2007 PatrickKneeCanevelloLonsbary PSPB

Journal of Personality and Social Psychology Copyright 2007 by the American Psychological Association
2007, Vol. 92, No. 3, 434 – 457 0022-3514/07/$12.00 DOI: 10.1037/0022-3514.92.3.434
The Role of Need Fulfillment in Relationship Functioning and Well-Being:

A Self-Determination Theory Perspective
Heather Patrick C. Raymond Knee

University of Rochester University of Houston
Amy Canevello Cynthia Lonsbary

University of Michigan St. Lawrence University
Self-determination theory posits 3 basic psychological needs: autonomy (feeling uncoerced in one’s
actions), competence (feeling capable), and relatedness (feeling connected to others). Optimal well-being
results when these needs are satisfied, though this research has traditionally focused on individual
well-being outcomes (e.g., E. L. Deci & R. M. Ryan, 2000). Three studies examined the role of need
fulfillment in relationship functioning and well-being. Study 1 found that fulfillment of each need
individually predicted both individual and relationship well-being, with relatedness being the strongest
unique predictor of relationship outcomes. Study 2 found that both partners’ need fulfillment uniquely
predicted one’s own relationship functioning and well-being. Finally, in Study 3, the authors used a diary
recording procedure and tested a model in which the association between need fulfillment and relation-
ship quality was mediated by relationship motivation. Those who experienced greater need fulfillment
enjoyed better postdisagreement relationship quality primarily because of their tendency to have more
intrinsic or autonomous reasons for being in their relationship.
Keywords: psychological needs, self-determination theory, relationships
Need theories are distinguishable by their perspective on the needs are acquired. According to Murray, a need is conceptualized
target of needs (i.e., physiological or psychological) and by their as anything that moves an individual to action. Thus, Murray’s list
definition of needs (i.e., nutriments necessary for growth versus of needs is quite extensive and accommodates drives toward
any motivating force). Hull’s (1943) drive theory is one salient positive psychological development (e.g., self-actualization) as
example of a physiological need theory. According to this per- well as drives toward less adaptive functioning (e.g., greed). The
spective, people have a set of innate physiological needs including self-determination theory (SDT; Deci & Ryan, 1985, 2000) per-
food, water, and sex. These physiological needs reflect tissue spective on needs lies somewhere between these two traditions.
deficits, result in drive states, and must be met for the organism to SDT has focused on psychological needs, which is consistent with
remain physically healthy (Hull, 1943). Thus, drive theory identi- the Murray (1938) perspective, and has characterized these needs
fies the target of the needs as physiological and defines the needs as innate, which is consistent with the Hullian tradition. SDT
as necessary nutriments for optimal physiological growth and further defines psychological needs as “nutriments that are essen-
development. The perspective outlined by Murray (1938) has tial for ongoing psychological growth, integrity, and well-being”
provided much of the basis for psychological need theory. In (Deci & Ryan, 2000, p. 229). On the basis of this definition,
contrast to the Hullian tradition, which suggests that needs are satisfaction of basic psychological needs is hypothesized to result
innate, Murray’s perspective on psychological needs suggests that in optimal functioning and well-being.
PSYCHOLOGICAL NEEDS: THE SDT PERSPECTIVE

Heather Patrick, Department of Medicine and Department of Clinical
and Social Sciences in Psychology, University of Rochester; C. Raymond SDT has identified three essential needs for optimal psycholog-
Knee, Department of Psychology, University of Houston; Amy Canevello, ical growth and well-being: competence, relatedness, and auton-
Research Center for Group Dynamics, Institute for Social Research, Uni- omy (Deci & Ryan, 2000). According to SDT, a need for compe-
versity of Michigan; Cynthia Lonsbary, Department of Psychology, St. tence reflects the need to feel effective in one’s efforts and capable
Lawrence University. of achieving desired outcomes. The need for relatedness involves
Study 3 was supported in part by National Institute of Mental Health a need to feel connected to and understood by others. Finally,
Grant MH61805 to C. Raymond Knee. We thank the Interpersonal Rela-
autonomy reflects the need to feel volitional in one’s actions, to
tions and Motivation Research Group at the University of Houston for help
at various stages of this research.
fully and authentically endorse one’s behaviors, and to act as the
Correspondence concerning this article should be addressed to Heather originator of one’s own behavior. This definition of psychological
Patrick, University of Rochester, Department of Clinical and Social Sci- needs and the specification of these particular psychological needs
ences in Psychology, Meliora Hall, P. O. Box 270266, Rochester, NY have been the source of considerable scientific debate. A growing
14627-0266. E-mail: [email protected] body of research based on both the SDT conceptualization of
434
NEED FULFILLMENT AND WELL-BEING 435
psychological needs as well as other perspectives has provided by SDT can be best clarified by comparing it with de Charms’s
evidence for the role of each of these needs in psychological health (1968) distinction between psychological “origins” and “pawns.”
and well-being (Carver & Scheier, 2000; Deci & Ryan, 2000; In contrast with origins, pawns do not feel as if they are the origins
Kernis, 2000). of their behavior, and they do not feel a sense of being fully
A broad literature has demonstrated the importance of ongoing engaged in their actions.
feelings of competence for optimal functioning and well-being.
For example, White (1959) theorized that feeling competent is an NEED FULFILLMENT AND WELL-BEING
integral contributor to self-confidence. Bandura’s (1977) work on
self-efficacy has found that believing that one can bring about SDT prescribes that overall psychological health requires the sat-
desired outcomes is an important determinant of psychological isfaction of all three needs. It is through the satisfaction of these needs
health. In a related vein, Carver and Scheier (1990) have shown that individuals are able to move toward the experience of achieving
that believing that one is effectively making progress toward one’s effectiveness, connectedness, and intrinsic motivation. Much of the
goals is psychologically beneficial. research on SDT’s notion of need fulfillment has found that satisfac-
The need to connect with and feel understood by others is a tion of these three needs is directly associated with well-being and that
distinct human need that is echoed in most theories of human each need contributes uniquely to well-being. This has been tested in
motivation and development (for review, see Reis & Patrick, several ways, including the use of an overall index of need fulfillment,
1996). Baumeister and Leary (1995) referred to this as the need to which aggregates across the three needs, and by testing the unique
belong, and they reviewed extensive evidence demonstrating its contribution of each need to outcomes of interest. For example,
vital role in human motivation. Other researchers have referred to studies incorporating daily recording procedures (e.g., diaries) have
the experience of relatedness as intimacy (Reis & Patrick, 1996; found that daily fluctuations in need fulfillment predict daily fluctu-
Reis & Shaver, 1988). Reis and Patrick (1996) included in their ations in well-being. In one study, Sheldon, Ryan, and Reis (1996)
definition of intimacy the feelings of being understood, validated, looked at both individual differences in perceived need fulfillment as
and cared for, and research has shown that experiencing these well as daily fluctuations in need fulfillment. Sheldon et al. (1996)
aspects of intimacy results in optimum psychological and relation- focused primarily on autonomy and competence needs, and found
ship functioning. Among married couples, Swann, De La Ronde, that, overall, individuals who generally experienced greater fulfill-
and Hixon (1994) found that feeling that one’s partner knows ment of autonomy and competence needs tended to have better days
oneself accurately is associated with experiencing greater inti- on average, as indicated by their tendency to experience more positive
macy. In studies of marital communication, feeling validated by affect and vitality and less negative affect and physical symptoms
one’s partner is associated with better relationship satisfaction (headaches, stomach discomfort, difficulty sleeping, etc.). In addition,
(Gottman, 1994), and feeling cared for by one’s partner is associ- daily fluctuations in need fulfillment were also associated with well-
ated with feelings of greater relationship security (Collins & being such that on days when participants experienced more need
Miller, 1994). Further evidence for the need for relatedness is fulfillment, they also experienced greater well-being. For both
evident in studies involving daily experiences. For example, between-person and within-person findings, autonomy- and
Watson (1988) found that the more opportunities participants had competence-need fulfillment each contributed unique variance in pre-
to interact with important others within a given day, the more dicting individual well-being. In a similar study examining all three
positive affect they experienced. needs, Reis, Sheldon, Gable, Roscoe, and Ryan (2000) reported
The need for autonomy has been perhaps the greatest source of results similar to those reported by Sheldon and colleagues (Sheldon
controversy in SDT’s conceptualization. This controversy has et al., 1996). Thus, both groups of researchers found evidence for the
stemmed largely from misconceptions regarding the definition of association between need fulfillment and well-being at both between-
autonomy (for review, see Ryan & Deci, 2000). From the SDT person and within-person levels of analysis and found an independent
perspective, autonomy refers to self-government or to the extent to contribution of each need to individual well-being.
