-
tribution versus reinforcement explanations. Journal of Personality
and Social Psychology, 35, 381-391.

Kenrick, D. T, & Johnson, G. A. (1979). Interpersonal attraction in
aversive environments: A problem for the classical conditioning par-
adigm? Journal of Personality and Social Psychology, 37, 572-579.

Kiecolt-Glaser, J. K., Fisher, L. D., Ogrocki, P., Stout, J. C., Speicher,
C. E., & Glaser, R. (1987). Marital quality, marital disruption, and
immune function. Psychosomatic Medicine, 49, 13-34.

Kiecolt-Glaser, J. K., Garner, W., Speicher, C., Penn, G. M., Holliday,
J., & Glaser, R. (1984). Psychosocial modifiers of immunocompe-
tence in medical students. Psychosomatic Medicine, 46, 7-14.

Kiecolt-Glaser, J. K., Ricker, D., George, J., Messick, G., Speicher,
C. E., Garner, W., & Glaser, R. (1984). Urinary cortisol levels, cellu-

526 ROY F. BAUMEISTER AND MARK R. LEARY

lar immunocompetency, and loneliness in psychiatric inpatients. Psy-
chosomatic Medicine, 46, 15-23.

Kirkpatrick, L. A., & Shaver, P. R. (1992). An attachment-theoretical
approach to romantic love and religious belief. Personality and Social
Psychology Bulletin, 18, 266-275.

Kramer, R. M., & Brewer, M. B. (1984). Effects of group identity on
resource use in a simulated commons dilemma. Journal of Personal-
ity and Social Psychology, 46, 1044-1057.

Kunz, P. R., & Woolcott, M. (1976). Season’s greetings: From my sta-
tus to yours. Social Science Research, 5, 269-278.

Lacoursiere, R. B. (1980). The life cycles of groups: Group developmen-
tal stage theory. New York: Human Sciences Press.

Lakey, B., & Cassady, P. B. (1990). Cognitive processes in perceived
social support. Journal of Personality and Social Psychology, 59, 337-
343.

Larkin, M. (1972). Experiential groups: The uses of interpersonal en-
counter, psychotherapy groups, and sensitivity training. Morristown,
NJ: General Learning Press.

Latane, B., Eckman, J., & Joy, V. (1966). Shared stress and interper-
sonal attraction. Journal of Experimental Social Psychology,
7(Suppl.), 80-94.

Latane, B., & Nida, S. ( 1 9 8 1 ) . Ten years of research on group size and
helping. Psychological Bulletin, 89, 308-324.

Latane, B., Williams, K., & Harkins, S. G. (1979). Many hands make
light the work: The causes and consequences of social loafing. Journal
of Personality and Social Psychology, 37, 822-832.

Lauderdale, P., Smith-Cunnien, P., Parker, J., & Inverarity, J. (1984).
External threat and the definition of deviance. Journal of Personality
and Social Psychology, 46, 1058-1068.

Lawson, A. (1988). Adultery: An analysis of love and betrayal. New
York: Basic Books.

Leary, M. R. (1983). Understanding social anxiety: Social, personality,
and clinical perspectives. Beverly Hills, CA: Sage.

Leary, M. R. (1990). Responses to social exclusion: Social anxiety, jeal-
ousy, loneliness, depression, and low self-esteem. Journal of Social
and Clinical Psychology, 9, 221-229.

Leary, M. R. (1994). Self-presentation: Impression management and
interpersonal behavior. Dubuque, IA: Brown & Benchmark.

Leary, M. R., & Downs, D. L. (in press). Interpersonal functions of the
self-esteem motive: The self-esteem system as a sociometer. In
M. Kernis (Ed.), Efficacy, agency, and self-esteem. New York: Plenum
Press.

Leary, M. R., & Forsyth, D. R. (1987). Attributions of responsibility
for collective endeavors. In C. Hendrick (Ed.), Review of personality
and social psychology. Vol. 8: Group processes (pp. 167-188). New-
bury Park, CA: Sage.