which people feel self-directed in their actions. However, auton- Researchers examining the link between need fulfillment and
omy has often been confused with independence. This has resulted well-being in specific settings have noted similar patterns of re-
in much debate regarding the seemingly competing presence of sults. In studies of need fulfillment in work settings, employee
autonomy and relatedness needs. From the SDT perspective, au- reports of satisfaction of their needs for autonomy, competence,
tonomy does not involve independence or detachment from others. and relatedness were associated with workers’ self-esteem and
Rather, it involves a sense of volition, agency, and initiative. Thus, overall health (Ilardi, Leone, Kasser, & Ryan, 1993) and with more
fulfillment of one’s need for autonomy does not preclude feeling vitality and less anxiety and physical symptoms (Baard, Deci, &
related to and connected with others. Indeed, Ryan and Lynch Ryan, 2000). These findings have emerged in both American and
(1989) found that autonomy is positively associated with related- Bulgarian samples (Deci et al., 2001). Similarly, among nursing
ness and well-being, and Hodgins, Koestner, and Duncan (1996) home residents, daily autonomy and relatedness-need fulfillment
found that those who functioned more autonomously had more were positively related to well-being and perceived health (Kasser
positive social experiences. Others have implied that this definition & Ryan, 1999; Vallerand & O’Connor, 1989).
of autonomy assumes that people’s behaviors occur completely
independent of influence from the social environment. To the SDT AND CLOSE RELATIONSHIPS
contrary, the SDT perspective on needs suggests that optimal
Self-Determined Motivation and Close Relationships
human functioning in general and need fulfillment in particular
arise out of social contexts that provide nutriments consistent with Much of the research on self-determination in close relation-
these needs. Perhaps the conceptualization of autonomy put forth ships has focused on relationship motivation, though more recent
436 PATRICK, KNEE, CANEVELLO, AND LONSBARY
research has begun to examine need fulfillment in close relation- fulfillment may be one mechanism through which these within-
ships, as described below. Blais, Sabourin, Boucher, and Vallerand person variations in attachment emerge.
(1990) examined the role of self-determined motivation in close Across three studies, strong evidence of within-person variation
relationships as defined by reasons for being in the relationship. in attachment emerged. Attachment was assessed for multiple
They found that being motivated to maintain one’s relationship for attachment figures including friends, parents, romantic partners,
intrinsic or more self-determined reasons was associated with roommates, and other important adult figures (e.g., a teacher), and
reporting more adaptive couple behaviors, which was in turn a substantial proportion of the variation in attachment across these
associated with greater couple happiness. Knee and colleagues attachment figures was attributable to within-person variation.
conceptualized self-determination as growth motivation in rela- More important, within-person variation in attachment was pre-
tionships (Knee, Patrick, Vietor, Neighbors, & Nanayakkara, dicted by need fulfillment. Specifically, participants were more
2002). Similar to Blais et al. (1990), Knee et al. (2002) found that securely attached to those who met their needs for autonomy,
having more intrinsic or self-determined reasons for being in one’s competence, and relatedness (La Guardia et al., 2000). This set of
relationship was associated with more adaptive behaviors, partic- studies represents an important first step to examining the role of
ularly with regard to coping with relationship problems. Those need fulfillment not only in personal well-being but also in rela-
who had more intrinsic reasons for being in their relationship tionship functioning and well-being. Attachment represents a
reported engaging in more active coping strategies such as dis- bridge between these two aspects of well-being because it is in part
cussing problems and finding ways to constructively address prob- about enduring traits of the person and is also in part a response to
lems. More recently, in a series of four studies, Knee, Lonsbary, the social context provided by close relationships. In a related
Canevello, and Patrick (2005) examined the role of self- series of studies, Ryan and colleagues (Ryan et al., 2005) exam-
determined motivation in dealing with relationship conflict. In ined the role of need fulfillment in emotional reliance on others.
Knee et al.’s (2005) Studies 1 and 2, the researchers found evi- Results demonstrated that there is substantial within-person vari-
dence for a hierarchical model of relationship motivation in which ability in emotional reliance across relationship partners and that
general self-determination predicted relationship motivation, emotional reliance is positively associated with need fulfillment
which in turn predicted satisfaction after disagreements (Study 1) such that people are more likely to emotionally rely on someone to
as well as understanding responses to conflict and defensive re- the extent that that person meets one’s needs for autonomy, com-
sponses to conflict (Study 2). In Knee et al.’s (2005) Studies 3 and petence, and relatedness. In addition, need fulfillment was shown
4, the researchers examined the role of both partners’ reasons for to mediate the association between emotional reliance and well-
being in the relationship and found that each partner’s relationship being such that those who sought emotional support from others
motivation contributed uniquely to both self-reported and observed experienced greater well-being outcomes primarily because of
responses to conflict. their tendency to rely on those who met their basic psychological
needs.
Need Fulfillment and Autonomy Support in Close Deci and colleagues (Deci et al., 2006) studied how perceived
Relationships autonomy support within close friendships was related to need
fulfillment and relationship quality variables. Within SDT, receiv-
One of the key assumptions of the SDT perspective on needs is ing autonomy support is theorized to be beneficial because auton-
that need fulfillment arises out of certain optimal social contexts. omy support provides satisfaction of the basic psychological
Thus, developing an understanding of the interplay between close needs. In Study 1, the perception of autonomy support was asso-
relationships and need fulfillment is a natural extension of this line ciated with greater need fulfillment, emotional reliance, attach-
of research. Indeed, recent research has begun to examine the role ment security, dyadic adjustment, and inclusion of the friend in the
of need fulfillment within one’s relationships (Deci, La Guardia, self. There was also evidence of mutuality within friendships such
Moller, Scheiner, & Ryan, 2006; La Guardia, Ryan, Couchman, & that friends perceived similar degrees of autonomy support, need
Deci, 2000; Ryan, La Guardia, Solky-Butzel, Chirkov, & Kim, fulfillment, emotional reliance, attachment security, dyadic adjust-
2005). In one of the first studies to examine the role of need ment, and inclusion of the friend in the self. After controlling for
fulfillment in relationships, La Guardia and colleagues examined dyad-level variance, the associations between perceived autonomy
the role of need fulfillment in attachment (La Guardia et al., 2000). support and the relationship quality variables were no longer
Early research on attachment emphasized the continuity of attach- significant, but the association between received autonomy support
ment patterns from early childhood through adulthood (Bowlby, and need fulfillment remained. In their Study 2, Deci et al. (2006)
1973, 1980). Support for conceptualizing attachment as a trait extended this line of research to examine how both friends’ per-
comes from research demonstrating stability in attachment over ceived autonomy support contributes to each friend’s own rela-
time (e.g., Crittenden, 1990) and similarity in attachment with tionship quality and personal well-being outcomes. Perceived au-
different attachment figures (e.g., parents, teachers, peers; Ryan, tonomy support was assessed in terms of one’s own perceptions of
Stiller, & Lynch, 1994). More recent research has examined both giving and receiving autonomy support. Results showed that
within-person variations in attachment (e.g., Shaver, Collins, & giving autonomy support uniquely predicted need satisfaction and
Clark, 1996) and has suggested that a person’s attachment to a relationship quality above that accounted for by receiving auton-
particular other is in part a function of that person’s general omy support. Regarding well-being outcomes, results indicated
working model of attachment but is also partly about that person’s that giving autonomy support was more strongly related to the
experiences with that individual (e.g., Kobak, 1994). Thus, indi- person’s well-being than was receiving autonomy support.
viduals may have different attachment styles with different rela- In the current research, we sought to further explore the asso-
tionship partners. La Guardia and colleagues proposed that need ciation between need fulfillment and relationship functioning and
well-being, particularly with romantic partners. We were primarily need fulfillment plays a role in one’s own relationship well-being
interested in how need fulfillment is associated with relationship (Study 2); (c) the role of need fulfillment in responses to naturally
quality, as indicated by satisfaction and commitment, and with occurring conflicts in one’s relationships (Study 3); and (d) the
perceptions of and responses to conflict within one’s relationship. mediating role of relationship motivation in the association be-
tween need fulfillment and relationship quality following disagree-
OTHER PERSPECTIVES ON NEED FULFILLMENT ments (Study 3). As in previous research, we also examined the
unique contribution of each need to these outcomes. SDT posits
IN CLOSE RELATIONSHIPS
that fulfillment of all three needs is necessary for optimal personal
Several other perspectives have discussed the importance of well-being, although it is not clear if all three needs are equally
need fulfillment in close relationships. However, many of these important when it comes to indicators of relationship functioning
perspectives have dealt largely with relatedness-type needs, ignor- and well-being. Indeed, La Guardia and colleagues found that
ing the importance of autonomy and competence needs as pre- when the needs were examined individually, each was signifi-
scribed by SDT. For example, as described by Le and Agnew cantly associated with attachment, although relatedness-need ful-
(2001), Drigotas and Rusbult (1992) identified five types of needs fillment was the strongest unique predictor of attachment (La
fulfilled by romantic relationships: intimacy needs, companionship Guardia et al., 2000).
needs, sexual needs, security needs, and emotional involvement In Study 1, we tested the associations between need fulfillment
needs. Intimacy needs relate to confiding in each other, sharing within one’s relationship and indicators of individual (e.g., self-
thoughts, and disclosing feelings. Companionship needs involve esteem) and relationship (e.g., satisfaction) well-being as well as the
spending time and enjoying activities together. Sexual needs in- unique contribution of each need to these outcomes. In Study 2, we
clude the full range of physical relations from holding hands to further explored the role of need fulfillment in relationship well-being
intercourse. Security needs involve relationship stability and the by testing whether one’s partner’s need fulfillment predicts one’s own
extent to which one can depend on the relationship to make life relationship well-being and whether certain combinations of partners’
feel more secure. Emotional involvement needs refer to the degree need fulfillment were particularly beneficial (i.e., the interaction be-
to which one partner’s affective states influence the other partner’s tween each partner’s need fulfillment in predicting outcomes). Fi-
emotional experiences. In SDT, these five needs would all fall nally, in Study 3, we sought to better understand the mechanisms
under relatedness. Thus, this characterization of relationship needs through which need fulfillment is associated with relationship well-
ignores needs for competence and autonomy in the context of being, particularly with regard to satisfaction and commitment fol-
romantic relationships. lowing conflict in daily life. Thus, in Study 3, we tested a model
Prager and Buhrmester (1998) examined whether intimacy con- integrating need fulfillment, relationship motivation, and relationship
tributes to individual need fulfillment. They cluster analyzed 237 quality following disagreements.
human needs derived from the writings of personality theorists and
found that 19 basic needs emerged with overall dimensions of STUDY 1
agentic, communal, and survival needs. On the surface, agentic The purpose of Study 1 was to examine how need fulfillment in
needs appear similar to SDT’s need for autonomy, communal one’s relationship is associated with both individual and relation-
needs appear similar to SDT’s need for relatedness, and survival ship functioning and well-being and to test the unique contribution
needs appear similar to SDT’s need for competence. On closer of each need to these outcomes. Consistent with previous research
examination, however, agentic needs included terms like power on the role of need fulfillment in well-being, indicators of indi-
prestige (e.g., the need to have an impact on others) and avoiding vidual well-being included trait self-esteem, positive and negative
self-esteem loss, which would, if anything, be the opposite of the affect, and vitality. (e.g., Reis et al., 2000; Sheldon et al., 1996).
need for autonomy according to SDT. Further, survival needs Indicators of relationship well-being included relationship quality
included aspects such as relief from anxiety and hunger, which do (satisfaction, commitment), perceptions of conflict, and reported
not seem to be at the same level of abstraction as SDT’s need for responses to conflict. Participants were involved in romantic rela-
competence. Prager and Buhrmester’s (1998) Need Fulfillment tionships at the time of assessment. We examined the association
Inventory, derived from the cluster analyses, settled on agency and between these outcomes and need fulfillment with respect to one’s
communion as the two dimensions of need fulfillment, consistent romantic partner. Several samples of data were gathered, with
with Bakan’s (1966) conception of the principle psychological constructs assessed in at least two and up to eight samples. For
dimensions of human existence. These two needs are conceptually efficiency of presentation and ease of discussion, associations
similar to SDT’s notion of autonomy and relatedness, though the across samples were combined using meta-analytic procedures,
agency and communion perspective fails to take into account correcting for unreliability of measurement.1
competence needs. Prager and Buhrmester (1998) found that inti-
macy predicted need fulfillment, which in turn predicted psycho- Method
logical well-being (e.g., greater life satisfaction, less anxiety).
Samples
OVERVIEW Eight samples that measured need fulfillment in one’s romantic
relationship were included in analyses. Participants in all samples
The current studies go beyond previous research by examining
(a) the role of need fulfillment in relationship well-being including
satisfaction, commitment, perceived conflict, and responses to 1
These data have not been published elsewhere and were gathered by
conflict (Studies 1 and 2); (b) the degree to which one’s partner’s the authors and associated colleagues.
were involved in romantic relationships at the time data were items rated from 1 (not at all true) to 7 (very true). The measure
collected. Table 1 provides descriptive information for each indi- assesses the extent to which participants feel that their romantic
vidual sample and for the combined sample. Sample sizes ranged partner supports their needs for autonomy, competence, and relat-
from 94 to 503, yielding an average of 240 participants per sample edness. An overall need fulfillment score is derived by averaging
and a total of 1,918 observations overall. Across samples, the mean all nine items, with higher scores indicating greater need fulfill-
age was 22 years, and the average relationship length was approx- ment. Subscale scores for autonomy, competence, and relatedness
imately 3 years. On average, 81% of participants in these samples needs are calculated by averaging autonomy, competence, and
were female. This is due in part to the greater likelihood of women relatedness items separately. Sample items are “When I am with
to volunteer to participate in studies on relationships at the uni- my partner, I feel free to be who I am” (autonomy), “When I am
versity where these data were collected. with my partner, I feel like a competent person” (competence), and
“When I am with my partner, I feel loved and cared about”
Meta-Analytic Technique (relatedness). Internal reliabilities (Cronbach’s ␣s) in these sam-
Estimates across samples were meta-analytically combined. ples ranged from .86 to .90 for overall need fulfillment, from .59
This procedure was used to most accurately and parsimoniously to .76 for autonomy, from .67 to .79 for competence, and from .75
estimate the effect size from the available data. In the first set of to .82 for relatedness. Consistent with La Guardia et al. (2000), we
analyses, Pearson’s r was the effect-size index used in the meta- also computed a composite index of autonomy and competence-
analytic estimates (Rosenthal & Rosnow, 1991). The distribution need fulfillment to rule out the possibility that findings for overall
of rs sampled from the population becomes more skewed as r in need fulfillment were simply an artifact of the role of relatedness-
the population increases (Rosenthal, 1994). Thus, Fisher’s Zr need fulfillment in relationship contexts and that findings regard-
transformation was used to correct for the increase in skew. Then, ing the unique role of relatedness-need fulfillment were not simply
each Zr was weighted by the inverse of the sampling error vari- a function of competing variance between autonomy and compe-
ance, which gave greater weight to more precise and reliable tence needs. Internal reliabilities (Cronbach’s ␣s) in these samples
effects resulting from larger, more representative estimates of the ranged from .75 to .85.
population correlation (Hedges & Olkin, 1985). All estimates were
corrected for attenuation due to measurement error of both vari- Individual Well-Being
ables in each effect. This approach was used as recommended by
Hunter and Schmidt (1990, 1994) to come as close as possible to Self-esteem. The Rosenberg (1965) Self-Esteem Scale was
estimating the size of the effect as it would appear under ideal used to assess global self-esteem. The measure consists of 10 items
circumstances. The standard correction following Hunter and answered on a 1 (strongly disagree) to 5 (strongly agree) scale.
Schmidt (1990) was implemented as follows: Items are averaged, and negative items are reverse-scored such that
higher scores indicate higher self-esteem. Internal reliability
ES⬘r ⫽ 关ESr兴/关√rxx ⫻ √ryy兴, (Cronbach’s ␣) ranged from .86 to .90 in the samples reported
where ES⬘r is the adjusted effect r, ESr is the uncorrected effect r, here.
and rxx and ryy are the reliability estimates for each variable in the Emotion. The Positive and Negative Affect Schedule (PA-
correlation. A similar correction was applied to the inverse vari- NAS; Watson, Clark, & Tellegen, 1988) was used to assess pos-
ance estimate. This meta-analytic procedure was repeated for the itive and negative emotion. The measure includes 10 positive and
second set of analyses in which partial correlation ( pr) was the 10 negative adjectives. Participants rated the extent to which they
effect size used. felt each emotion on a 1 (very slightly or not at all) to 5 (extremely)
scale. Internal reliabilities (Cronbach’s ␣s) in these samples ranged
Measures from .86 to .89 for positive affect and from .83 to .89 for negative
affect.
Need Fulfillment Vitality. Participants completed a measure of trait-based sub-
Participants completed the Basic Need Satisfaction in Relation- jective vitality (Ryan & Frederick, 1997). The measure consists of
ships Scale (La Guardia et al., 2000). The measure consists of nine seven items designed to measure general feelings of vitality and
Table 1
Study 1: Descriptive Statistics Overall and by Sample
Mean relationship Mean number of

Sample N % female Mean age length (years) previous relationships
1 204 82 23.1 (4.9) 2.6 (2.5) 1.9 (1.0)

2 163 75 24.5 (6.0) 3.3 (4.1) 1.9 (1.5)
3 94 88 23.2 (7.5) 4.3 (6.7) 1.6 (1.0)
4 368 82 23.0 (5.3) 3.0 (3.8) 2.0 (1.5)
5 217 82 21.1 (3.9) 2.0 (2.1) 1.7 (1.4)
6 113 83 22.2 (4.5) 3.8 (3.9) 1.4 (1.4)
7 503 78 21.1 (4.9) 2.3 (3.2) 1.8 (1.8)
8 266 81 22.6 (5.2) 2.5 (2.6) 1.8 (1.3)
Total 1,918 81 22.6 (4.8) 2.7 (3.3) 1.8 (1.5)
Note. Numbers presented parenthetically are standard deviations for the respective samples.
energy, and items are answered on a 1 (not at all true) to 7 (very and defensive items separately. Sample items for the understand-
true) scale. Sample items include “I feel alive and vital,” and “I ing subscale are “Explore other points of view” and “Understand
look forward to each new day.” Items are averaged, and negative your relationship better.” Sample items for the defensive subscale
items are reverse-scored such that higher scores indicate greater are “Want to leave or walk away,” and “Feel distant or detached
vitality. Internal reliability (Cronbach’s ␣) was .91 in the samples from your partner.” Internal reliabilities (Cronbach’s ␣s) ranged
reported here. from .80 to .84 for understanding responses and from .72 to .80 for
defensive responses.