Leary, M. R., Knight, P. D., & Johnson, K. A. (1987). Social anxiety
and dyadic conversation: A verbal response analysis. Journal of So-
cial and Clinical Psychology, 5, 34-50.

Leary, M. R., & Kowalski, R. M. (1990). Impression management: A
literature review and two-component model. Psychological Bulletin,
107, 34-47.

Lewin, K. ( 1 9 5 1 ) . Field theory in social sciences. New York: Harper &
Row.

Lieberman, M. A., Yalom, I. D., & Miles, M. B. (1973). Encounter
groups: First facts. New York: Basic Books.

Linville, P. W., & Jones, E. E. (1980). Polarized appraisals of out-group
members. Journal of Personality and Social Psychology, 38,689-703.

Locksley, A., Ortiz, V., & Hepburn, C. (1980). Social categorization
and discriminatory behavior: Extinguishing the minimal intergroup
discrimination effect. Journal of Personality and Social Psychology,
39. 773-783.

Lofland, L. H. (1982). Loss and human connection: An exploration
into the nature of the social bond. In W. Ickes & E. S. Knowles (Eds.),

Personality, roles, and social behavior (pp. 219-242). New York:
Springer-Verlag.

Lowenthal, M. F, & Haven, C. (1968). Interaction and adaptation: In-
timacy as a critical variable. American Sociological Review, 33, 20-
30.

Lynch, J. J. (1979). The broken heart: The medical consequences of
loneliness. New York: Basic Books.

Mann, L. (1980). Cross-cultural studies of small groups. In H. Triandis
& R. Brislin (Eds.), Handbook of cross-cultural psychology: Social
psychology (Vol. 5, pp. 155-209). Boston: Allyn & Bacon.

Manucia, G. K., Baumann, D. J., & Cialdini, R. B. (1984). Mood in-
fluences on helping: Direct effects or side effects? Journal of Person-
ality and Social Psychology, 46, 357-364.

Maslow, A. H. (1968). Toward a psychology of being. New York: Van
Nostrand.

Masserman, J. H., Wechkin, S., & Terris, W. (1964). “Altruistic” be-
havior in rhesus monkeys. American Journal of Psychiatry, 121, 584-
585.

Mathes, E. W., Adams, H. E., & Davies, R. M. (1985). Jealousy: Loss
of relationship rewards, loss of self-esteem, depression, anxiety, and
anger. Journal of Personality and Social Psychology, 48, 1552-1561.

May, J. L., & Hamilton, P. A. (1980). Effects of musically evoked affect
on women’s interpersonal attraction and perceptual judgments of
physical attractiveness of men. Motivation and Emotion, 4,117-228.

McAdams, D. P. (1985). Motivation and friendship. In S. Duck &
D. Perlman (Eds.), Understanding personal relationships: An inter-
disciplinary approach (pp. 85-105). Beverly Hills, CA: Sage.

McAdams, D. P., & Bryant, F. B. (1987). Intimacy motivation and sub-
jective mental health in a nationwide sample. Journal of Personality,
55,395-413.

McCauley, C. R., & Segal, M. E. (1987). Social psychology of terrorist
groups. In C. Hendrick (Ed.), Group processes and intergroup re-
lations: Review of personality and social psychology (Vol. 9, pp. 231 –
256). Newbury Park, CA: Sage.

McLeod, E. (1982). Women working: Prostitution today. London:
Croom Helm.

Meindl, J. R., & Lerner, M. J. (1984). Exacerbation of extreme re-
sponses to an out-group. Journal of Personality and Social Psychol-
ogy, 47, 71-84.

Miceli, M. (1992). How to make someone feel guilty: Strategies of guilt
inducement and their goals. Journal for the Theory of Social Behav-
iour, 22, 81-104.

Mijuskovic, B. (1980). Loneliness: An interdisciplinary approach. In
J. Hartog, J. R. Audy, & Y. A. Cohen (Eds.), The anatomy of loneli-
ness (pp. 65-94). New York: International Universities Press.

Milardo, R. M., Johnson, M. P., & Huston, T. L. (1983). Developing
close relationships: Changing patterns of interaction between pair
members and social networks. Journal of Personality and Social Psy-
chology, 44, 964-976.