Attachment Style
There are currently a variety of self-report measures available to Results and Discussion
assess adult attachment. Brennan, Clark, and Shaver (1998) pro-
vided evidence for the validity of a 36-item measure derived from Effect Size: Correlations (Pearson’s r)
a factor analysis of most of the existing self-report measures of
adult romantic attachment. Accordingly, in this study, participants Table 2 provides effect size estimates (Pearson’s r; Rosenthal &
completed the Experiences in Close Relationships (ECR; Brennan Rosnow, 1991) along with the number of samples and participants
et al., 1998) measure of adult attachment. The measure yields two on which each estimate is based. In addition, 95% confidence
subscales, Avoidance and Anxious, which are scored continuously intervals (CIs) were calculated along with significance tests of the
such that all participants receive a score on both dimensions. overall effect estimate. Finally, the range of individual correlations
Internal reliabilities (Cronbach’s ␣s) in the samples reported here from each sample is provided along with the standard error of the
ranged from .90 to .93 for Avoidance and from .90 to .91 for means.
Anxious. Attachment was included in this study because of its As shown, need fulfillment was positively associated with self-
unique role as both a trait and a state (i.e., a characteristic that esteem, positive affect, and vitality and was negatively associated
emerges in response to the context provided in particular relation- with negative affect. For attachment variables, need fulfillment
ships), thus serving as an indicator of both personal and relational was negatively associated with both avoidant and anxious attach-
functioning and well-being. ment. Regarding relationship functioning and well-being variables,
need fulfillment was positively associated with relationship satis-
faction and commitment and with reporting more understanding
Relationship Well-Being responses to conflict. Need fulfillment was also associated with
Satisfaction. Satisfaction was assessed by the Quality of Re- perceiving less conflict and with reporting less defensive responses
lationship Index (QRI), adapted from the Quality of Marriage to conflict. Though not shown in Table 2, a similar pattern of
Index (Norton, 1983). The QRI consists of six Likert-type items findings emerged for the composite of autonomy and competence
that assess the extent to which individuals are satisfied and happy needs and for each of the needs individually.
with their relationship (e.g., “My relationship with my partner Regarding the meta-analytically averaged correlations among
makes me happy”). Items are averaged such that higher scores the need fulfillment subscales, autonomy and relatedness were
reflect higher relationship satisfaction. Internal reliability (Cron- strongly correlated across samples (r ⫽ .86, p ⬍ .0001, 95% CI ⫽
bach’s ␣) in the samples reported here ranged from .85 to .94. .80 to .92, CI z ⫽ 41.83), and competence and relatedness were
Commitment. Commitment was assessed with five items on strongly correlated across samples (r ⫽ .84, p ⬍ .0001, 95% CI ⫽
9-point Likert-type scales (Rusbult, Verette, Whitney, Slovik, & .78 to .90, CI z ⫽ 40.55). Because the correlation between auton-
Lipkus, 1991). The items emphasize both feelings of commitment omy and competence was higher than the reliability of the scale(s)
and likelihood of becoming less committed (e.g., “How likely is it in several samples, we computed an uncorrected meta-analytically
that you will date someone other than your partner in the next averaged correlation between these two subscales that does not
year?”). Items are averaged such that higher scores reflect greater take into account the reliability of the scales. Even with this
commitment. Internal reliability (Cronbach’s ␣) ranged from .88 to conservative approach, autonomy and competence needs were
.92 in the samples reported here. strongly correlated (r ⫽ .68, p ⬍ .0001, CI ⫽ .64 to .73, CI z ⫽
Perceived conflict. Perceptions of conflict were assessed with 36.22). We also examined associations between need fulfillment
13 items rated from 1 (always agree) to 7 (always disagree), based and several other potential covariates. Regarding gender, there was
on the consensus subscale of the Dyadic Adjustment Scale a significant but modest association between need fulfillment and
(Spanier, 1976). Participants reported the degree to which they sex such that women experienced somewhat more overall need
have disagreements with their partner on each of 13 issues (e.g., fulfillment (r⫽ ⫺.10, p ⬍ .001); a similar pattern emerged for
demonstrations of affection). Internal reliability (Cronbach’s ␣) in each of the needs individually. There was also a significant but
these samples ranged from .82 to .88. modest association between overall need fulfillment and age such
Understanding and defensive responses to conflict. Self- that younger individuals experienced somewhat more need fulfill-
reported responses to conflict were assessed with 12 items devel- ment (r ⫽ ⫺.05, p ⬍ .05); a similar correlation emerged for
oped to represent attempts to better understand the conflict or relatedness but not for autonomy or competence. There were no
avoid it. For each item, participants completed the statement, significant correlations between need fulfillment and number of
“After you and your partner have a disagreement or misunder- previous relationships or relationship length; thus, it was unlikely
standing, to what extent do you tend to feel that it led you to that participants experienced greater need fulfillment as a function
__________.” Items were rated from 1 (not at all) to 7 (very of having more relationship experiences or being involved in
much). Subscale scores were created by averaging understanding longer term relationships (all rs ⬍ .04).
Table 2
Study 1: Effect Size Estimates for Meta-Analytically Combined Associations Between Overall
Need Fulfillment and Indicators of Individual and Relationship Well-Being
Characteristic k N r 95% CI CI z Range of r SE of mean
Individual well-being
Self-esteem 8 1,918 .42*** .36 to .47 16.88 .27 to .47 .03
Positive affect 4 955 .42*** .34 to .49 12.05 .32 to .54 .04
Negative affect 4 955 ⫺.43*** ⫺.50 to ⫺.35 12.27 ⫺.30 to ⫺.47 .04
Vitality 2 484 .42*** .32 to .52 8.76 .32 to .44 .05
Attachment
Avoidant 4 1,345 ⫺.56*** ⫺.62 to ⫺.50 20.18 ⫺.43 to ⫺.60 .03
Anxious 4 1,345 ⫺.50*** ⫺.56 to ⫺.44 17.93 ⫺.34 to ⫺.60 .03
Relationship well-being
Relationship quality
Satisfaction 8 1,918 .72*** .67 to .77 35.11 .52 to .73 .03
Commitment 7 1,561 .57*** .52 to .63 22.57 .39 to .63 .03
Perceived conflict 5 1,354 ⫺.61*** ⫺.67 to ⫺.55 22.26 ⫺.49 to ⫺.63 .03
Responses to conflict
Understanding 5 1,467 .34*** .28 to .40 11.54 .23 to .34 .03
Defensiveness 5 1,467 ⫺.51*** ⫺.58 to ⫺.44 17.56 ⫺.33 to ⫺.51 .03
Note. Results were similar for the autonomy– competence composite and the autonomy, competence, and
relatedness subscales. k ⫽ number of samples; N ⫽ total number of observations; r ⫽ correlation coefficient;
95% CI ⫽ 95% confidence interval for partial correlation coefficient; CI z ⫽ z test of the mean effect size.
***
p ⬍ .001.
Effect Size: pr testing the unique contribution of autonomy, competence, and

relatedness-need fulfillment in predicting individual and relation-
The first set of analyses provided evidence for the role of need ship functioning and well-being. The prs reflect the association
fulfillment in individual and relationship well-being as well as between each need and the outcomes, controlling for each of the
attachment. However, these analyses did not address the question other needs. The table provides information on the number of
of the unique contribution of each need to individual and relation- samples and participants on which each estimate is based. In
ship indicators and whether particular needs play a more dominant addition, 95% CIs were calculated along with significance tests of
role in particular outcomes. To address this issue, we calculated the overall effect estimates. For those interested, the range of prs
prs for each sample and meta-analytically combined effect sizes, and the standard error of the means from each sample are provided
as described above. Table 3 provides effect size estimates ( prs) in the Appendix.