Miller, N. B., Cowan, P. A., Cowan, C. P., Hetherington, E. M., &
Clingempeel, W. G. (1993). Externalizing in preschoolers and early
adolescents: A cross-study replication of a family model. Develop-
mental Psychology, 29, 3-18.

Miller, R. S., & Leary, M. R. (1992). Social sources and interactive
functions of emotion: The case of embarrassment. In M. Clark (Ed.),
Review of personality and social psychology: Vol. 14. Emotion and
social behavior (pp. 202-221). Newbury Park, CA: Sage.

Monson, T. C., Keel, R., Stephens, D., & Genung, V. (1982). Trait at-
tributions: Relative validity, covariation with behavior, and prospect
of future interaction. Journal of Personality and Social Psychology,
42, 1014-1024.

Moreland, R. L. (1987). The formation of small groups. In C. Hendrick
(Ed.), Group processes: Review of personality and social psychology
(Vol. 8, pp. 80-110). Newbury Park, CA: Sage.

THE NEED TO BELONG 527

Moreland, R. L., & Levine, J. M. (1989). Newcomers and oldtimers in
small groups. In P. Paulus (Ed.), Psychology of group influence (Vol.
2, pp. 143-186). Hillsdale, NJ: Erlbaum.

Morgenthau, H. (1962). Love and power. Commentary, 33, 247-251.
Morris, D. R. (1965). The washing of the spears: The rise and fall of the

Zulu nation. New York: Simon & Schuster.
Myers, D. (1992). The pursuit of happiness. New York: Morrow.
Nahemow, L., & Lawton, M. P. (1975). Similarity and propinquity in

friendship formation. Journal of Personality and Social Psychology,
32,205-213.

O’Grady, K. E. (1989). Physical attractiveness, need for approval, so-
cial self-esteem, and maladjustment. Journal of Social and Clinical
Psychology, 8, 62-69.

Olmos, E. (Producer). (1994, April 8). Lives in hazard. New York:
National Broadcasting Company.

Orbell, J. M., van de Kragt, A., & Dawes, R. M. (1988). Explaining
discussion-induced cooperation. Journal of Personality and Social
Psychology, 54, 811-819.

Ostrom, T. M., Carpenter, S. L., Sedikides, C., & Li, F. (1993). Differ-
ential processing of in-group and out-group information. Journal of
Personality and Social Psychology, 64, 21-34.

Paloutzian, R. E, & Janigian, A. S. (1987). Models and methods in
loneliness research: Their status and direction. In M. Hoja & R. Cran-
dall (Eds.), Loneliness: Theory, research, and applications (pp. 31-
36). San Rafael, CA: Select Press.

Panksepp, J., Siviy, S. M., & Normansell, L. A. (1985). Brain opioids
and social emotions. In M. Reite & T. Field (Eds.), Thepsychobiology
of attachment and separation (pp. 3-49). New York: Academic Press.

Pearce, P. L. (1980). Strangers, travelers, and Greyhound terminals:
A study of small-scale helping behaviors. Journal of Personality and
Social Psychology, 38, 935-940.

Pearlman, C. A. (1970). Separation reactions of married women.
American Journal of Psychiatry, 126, 946-950.

Perlman, D. (1987). Further reflections on the present state of loneli-
ness research. In M. Hoja & R. Crandall (Eds.), Loneliness: Theory,
research, and applications (pp. 17-26). San Rafael, CA: Select Press.

Perloff, L. S., & Fetzer, B. K. (1986). Self-other judgments and per-
ceived vulnerability to victimization. Journal of Personality and So-
cial Psychology, 50, 502-510.

Pines, M., & Aronson, E. (1983). Antecedents, correlates, and conse-
quences of sexual jealousy. Journal of Personality, 51, 108-135.

Price, S. J., & McKenry, P. C. (1988). Divorce. Beverly Hills, CA: Sage.
Pryor, J. M., & Ostrom, T. M. (1981). The cognitive organization of

social life: A converging-operations approach. Journal of Personality
and Social Psychology, 41, 628-641.