Table 3
Study 1: Effect Size Estimates (Partial Correlations: pr) for Meta-Analytically Combined Associations Between Need Fulfillment and
Indicators of Individual and Relationship Well-Being
Autonomy Competence Relatedness
Characteristic k N pr 95% CI CI z pr 95% CI CI z pr 95% CI CI z
Self-esteem 8 1,918 .14*** .08 to .20 4.69 .39*** .33 to .44 14.26 .04 ⫺.20 to .09 1.29
Positive affect 4 955 ⫺.01 ⫺.09 to .07 0.19 .24*** .16 to .32 6.04 .25*** .17 to .32 6.50
Negative affect 4 955 ⫺.17*** ⫺.25 to ⫺.09 4.21 ⫺.19*** ⫺.27 to ⫺.11 4.76 ⫺.16*** ⫺.23 to ⫺.08 4.11
Vitality 2 484 ⫺.00 ⫺.11 to .10 0.08 .21*** .10 to .32 3.72 .21*** .11 to .32 4.02
Attachment
Avoidant 4 1,345 ⫺.17*** ⫺.23 to ⫺.10 4.99 ⫺.01 ⫺.08 to .06 0.30 ⫺.42*** ⫺.49 to ⫺.36 14.07
Anxious 4 1,345 ⫺.17*** ⫺.23 to ⫺.10 4.99 ⫺.13*** ⫺.20 to ⫺.06 3.84 ⫺.24*** ⫺.30 to ⫺.18 7.51
Satisfaction 8 1,918 .11*** .06 to .17 3.94 .09** .03 to .14 3.18 .65*** .59 to .70 28.44
Commitment 7 1,561 .07* .01 to .13 2.11 .05 ⫺.01 to .11 1.56 .59*** .53 to .65 22.45
Perceived conflict 5 1,354 ⫺.21*** ⫺.28 to ⫺.14 5.87 ⫺.12*** ⫺.19 to .05 3.51 ⫺.35*** ⫺.41 to ⫺.28 10.76
Understanding 5 1,467 .07* .01 to .14 2.11 .10** .03 to .17 2.95 .21*** .14 to .28 6.33
Defensive 5 1,467 ⫺.15*** ⫺.22 to ⫺.08 4.09 ⫺.12*** ⫺.19 to .05 3.51 ⫺.22*** ⫺.29 to ⫺.16 6.65
Note. k ⫽ number of samples; N ⫽ total number of observations; 95% CI ⫽ 95% confidence interval for pr; CI z ⫽ z test of the mean effect size.
*
p ⬍ .05. ** p ⬍ .01. *** p ⬍ .001.
As shown, competence-need fulfillment was shown to play the vided when these subscales were entered as separate subscales. To
most consistent and unique role in individual well-being indica- address this possibility, we repeated all of the above analyses using
tors. Competence was positively and uniquely associated with a composite of autonomy and competence in addition to the
self-esteem, positive affect, and vitality and was negatively and relatedness subscale. Results for this set of analyses were nearly
uniquely associated with negative affect. Autonomy was positively identical to the original analyses in which autonomy and compe-
and uniquely associated with self-esteem and was negatively as- tence were treated as separate predictors.
sociated with negative affect. Finally, relatedness-need fulfillment It was also possible that the associations that emerged between
was uniquely and positively associated with positive affect and need fulfillment and some of the relationship variables, notably
vitality. We now turn to the analyses for attachment. Consistent satisfaction and commitment, may have been driven by conceptual
with what La Guardia and her colleagues reported, relatedness- overlap of item content between the need fulfillment measure and
need fulfillment was uniquely and negatively associated with the the measures of satisfaction and commitment incorporated here.
two attachment dimensions, as was autonomy-need fulfillment (La There are both conceptual and empirical reasons for why we
Guardia et al., 2000). Results from analyses testing the unique believe this was not the case. Empirically, with the exception of
contribution of each need to indicators of relationship functioning one extreme value, the range of correlations between need fulfill-
and well-being were also intriguing. As shown in Table 3, each ment and satisfaction and between need fulfillment and commit-
need was uniquely and positively associated with satisfaction and ment are not notably different from the range of correlations
understanding responses to conflict and was uniquely negatively between need fulfillment and any of the other outcome variables,
associated with perceived conflict and defensiveness. Only auton- including those for individual well-being (see Table 2). If over-
omy and relatedness needs were uniquely (positively) associated lapping item content was responsible for the correlations, one
with commitment. would expect correlations between need fulfillment and satisfac-
We were also interested in the magnitude of the contribution of tion and commitment to be larger in magnitude than correlations
each need to these outcomes. Using the Z test procedure for between need fulfillment and other variables. Conceptually, there
nonindependent correlation coefficients described by Meng, is, of course, some similarity between these constructs. Need
Rosenthal, and Rubin (1992), we compared the coefficients for fulfillment, satisfaction, and commitment all tap positive experi-
autonomy versus relatedness-need fulfillment and competence ver- ences in one’s relationship, and one would expect some overlap in
sus relatedness-need fulfillment. Results showed that both auton- these types of experiences. Despite these similarities, there are
omy (Z ⫽ 8.23, p ⬍ .0001) and competence (Z ⫽ 25.33, p ⬍ some important conceptual distinctions worth noting. Most impor-
.0001) were significantly stronger predictors of self-esteem than tant, need fulfillment items address how one feels with one’s
was relatedness-need fulfillment. Relatedness was a stronger pre- partner regarding autonomy, competence, and relatedness (e.g.,
dictor of positive affect (Z ⫽ 14.94, p ⬍ .0001) and vitality (Z ⫽ “When I am with my partner, I am free to be who I am”; “When
8.61, p ⬍ .0001) than was autonomy-need fulfillment. Of partic- I am with my partner I feel like a competent person”; “When I am
ular interest for this study was the magnitude of the contribution of with my partner I feel loved and cared for”). These are conceptu-
relatedness-need fulfillment to relationship-relevant variables. Re- ally distinct items from those assessing the extent to which one is
garding attachment, relatedness-need fulfillment was a stronger happy with one’s relationship more generally (e.g., “My relation-
predictor of anxious attachment than was autonomy (Z ⫽ 4.75, ship with my partner makes me happy”) and the extent to which
p ⬍ .0001) or competence-need fulfillment (Z ⫽ 6.79, p ⬍ .0001). one intends to stay in one’s relationship (e.g., “For how much
The same pattern emerged for avoidant attachment (autonomy, longer do you want your relationship to last?”).
Z ⫽ 16.46, p ⬍ .0001; competence, Z ⫽ 24.70, p ⬍ .0001). Together, these results support much of the existing literature on
Relatedness was also a stronger predictor of satisfaction (auton- the association between need fulfillment and individual well-being
omy, Z ⫽ 39.90, p ⬍ .0001; competence, Z ⫽ 37.67, p ⬍ .0001), and shed light on the role of need fulfillment from one’s partner in
commitment (autonomy, Z ⫽ 35.47, p ⬍ .0001; competence, Z ⫽ relationship functioning and well-being. These results suggest that
33.50, p ⬍ .0001), perceived conflict (autonomy, Z ⫽ 9.34, p ⬍ relatedness-need fulfillment may play a particularly stronger role
.0001; competence, Z ⫽ 14.03, p ⬍ .0001), understanding re- in relationship functioning and well-being.
sponses to conflict (autonomy, Z ⫽ 10.01, p ⬍ .0001; competence,
Z ⫽ 7.13, p ⬍ .0001), and defensive responses to conflict (auton- STUDY 2
omy, Z ⫽ 4.98, p ⬍ .0001; competence, Z ⫽ 6.47, p ⬍ .0001).
Thus, there was clear and convincing evidence that relatedness was Study 1 provided initial evidence for the role of need fulfillment
the strongest unique predictor of relationship functioning and in relationship functioning and well-being. In Study 2, we further
well-being outcomes. explored these associations by examining these processes among
As noted above, autonomy and competence needs were highly couples. It is possible that both partners’ perceived need fulfill-
correlated. In fact, when using a less conservative meta-analytic ment uniquely contributes to relationship functioning and well-
approach, in which the two samples contained higher correlations being such that one benefits not only from experiencing need
between the scales than between the reliabilities of the scales, we fulfillment oneself but also from one’s partner experiencing need
found that the meta-analytically combined correlation between fulfillment. This potential for partners to influence each other is
autonomy and competence was quite high (r ⫽ .96, p ⬍ .0001, one of the defining characteristics of close relationships (Kelley &
CI ⫽ .88 to 1.0, CI z ⫽ 49.29). Therefore, it was possible that the Thibaut, 1978). Thus, in Study 2, we gathered data from both
lack of unique association between autonomy and competence partners using a sample of couples. This design allowed for simul-
needs and many of the outcomes of interest was simply a function taneous estimation of the role of both partners’ need fulfillment in
of shared variance between autonomy and competence being di- relationship functioning and well-being. The Actor–Partner Inter-
dependence Model (APIM; Kashy & Kenny, 2000) allows for the faction and commitment. Finally, internal reliabilities (Cronbach’s
capacity to directly model the mutual influence that may occur ␣s) for perceived conflict, understanding responses to conflict, and
between individuals in a dyadic relationship (Campbell & Kashy, defensive responses to conflict were .88, .81, and .76, respectively.