Rabbie, J. M., & Horwitz, M. (1969). Arousal of ingroup-outgroup
bias by a chance win or loss. Journal of Personality and Social Psy-
chology, 13, 269-277.

Reis, H. T. (1990). The role of intimacy in interpersonal relations.
Journal of Social and Clinical Psychology, 9, 15-30.

Reis, H. T., Wheeler, L., Kernis, M. H., Spiegel, N., & Nezlek, J. (1985).
On specificity in the impact of social participation on physical and
psychological health. Journal of Personality and Social Psychology,
48,456-471.

Reiss, I. L. (1986). A sociological journey into sexuality. Journal of
Marriage and the Family, 48, 233-242.

Rofe, Y. (1984). Stress and affiliation: A utility theory. Psychological
Review, 91, 235-250.

Rogers, C. R. (1959). A theory of therapy, personality, and interper-
sonal relationships, as developed in the client-centered framework. In
S. Koch (Ed.), Psychology: A study of a science (Vol. 3, pp. 184-256).
New York: McGraw-Hill.

Rook, K. S. (1987a). Reciprocity of social exchange and social satisfac-

tion among older women. Journal of Personality and Social Psychol-
ogy, 52, 145-154.

Rook, K. S. (1987b). Social support versus companionship: Effects on
life stress, loneliness, and evaluations by others. Journal of Personal-
ity and Social Psychology, 52, 1132-1147.

Rosen, R. (1979). Some crucial issues concerning children of divorce.
Journal of Divorce, 3, 19-25.

Rosen, S., Mickler, S. E., & Collins, J. E. (1987). Reactions of would-be
helpers whose offer of help is spurned. Journal of Personality and
Social Psychology, 53, 288-297.

Rosenberg, T. (1991). Children of Cain: Violence and the violent in
Latin America. New brk: Penguin Books.

Ross, C. (1976). The Wars of the Roses: A concise history. New York:
Thames & Hudson.

Rothberg, J. M., & Jones, F. D. (1987). Suicide in the U.S. Army: Epi-
demiological and periodic aspects. Suicide and Life-Threatening Be-
havior, 17, 119-132.

Roy, M. (Ed.). (1977). Battered women. New York: Van Nostrand.
Ruehlman, L. S., & Wolchik, S. A. (1988). Personal goals and interper-

sonal support and hindrance as factors in psychological distress and
well-being. Journal of Personality and Social Psychology, 55, 293-
301.

Rusbult, C. E. (1980). Commitment and satisfaction in romantic asso-
ciations: A test of the investment model. Journal of Experimental
Social Psychology, 16, 172-186.

Rusbult, C. E., Verette, J., & Drigotus, S. M. (1994). Absolute commit-
ment level, mutuality of commitment, and couple adjustment in mar-
ital relationships. Unpublished manuscript, University of North Car-
olina at Chapel Hill.

Rusbult, C. E., Zembrodt, I. M., & Gunn, L. K. (1982). Exit, voice,
loyalty, and neglect: Responses to dissatisfaction in’romantic involve-
ments. Journal oj Personality and Social Psychology, 43, 1230-1242.

Russell, D., Cutrona, C., Rose, J., & Yurko, K.. (1984). Social and emo-
tional loneliness: An examination of Weiss’s typology of loneliness.
Journal of Personality and Social Psychology, 46, 1313-1321.

Rutkowski, G. K., Cruder, C. L., & Romer, D. (1983). Group cohesive-
ness, social norms, and bystander intervention. Journal of Personal-
ity and Social Psychology, 44, 545-552.

Rutter, M. (1979). Maternal deprivation, 1972-1978: New findings,
new concepts, new approaches. Child Development, 50, 283-305.

Rutter, M., &Garmezy, N. (1983). Developmental psychopathology. In
E. M. Hetherington (Ed.), Handbook ofchild psychology (Vol. 4, pp.
775-911). New York: Wiley.

Ryan, R. M. (1991). The nature of the self in autonomy and related-
ness. In J. Strauss & G. R. Goethals (Eds.), The self: Interdisciplinary
approaches (pp. 208-238). New York: Springer-Verlag.

Saklofske, D. H., & Yackulic, R. A. (1989). Personality predictors of
loneliness. Personality and Individual Differences, 10, 467-472.

Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways
and turning points through life. Cambridge, MA: Harvard University
Press.

Sarbin, T. (1982). The dangerous individual: An outcome of social
identity transformations. In V. L. Allen & K. E. Scheibe (Eds.), The
social context of conduct (pp. 113-118). New York: Pergamon Press.

Schachter, S. (1951). Deviance, rejection, and communication. Journal
of Abnormal and Social Psychology, 46, 190-207.

Schachter, S. (1959). The psychology of affiliation. Stanford, CA: Stan-
ford University Press.

Schlenker, B. R. (1980). Impression management: The self-concept, so-
cial identity, and interpersonal relations. Monterey, CA: Brooks/
Cole.

Schoenrade, P. A., Batson, C. D., Brandt, J. R., & Loud, R. E. (1986).
Attachment, accountability, and motivation to benefit another not in
distress. Journal of ‘Personality and Social Psychology, 51, 557-563.

528 ROY F. BAUMEISTER AND MARK R. LEARY

Scudder, K. J. (1952). Prisoners are people. Garden City, NY:
Doubleday.

Sedikides, C, Olsen, N., & Reis, H. T. (1993). Relationships as natural
categories. Journal of Personality and Social Psychology, 64, 71-82.

Shaver. P., & Buhrmester, D. (1983). Loneliness, sex-role orientation,
and group life: A social needs perspective. In P. Paulus (Ed.), Basic
group processes (pp. 259-288). New York: Springer-Verlag.

Shaver, P., Hazan, C., & Bradshaw, D. (1988). Love as attachment: The
integration of three behavioral systems. In R. J. Sternberg & M. L.
Barnes (Eds.), The psychology of love (pp. 68-99). New Haven, CT:
Yale University Press.

Shepperd, J. A., & Arkin, R. M. (1990). Shyness and self-presentation.
In W. R. Crozier (Ed.), Shyness and embarrassment: Perspectives
from social psychology (pp. 286-314). Cambridge, England: Cam-
bridge University Press.

Sherif, M., Harvey, O. H., White, B. J., Hood, W. R., & Sherif, C. W.
(1988). The Robbers Cave experiment: Intergroup conflict and coop-
eration. Middletown, CT: Wesleyan University Press. (Original work
published 1961)

Snyder, A. 1. ( 1 9 7 8 ) . Periodic marital separation and physical illness.
American Journal of Orthopsychiatry, 48, 637-643.

Snyder, C. R., Higgins, R. L., & Stucky, R. J. ( 1 9 8 3 ) . Excuses: Mas-
querades in search of grace. New York: Wiley.

Solomon, Z., Waysman, M., & Mikulincer, M. (1990). Family func-
tioning, perceived social support, and combat-related psychopathol-
ogy: The moderating role of loneliness. Journal of Social and Clinical
Psychology. 9, 456-472.

Sours, J. A. (1974). The anorexia nervosa syndrome. International
Journal of Psychoanalysis, 55, 567-576.

Spanier, G. B., & Casto, R. F. ( 1 9 7 9 ) . Adjustment to separation and
divorce: A qualitative analysis. In G. Levinger & O. C. Moles (Eds.),
Divorce and separation: Context, causes, and consequences (pp. 211-
227). New York: Basic Books.

Spanier, G. B., & Lewis, R. A. (1980). Marital quality: A review of the
seventies. Journal of Marriage and the Family, 42, 825-839.

Spivey, E. (1990). Social exclusion as a common factor in social anxiety,
loneliness, jealousy, and social depression. Unpublished master’s the-
sis, Wake Forest University, Winston-Salem, NC.

Stark, R., & Bainbridge, W. S. (1985). The future of religion: Secular-
ization, revival, and cult formation. Berkeley: University of California
Press.

Staub, E. (1989). The roots of evil: The origins of genocide and other
group violence. Cambridge, England: Cambridge University Press.