2002). Further, APIM allows for tests of interactions to determine In this study, we also included a measure of perceived closeness.
whether certain combinations of both partners’ scores on predictor We assessed this with the Inclusion of Other in the Self (IOS)
variables are particularly beneficial for relationship functioning Scale pictorial instrument (Aron, Aron, & Smollan, 1992). The
and well-being. IOS Scale taps aspects of feeling connected and behaving inter-
Deci and colleagues (Deci et al., 2006) examined both giving dependently. The measure consists of a series of two circles
and receiving autonomy support as unique predictors of relation- (labeled self and other) that overlap to equally increasing degrees
ship quality between friends. They hypothesized and found that, in seven stages. Participants selected the picture that best describes
given the reciprocal nature of close relationships, there should be their relationship, and their selection was translated into a score
some degree of mutuality in terms of autonomy support between from 1 to 7, with a higher score reflecting more IOS.
friends. That is, friends’ perceptions of receiving autonomy sup-
port are correlated. More important, they hypothesized and found Results and Discussion
that one’s friend receiving autonomy support benefited one’s own
perceptions of the friendship, including one’s feelings of need Preliminary Analyses
fulfillment within the friendship. In the current research, we sought
Table 4 provides within-couple correlations along the diagonal
to examine similar processes with regard to both partners’ percep-
as well as correlations between variables for men (above the
tions of need fulfillment. Here, we assessed perceived need ful-
diagonal) and women (below the diagonal). As shown, overall
fillment from both members of the dyad, whereas Deci et al.
need fulfillment was positively associated with satisfaction and
(2006) operationalized “autonomy support given” as one’s own
commitment and was negatively associated with perceived conflict
perception of whether one’s friend experiences autonomy support
and defensive responses to conflict for both men and women. The
within the friendship. Deci et al. (2006) hypothesized that giving
same pattern emerged for autonomy, competence, and relatedness
autonomy support was another way in which individuals experi-
needs. Need fulfillment was also positively associated with under-
enced need fulfillment (i.e., an individual giving autonomy support
standing responses to conflict among women, but this correlation
may experience competence in response to the friend receiving
was not statistically significant for men. Within-couple correla-
autonomy support, relatedness as a function of caring for one’s
tions revealed significant associations between partners for overall
friend, and autonomy in response to volitionally doing something
need fulfillment, autonomy, relatedness, satisfaction, perceived
that one valued through giving to the friend). In the same way, we
conflict, and defensive responses to conflict. Partners’
expected that both partners’ need fulfillment would uniquely con-
competence-need fulfillment, commitment, and understanding re-
tribute to one’s own relationship functioning and well-being.
sponses to conflict were not significantly correlated.
Method Analytic Strategy

Participants and Procedure The structure of these data was nested because data were col-
lected from both partners. APIM (Campbell & Kashy, 2002;
Sixty-six couples who had been in their relationship for 1 month Kashy & Kenny, 2000) was used to model the nonindependence of
or longer were recruited from introductory psychology classes. dyadic data and to test whether one’s partner’s need fulfillment
Participants ranged in age from 18 to 55 years (M ⫽ 23.0, SD ⫽ uniquely predicts one’s own relationship functioning and well-
5.4) and had been in their current relationship for an average of 2.5 being. In these analyses, an actor effect occurs when one’s own
years (SD ⫽ 2.0). The sample was ethnically diverse with 39% score on need fulfillment predicts one’s own score on the criterion
Caucasian, 18% Asian, 17% Hispanic, 16% African-American, (e.g., satisfaction); a partner effect occurs when one’s partner’s
and 8% who chose “Other.” With regard to relationship status, score on need fulfillment predicts one’s own score on the criterion.
48% of participants were exclusively dating, 20% were married, We used the PROC MIXED routine in SAS with restricted like-
18% were nearly engaged, 10% were engaged, and 2% were lihood estimation to estimate the coefficients (see Campbell &
casually dating. Participants completed questionnaire packets in a Kashy, 2002). Gender was included to control for potential vari-
Latin square design; they were asked to complete the packets in a ation between men and women. Interactions between actor and
single setting and to answer the questions independently from their partner effects were included in a separate step to examine how the
romantic partner. combination of partners’ need fulfillment was related to relation-
ship quality, perceived conflict, and responses to conflict.
Measures
Main Effects
The measures used in Study 2 were the same as those used in
Study 1 to assess need fulfillment and indicators of relationship The first set of equations included the main effects for actor and
functioning and well-being. Internal reliabilities (Cronbach’s ␣s) partner need fulfillment and gender. PROC MIXED estimates
for need fulfillment were .88, .82, .62, .79, and .76 for overall need coefficients for a single criterion at a time, and thus satisfaction,
fulfillment, the autonomy– competence composite, autonomy, commitment, perceived conflict, understanding responses to con-
competence, and relatedness, respectively. For relationship qual- flict, and defensive responses to conflict were examined sepa-
ity, internal reliabilities (Cronbach’s ␣s) were .90 for both satis- rately. Analyses of actor and partner effects were conducted sep-
Table 4
Study 2: Correlations among Need Fulfillment, Relationship Quality, Perceived Conflict, and Responses to Conflict
Characteristic 1a 1b 1c 1d 2a 2b 3 4a 4b
1. Need fulfillment
1a. Overall .38** .87*** .91*** .91*** .66*** .44*** ⫺.36** .04 ⫺.59***
1b. Autonomy .84*** .30** .69*** .72*** .56*** .40*** ⫺.31** .04 ⫺.51***
1c. Competence .82*** .58*** .06 .75*** .57*** .33** ⫺.35** .00 ⫺.55***
1d. Relatedness .77*** .56*** .36** .49*** .65*** .46*** ⫺.32** .07 ⫺.55***
2. Relationship quality
2a. Satisfaction .41*** .32** .14 .57*** .50*** .56*** ⫺.47*** .15 ⫺.59***
2b. Commitment .33** .24* .19 .38** .48*** .14 ⫺.39*** .01 ⫺.45***
3. Perceived conflict ⫺.50*** ⫺.38** ⫺.28* ⫺.58*** ⫺.61*** ⫺.19 .37** ⫺.50*** .54***
4. Responses to conflict
4a. Understanding .44*** .36** .19 .54*** .39** .24* ⫺.23 .21 ⫺.38**
4b. Defensiveness ⫺.63*** ⫺.52*** ⫺.45*** ⫺.57*** ⫺.52*** ⫺.26* .51*** ⫺.55*** .32**
Note. Values along the diagonal represent within-couple correlations. Values above the diagonal are correlations for men; values below the diagonal are
correlations for women.
*
p ⱕ .05. ** p ⬍ .01. *** p ⬍ .001.
arately for each indicator of need fulfillment (i.e., overall need more satisfied in their relationship to the extent that both partners
fulfillment and autonomy, competence, and relatedness needs in- experienced greater relatedness-need fulfillment.
dividually). Table 5 provides coefficients for actor and partner There was also a significant Actor ⫻ Partner interaction for
effects of overall need fulfillment predicting relationship function- relatedness-need fulfillment when predicting perceived conflict,
ing and well-being. As shown, actor overall need fulfillment was t(62) ⫽ ⫺2.22, p ⬍ .05, pr ⫽ ⫺.27. Figure 2 provides the simple
positively associated with satisfaction and commitment and was regression lines of perceived conflict as a function of actor
negatively associated with perceived conflict and defensiveness. relatedness-need fulfillment at high and low levels of partner
More important, partner overall need fulfillment was uniquely and relatedness-need fulfillment. As shown, tests of simple slopes
positively associated with satisfaction and was negatively associ- indicated that the negative association between actor relatedness-
ated with perceived conflict and defensiveness. A similar pattern
of findings emerged for both actor and partner effects for the
autonomy and competence composite and for the autonomy and 2
Main effects for gender also emerged. Results are reported here for
competence needs individually. Findings for relatedness needs analyses involving overall need fulfillment, although a similar pattern
were somewhat different. Specifically, in addition to the actor emerged for each need individually. Gender was associated with satisfac-
relatedness associations reported for overall need fulfillment, actor tion, perceived conflict, and defensiveness such that women were more
relatedness was positively associated with understanding, and the satisfied than men, t(63) ⫽ ⫺2.24, p ⬍ .05, ␤ ⫽ ⫺.16, pr ⫽ ⫺.27, women
only significant relatedness partner effect was for satisfaction. perceived less conflict than men, t(63) ⫽ 2.87, p ⬍ .01, ␤ ⫽ .23, pr ⫽ .34,
Thus, it appears that both actor and partner need fulfillment and women were more defensive than men, t(63) ⫽ ⫺2.44, p ⬍ .05, ␤ ⫽
uniquely contribute to relationship functioning and well-being, ⫺.15, pr ⫽ .29. There were also a few interactions between actor effects
primarily for autonomy and competence. For relatedness, it ap- and gender. First, there was a significant Gender ⫻ Actor Overall Need
pears that actor perceptions play a more dominant role.2 Fulfillment interaction in predicting understanding, t(94.3) ⫽ ⫺2.53, p ⬍
.05, pr ⫽ ⫺.25. Tests of simple slopes revealed that the association
between actor need fulfillment and understanding responses to conflict was
Interactions significant for women, t(102) ⫽ 3.24, p ⬍ .01, pr ⫽ .31, but not for men,
t(117) ⫽ ⫺0.10, ns. This same pattern of findings emerged for autonomy,
We also tested whether the combination of partners’ perceived
competence, and relatedness needs. There were also two significant Gen-
need fulfillment was related to the outcomes of interest. These der ⫻ Competence-Need Fulfillment interactions: one for actor compe-
analyses yielded several interesting findings, primarily for relat- tence predicting satisfaction, t(110) ⫽ 3.02, p ⬍ .01, pr ⫽ .28, and one for
edness. First, in predicting satisfaction, a significant interaction partner competence predicting defensive responses to conflict, t(119) ⫽
between actor and partner relatedness-need fulfillment emerged, 2.77, p ⬍ .01, pr ⫽ .25. Tests of simple slopes revealed that the association
t(62) ⫽ 3.19, p ⬍ .01, pr ⫽ .37. We selected data points for between actor competence-need fulfillment and satisfaction was significant
estimating regression lines at ⫾ 1 SD for predictors of the equation for men, t(114) ⫽ 5.52, p ⬍ .0001, pr ⫽ .46, but not for women, t(117) ⫽
(Aiken & West, 1991). Figure 1 provides simple regression lines 0.89, ns. In predicting defensive responses to conflict, tests of simple
of satisfaction as a function of actor relatedness-need fulfillment at slopes showed that the association between partner competence-need ful-
high and low levels of partner relatedness-need fulfillment. As fillment and defensive responses to conflict was significant for women,
t(121) ⫽ ⫺4.04, p ⬍ .0001, pr⫽ ⫺.34, but not for men, t(121) ⫽ 0.10, ns.
shown, tests of simple slopes revealed that the association between
Finally, there was a significant Gender ⫻ Actor interaction for relatedness-
actor relatedness-need fulfillment and satisfaction was stronger need fulfillment that predicted perceptions of conflict, t(86.2) ⫽ 1.94, p ⫽
when one’s partner also experienced greater relatedness-need ful- .05, pr ⫽ .20. Tests of simple slopes revealed that the association between
fillment, t(101) ⫽ 6.75, p ⬍ .0001, pr ⫽ .58, relative to when actor relatedness-need fulfillment and perceptions of conflict was signifi-
one’s partner experienced less relatedness-need fulfillment, cant for women, t(116) ⫽ ⫺3.93, p ⬍ .0001, pr ⫽ ⫺.34, but not for men,
t(101) ⫽ 4.18, p ⬍ .0001, pr ⫽ .38, suggesting that partners are t(116) ⫽ ⫺1.50, ns.
Table 5
Study 2: APIM Analyses Testing the Unique Contribution of Actor and Partner Overall Need
Fulfillment to Relationship Variables
Overall need fulfillment
Actor Partner
Characteristic B SE B ␤ pr B SE B ␤ pr
Satisfaction 0.49 .07 .53*** .49 0.29 .07 .31*** .32
Commitment 0.47 .10 .43*** .39 0.05 .10 .05 .05
Perceived conflict ⫺0.37 .01 ⫺.38*** ⫺.35 ⫺0.17 .09 ⫺.17* ⫺.17
Understanding 0.19 .11 .14 .15 0.17 .11 .12 .13
Defensiveness ⫺0.79 .10 ⫺.47*** ⫺.56 ⫺0.21 .10 ⫺.14* ⫺.18
Note. Results were similar for the autonomy– competence composite and for both the autonomy and compe-
tence need fulfillment subscales. For relatedness, there was also a positive association between actor relatedness-
need fulfillment and understanding responses to conflict, and the only significant partner effect for relatedness
needs was in predicting satisfaction.
*
p ⱕ .05. *** p ⬍ .001.
need fulfillment and perceived conflict was stronger when one’s conflict in their relationship when both partners experienced
partner also experienced greater relatedness-need fulfillment, greater relatedness-need fulfillment.
t(95.5) ⫽ ⫺4.14, p ⬍ .0001, pr ⫽ ⫺.39, relative to when one’s Finally, in predicting defensive responses to conflict, there was
partner experienced less relatedness-need fulfillment, t(123) ⫽ a significant Actor ⫻ Partner interaction for relatedness-need
⫺2.11, p ⬍ .05, pr ⫽ ⫺.19. Thus, partners tended to perceive less fulfillment, t(62) ⫽ ⫺2.12, p ⬍ .05, pr ⫽ ⫺.26. Figure 3 provides
6.5
6
Satisfaction
Low Partner NF
5.5
High Partner NF
4.5
4
Low High
Actor Relatedness Need Fulfillment
Figure 1. Actor ⫻ Partner interaction for relatedness-need fulfillment (NF) predicting relationship satisfaction.
3.5
3
Perceived Conflict
Low Partner NF
2.5
High Partner NF
1.5
1
Low High
Figure 2. Actor ⫻ Partner interaction for relatedness-need fulfillment (NF) predicting perceived conflict.
the simple regression lines of defensive responses as a function of Thus, relatedness-need fulfillment accounts for unique variance in
actor relatedness-need fulfillment at high and low levels of partner these outcomes beyond what is accounted for by closeness.
relatedness-need fulfillment. As shown, tests of simple slopes There was also one significant Actor ⫻ Partner interaction for
revealed that the negative association between actor relatedness- competence needs in predicting understanding responses to con-
need fulfillment and defensiveness was stronger when one’s part- flict, t(62) ⫽ ⫺1.99, p ⫽ .05, pr ⫽ ⫺.25. Tests of simple slopes
ner experienced greater relatedness-need fulfillment, t(101) ⫽ revealed that the positive association between actor competence-
⫺5.68, p ⬍ .0001, pr ⫽ ⫺.49, and was weaker when one’s partner need fulfillment and understanding responses to conflict was sig-
experienced less relatedness-need fulfillment, t(125) ⫽ ⫺4.43, nificant when one’s partner had lower feelings of competence-
p ⬍ .0001, pr ⫽ ⫺.37. Thus, partners tended to be particularly less need fulfillment, t(79.7) ⫽ 2.11, p ⬍ .05, pr ⫽ .23, but not when
defensive when both experienced greater relatedness-need fulfill- one’s partner had higher feelings of competence-need fulfillment,
ment. t(90.2) ⫽ ⫺0.97, ns. This is an interesting finding, particularly in
Given the conceptual similarity between relatedness-need ful- light of the fairly consistent pattern of results that emerged for
fillment and perceived closeness, we were concerned that these relatedness needs, in which optimal outcomes emerged when both
findings could be accounted for by a tendency for individuals partners experienced greater need fulfillment. On the basis of this
experiencing greater relatedness-need fulfillment to also perceive finding, it appears that, for competence needs, it is sufficient
more closeness in their relationship. Indeed, for both men and (indeed beneficial) for only one partner to feel that his or her needs
women, relatedness-need fulfillment was significantly correlated are being met.
with closeness (rs ⬎ .27, ps ⬍ .05). The above analyses were thus It is difficult to say why this might be. In examining the
repeated, controlling for the main effects of actor and partner within-couple correlations, competence is the only need on which
perceived closeness as assessed by the IOS Scale (Aron et al., partners seem to be fairly independent (r ⫽ .06). Thus, perceptions
1992). All main effects for the unique contribution of actor and of competence may not be derived from the relationship in similar
partner relatedness-need fulfillment remained significant. In addi- ways for both partners. Although all three needs carry with them
tion, and perhaps more important, the interaction between actor at least some social component in that their fulfillment is in part a
and partner perceived closeness did not account for the significant function of the social context, perhaps competence involves more
Actor ⫻ Partner interactions for relatedness-need fulfillment. self-focus or self-evaluation, and thus, in terms of one’s own
4.5
4
Defensive Responses
Low Partner NF
3.5
High Partner NF
2.5
2
Low High
Figure 3. Actor ⫻ Partner interaction for relatedness-need fulfillment (NF) predicting reported defensive
responses to conflict.
outcomes, it may be beneficial for only oneself to experience the challenges that often arise within close relationships in ways
satisfaction of this need. In contrast, because relatedness has to do that benefit the relationship as a whole, as well as the individuals
with one’s connection to others, it may be particularly important who compose it.