Staw, B. M., Sandelands, D. E., & Dutton, J. E. (1981). Threat-rigidity
effects in organizational behavior: A multi-level analysis. Administra-
tive Science Quarterly, 26, 501-524.

Stein, A. A. (1976). Conflict and cohesion: A review of the literature.
Journal of Conflict Resolution, 20, 143-172.

Sternberg, R. J. (1986). A triangular theory of love. Psychological Re-
view, 93, 119-135.

Strube, M. J. (1988). The decision to leave an abusive relationship:
Empirical evidence and theoretical issues. Psychological Bulletin,
104, 236-250.

Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New
York: Norton.

Symanski, R. (1980). Prostitution in Nevada. In E. Muga (Ed.), Stud-
ies in prostitution (pp. 246-279). Nairobi, Kenya: Kenya Literature
Bureau.

Tajfel, H. (1970). Experiments in intergroup discrimination. Scientific
American, 223, 96-102.

Tajfel, H., & Billig, M. (1974). Familiarity and categorization in in-
tergroup behavior. Journal of Experimental Social Psychology, 10,
159-170.

Tajfel, H., Flament, C., Billig, M. G., & Bundy, R. F. (1971). Social

categorization and intergroup behaviour. European Journal of Social
Psychology, I, 149-177.

Tambor, E. S., & Leary, M. R. ( 1 9 9 3 ) . Perceived exclusion as a common
factor in social anxiety, loneliness, jealousy, depression, and low self-
esteem. Manuscript submitted for publication.

Tangney, J. P. (1992). Situational determinants of shame and guilt in
young adulthood. Personality and Social Psychology Bulletin, 18,
199-206.

Tesser, A. ( 1 9 9 1 ) . Emotion in social comparison and reflection pro-
cesses. In J. Suls & T. A. Wills (Eds.), Social comparison: Contempo-
rary theory and research (pp. 117-148). Hillsdale, NJ: Erlbaum.

Tesser, A., Millar, M., & Moore, J. (1988). Some affective consequences
of social comparison and reflection processes: The pain and pleasure
of being close. Journal of Personality and Social Psychology, 54, 49-
61.

Tice, D. M., Butler, J. L., Muraven, M. B., & Stillwell, A. M. (1994).
When modesty prevails: Differential favorability of self-presentation
to friends and strangers. Manuscript submitted for publication.

Toch, H. ( 1 9 7 5 ) . Men in crisis. Chicago: Aldine.
Toch, H. ( 1 9 7 7 ) . Living in prison: The ecology of survival. New York:

Free Press.
Toch, H. (1992). Violent men. Washington, DC: American Psychologi-

cal Association.
Toi, M., & Batson, C. D. ( 1 9 8 2 ) . More evidence that empathy is a

source of altruistic motivation. Journal of Personality and Social Psy-
chology, 43, 281-292.

Trout, D. L. (1980). The role of social isolation in suicide. Suicide and
Life-Threatening Behavior, W, 10-23.

Turner, J. C. (1985). Social categorization and the self-concept: A social
cognitive theory of group behavior. In E. J. Lawler (Ed.), Advances in
group processes: Theory and research (Vol. 2, pp. 77-121). Green-
wich, CT: JAI Press.

Vallacher, R. R., & Nowak, A. (1994). The chaos in social psychology.
In R. Vallacher & A. Nowak (Eds.), Dynamical systems in social psy-
chology (pp. 1-16). San Diego, CA: Academic Press.

Vangelisti, A. L., Daly, J. A., & Rudnick, J. R. ( 1 9 9 1 ) . Making people
feel guilty in conversations: Techniques and correlates. Human Com-
munication Research, 18, 3-39.

Vaughan, D. (1986). Uncoupling. New York: Oxford University Press.
Vaux, A. (1988). Social and emotional loneliness: The role of social and

personal characteristics. Personality and Social Psychology Bulletin,
14, 722-734.

Veitch, R., & Griffin, W. (1976). Good news, bad news: Affective and
interpersonal effects. Journal of Applied Social Psychology, 6, 69-75.

Vinokur, A. D., & van Ryn, M. (1993). Social support and undermin-
ing in close relationships: Their independent effects on mental health
of unemployed persons. Journal of Personality and Social Psychol-
ogy, 65, 350-359.

Wade, W. C. (1987). The fiery cross: The Kit Klux Klan in America.
New “York: Touchstone/Simon & Schuster.

Wegner, D. M. (1986). Transactive memory: A contemporary analysis
of the group mind. In B. Mullen & G. R. Goethals (Eds.), Theory of
group behavior (pp. 185-208). New York: Springer-Verlag.

Wegner, D. M., Erber, R., & Raymond, P. ( 1 9 9 1 ) . Transactive memory
in close relationships. Journal of Personality and Social Psychology,
61, 923-929.

Weiss, R. S. (1973). Loneliness: The experience of emotional and social
isolation. Cambridge, MA: MIT Press.

Weiss, R. S. (1979). The emotional impact of marital separation. In
G. Levinger & O. C. Moles (Eds.), Divorce and separation: Context,
causes, and consequences (pp. 201-210). New York: Basic Books.

West, S., Newsom, J. T, & Fenaughty, A. M. (1992). Publication trends
in Journal of Personality and Social Psychology: Stability and change

THE NEED TO BELONG 529

in topics, methods, and theories across two decades. Personality and
Social Psychology Bulletin, 18, 473-484.

Wheeler, L., & Nezlek, J. (1977). Sex differences in social participation.
Journal of Personality and Social Psychology, 35, 742-754.

Wheeler, L., Reis, H. T, & Nezlek, J. (1983). Loneliness, social interac-
tion, and sex roles. Journal of Personality and Social Psychology, 45,
943-953.

Wilbur, J. R., & Wilbur, M. (1988). The noncustodial parent: Dilem-
mas and interventions. Journal of Counseling and Development, 66,
434-437.

Wilder, D. A., & Thompson, J. E. (1980). Intergroup contact with in-
dependent manipulations of in-group and out-group interaction.
Journal of Personality and Social Psychology, 38, 589-603.

Williams, J. G., & Solano, C. H. (1983). The social reality of feeling
lonely: Friendship and reciprocation. Personality and Social Psychol-
ogy Bulletin, 9, 237-242.

Winfield, F. E. (1985). Commuter marriage. New York: Columbia Uni-
versity Press.

Yalom, I. (1985). The theory and practice of group psychotherapy. New
York: Basic Books.

Zander, A. (1971). Motives and goals in groups. New York: Academic
Press.

Received August 27,1993
Revision received November 4, 1994

Accepted November 7, 1994 •

AMERICAN PSYCHOLOGICAL ASSOCIATION
SUBSCRIPTION CLAIMS INFORMATION Today’s Date:.

We provide this form to assist members, institutions, and nonmember individuals with any subscription problems. With the
appropriate information we can begin a resolution. If you use the services of an agent, please do NOT duplicate claims through
them and directly to us. PLEASE PRINT CLEARLY AND IN INK IF POSSIBLE.

PRINT FULL NAME OR KEY NAME OF INSTITUTION

ADDRESS

MEMBER OR CUSTOMER NUMBER (MAY BE FOUND ON ANY PAST ISSUE LABEL)

DATE YOUR ORDER WAS MAILED (OR PHONED)

PREPAID CHECK CHARGE
CHECK/CARD CLEARED DATE:_

CITY STATECOUNTRY

YOUR NAME AND PHONE NUMBER

TITLE

(If possible, send a copy, front and back, of your cancelled check to help us in our research
of your claim.)

ISSUES: MISSING DAMAGED

VOLUME OR YEAR NUMBER OR MONTH

Thank you. Once a claim is received and resolved, delivery of replacement issues routinely takes 4-6 weeks.

^——— (TO BE FILLED OUT BY APA STAFF) ——1—^__———._

DATE RECEIVED:.
ACTION TAKEN: _
STAFF NAME:

DATE OF ACTION: _
INV. NO. & DATE:
LABEL NO. & DATE:

Send this form to APA Subscription Claims, 750 First Street, NE, Washington, DC 20002-4242

PLEASE DO NOT REMOVE. A PHOTOCOPY MAY BE USED.

Intimate
Relationships

NINTH EDITION

Rowland S. Miller
Sam Houston State University

miL04267_fm_i-xvi.indd 1 12/01/21 7:55 PM

INTIMATE RELATIONSHIPS

Published by McGraw Hill LLC, 1325 Avenue of the Americas, New York, NY 10121. Copyright ©2022
by McGraw Hill LLC. All rights reserved. Printed in the United States of America. No part of this
publication may be reproduced or distributed in any form or by any means, or stored in a database or
retrieval system, without the prior written consent of McGraw Hill LLC, including, but not limited to, in
any network or other electronic storage or transmission, or broadcast for distance learning.

Some ancillaries, including electronic and print components, may not be available to customers outside
the United States.

This book is printed on acid-free paper.

1 2 3 4 5 6 7 8 9 LCR 26 25 24 23 22 21

ISBN 978-1-265-63977-8
MHID 1-265-63977-9

Cover Image: National Gallery of Art, Washington

All credits appearing on page or at the end of the book are considered to be an extension of the
copyright page.

The Internet addresses listed in the text were accurate at the time of publication. The inclusion of a
website does not indicate an endorsement by the authors or McGraw Hill LLC, and McGraw Hill LLC
does not guarantee the accuracy of the information presented at these sites.

mheducation.com/highered

miL39779_ise_ii.indd 2 11/01/21 5:08 PM

iii

Contents

PREFACE  ix
ABOUT THE AUTHOR  xii

1. The Building Blocks of Relationships 1

the nature and importance of intimacy 2

the influence of culture 6

the influence of experience 14

the influence of individual differences 20

the influence of human nature 37

the influence of interaction 41

the dark side of relationships 42

for your consideration 42

key terms 43

chapter summary 43

suggestions for satisfaction 45

references 45

2. Research Methods 59

the short history of relationship science 60

developing a question 64

obtaining participants 64

choosing a design 68

the nature of our data 70

the ethics of such endeavors 76

miL04267_fm_i-xvi.indd 3 12/01/21 7:55 PM

iv Contents

interpreting and integrating results 78

a final note 80

for your consideration 80

key terms 81

chapter summary 81

suggestions for satisfaction 82

references 82

3. Attraction 87

the fundamental basis of attraction 87

proximity: liking those near us 88

physical attractiveness: liking those who are lovely 94

reciprocity: liking those who like us 105

similarity: liking those who are like us 107

so, what do men and women want? 116

for your consideration 119

key terms 119

chapter summary 119

suggestions for satisfaction 121

references 121

4. Social Cognition 133

first impressions (and beyond) 133

the power of perceptions 140

impression management 156

so, just how well do we know our partners? 161

for your consideration 166

key terms 166

chapter summary 167

suggestions for satisfaction 168

references 169

miL04267_fm_i-xvi.indd 4 12/01/21 7:55 PM

Contents v

5. Communication 179

nonverbal communication 181

verbal communication 193

dysfunctional communication and what to do about it 203

for your consideration 209

key terms 209

chapter summary 209

suggestions for satisfaction 211

references 211

6. Interdependency 221

social exchange 221

the economies of relationships 229

are we really this greedy? 241

the nature of commitment 249

for your consideration 254

key terms 255

chapter summary 255

suggestions for satisfaction 257

references 257

7. Friendship 266

the nature of friendship 266

friendship across the life cycle 275

differences in friendship 279

friendship difficulties 285

for your consideration 295

key terms 296

chapter summary 296

suggestions for satisfaction 297

references 298

miL04267_fm_i-xvi.indd 5 12/01/21 7:55 PM

vi Contents

8. Love 308

a brief history of love 308

types of love 310

individual and cultural differences in love 327

does love last? 331

for your consideration 335

key terms 336

chapter summary 336

suggestions for satisfaction 337

references 337

9. Sexuality 343

sexual attitudes 343

sexual behavior 348