for both partners to experience a high degree of connection to
experience optimal relationship functioning and well-being. An- STUDY 3
other possible explanation may have to do with the nature of the
understanding responses to conflict variable. Perhaps feeling com- Studies 1 and 2 were based on cross-sectional data and focused
petent in one’s relationship provides the efficacy needed to be able exclusively on reported responses to conflict and disagreements. In
to respond to conflict in more constructive ways— by attempting Study 3, we sought to extend these findings to examine the
to understand the conflict and one’s partner better. Whether one’s association between need fulfillment and relationship quality fol-
partner experiences this same sense of efficacy may thus have less lowing disagreements in daily life. In addition, we sought to better
impact on one’s own experience of this dimension of relationship explain the mechanism(s) through which need fulfillment is related
well-being. to these outcomes. One mechanism through which need fulfillment
Overall, findings from Study 2 illustrate the importance of both may be related to relationship functioning and well-being is
partners’ need fulfillment in relationship functioning and well- through its association with relationship motivation. Put another
being. We found evidence for the unique role of both actor and way, being in a relationship that provides the contextual support
partner need fulfillment in predicting satisfaction, perceptions of for one’s needs for autonomy, competence, and relatedness may be
conflict, and defensive responses to conflict. In addition, for associated with experiencing greater intrinsic motivation for being
relatedness-need fulfillment, we found evidence that relationships in the relationship, which in turn is associated with responding to
benefit more in terms of this same set of outcomes when both relationship disagreements in more adaptive ways.
partners experience more relatedness. Thus, when both partners SDT assumes that people have an inherent tendency to move
experience greater relatedness, the relationship may become more toward integration (i.e., assimilating one’s goals, values, and be-
intrinsically rewarding. In light of the findings from La Guardia haviors into a coherent sense of self) and intrinsic motivation.
and colleagues (La Guardia et al., 2000), it may also be that when Rather than assuming that these integrative tendencies are auto-
both partners experience greater relatedness, the relationship has a matic, SDT takes a dialectical view and acknowledges that inte-
stronger, more secure base and partners are thus able to negotiate grative processes depend on the contextual supports for basic
psychological needs (Deci & Ryan, 1991). Thus, need fulfillment more serious disagreements. We chose to define disagreement in
is a prerequisite to optimal functioning. To the extent that an this way because we were primarily interested in examining peo-
individual experiences greater need fulfillment, he or she will also ple’s responses to a range of conflicts experienced in relationships.
experience greater intrinsic motivation. A growing body of re- Each diary record assessed the time and length of discussion, the
search has found that intrinsic motivation is most likely to occur in time the record was completed, and satisfaction and commitment
circumstances that support autonomy, competence, and relatedness immediately following the disagreement. At the end of the 10-day
needs (Deci & Ryan, 1985; Ryan, 1995). Although much of the period, participants completed a follow-up questionnaire to assess
research on SDT has addressed general motivation orientations, perceived accuracy of records.
the theory also specifies that contexts that support (or hinder) need
fulfillment may also be more (or less) conducive to intrinsic
motivation in that particular domain (Ryan, 1995). Intrinsic moti- Measures
vation in relationships has been shown to be associated with
greater agreement between partners and greater relationship satis-
Need Fulfillment
faction (Blais et al., 1990) as well as both reported and observed Perceived need fulfillment was assessed using the same measure
responses to conflict and feelings of satisfaction (Knee et al., described in Study 1. Internal reliabilities (Cronbach’s ␣s) in Study
2005). Previous research has thus demonstrated that need fulfill- 3 were .87, .86, .71, .81, and .74 for overall need fulfillment, the
ment is associated with more intrinsic motivation, and a separate autonomy– competence composite, autonomy, competence, and
body of research has demonstrated that being intrinsically moti- relatedness, respectively.
vated within one’s relationship is associated with a range of
positive relationship outcomes including dyadic functioning, sat-
isfaction, and responses to conflict. However, research has not yet Relationship Motivation
examined these processes simultaneously. These premises involv-
ing the link between need fulfillment and relationship motivation The Couple Motivation Questionnaire (Blais et al., 1990) was
and the link between relationship motivation and positive relation- used to assess relationship motivation in the form of one’s reasons
ship outcomes served as the basis for Study 3. for being in the relationship. The questionnaire begins with the
stem “Why are you in the relationship?” Each of the 18 items then
provides a reason for being in the relationship, and responses are
Method indicated using a 1 (Does not correspond at all) to 7 (Corresponds
exactly) scale. A simplex pattern was evident among the subscales
Participants
such that those reflecting more autonomous or intrinsic reasons
Participants were 120 undergraduates in heterosexual romantic were more positively related to one another and those reflecting
relationships lasting at least 1 month. The sample was ethnically less intrinsic reasons were more positively related to one another.
diverse with 42% Caucasian, 28% Hispanic, 15% African- Further, subscales reflecting more autonomous reasons were neg-
American, 8% Asian, and 7% who chose “Other.” The sample atively related to those reflecting less autonomous reasons. Con-
consisted mostly of individuals in serious dating relationships, sistent with Blais et al. (1990), an index of relationship motivation
with most participants exclusively dating (49%), nearly engaged was computed by weighting the items according to where they fell
(28%), or engaged (5%) and with others casually dating (8%) or on the relative autonomy continuum. (For further details on how
married (10%). The sample was biased toward women with 84% the weights were derived, please see Blais et al., 1990.) Sample
women (n ⫽ 101) and 16% men (n ⫽ 19). The average age of items are, “There is nothing motivating me to stay in my relation-
participants was 21.6 years old (SD ⫽ 3.5 years). Participants had ship with my partner” (weighted –3), “Because people who are
been in their current relationship between 1 month and over 14 important to me are proud of our relationship and I would not want
years (M ⫽ 2.5 years, SD ⫽ 2.3 years). to disappoint them” (weighted –2), “Because I would feel guilty if
I separated from my partner” (weighted –1), “Because this is the
person I have chosen to share life plans that are important to me”
Procedure
(weighted ⫹1), “Because I value the way my relationship with my
Participants were first given a battery of questionnaires in a partner allows me to improve myself as a person” (weighted ⫹2),
Latin square design to measure need fulfillment, relationship mo- and “Because I love the many fun and exciting times I share with
tivation, various demographics, and a variety of other constructs my partner” (weighted ⫹3). An overall index of relationship motiva-
included for other purposes.3 They were then given diary records tion was computed from the weighted subscales, with higher scores
to be completed after each disagreement over a period of 10 days. indicating more autonomous or intrinsic relationship motivation. In-
Disagreement was broadly defined as any interaction in which it ternal reliability (Cronbach’s ␣) in this sample was .76.
was apparent to them that they and their partner disagreed. This
definition was clarified by describing that a disagreement (a) 3
These data are part of a larger dataset on implicit theories of relation-
involves at least some discussion (e.g., they and their partner talk
ships. Portions of these data were described in Knee, Patrick, Vietor, and
about a difference of opinion); (b) involves a difference of opinion Neighbors (2004). Those data were limited to implicit theories of relation-
that includes some sort of interaction, even if for only a few ships as moderators of how experienced conflict is associated with rela-
seconds and even if only verbal (e.g., on the telephone); and (c) is tionship quality. Other portions of these data were described in Knee et al.
not necessarily a major conflict or fight, because we were equally (2005). Those data were limited to the role of trait and relationship
interested in everyday minor differences of opinion as well as autonomy in relationship satisfaction.
IOS interactions with their partner. Items were answered on a 7-point

scale in which 1 is a low anchor (not at all) and 7 is a high anchor
As in Study 2, the IOS Scale (Aron et al., 1992) was included to (very much). In addition, participants were asked to indicate how
assess perceived closeness. many hours per day they interacted with their partner, on average,
over the diary-recording period.
Postdisagreement Satisfaction
Satisfaction after disagreement was assessed on each diary Results and Discussion
record using an abbreviated form of the QRI, adapted from the
Quality of Marriage Index (Norton, 1983), as described in Study 1. Preliminary Analyses
Items were selected on the basis of their relevance to daily inter-
Participants recorded 908 disagreements over the 10-day period,
actions and were modified with the phrase “right now.” The four
with an average of 5.43 per person. On average, disagreements
satisfaction items were “Right now, my relationship with my
lasted approximately 21 min (SD ⫽ 48), and 140 min (SD ⫽ 184)
partner is stable,” “Right now, our relationship is strong,” “Right
elapsed between the time that the disagreement occurred and the
now, my relationship with my partner makes me happy,” and
time that the diary record was completed. Participants did not feel
“Right now, I really feel like part of a team with my partner.”
it was especially difficult to record their disagreements (M ⫽ 2.75,
These abbreviated diary items were averaged (on each record)
SD ⫽ 1.46), felt their diary records were fairly accurate, (M ⫽
such that higher scores reflected higher relationship satisfaction
5.71, SD ⫽ 0.94), estimated that they were able to record 88.46%
after disagreement. Internal reliability (Cronbach’s ␣) in this sam-
of disagreements on average, felt that keeping records did not
ple was .95.
especially increase (M ⫽ 2.09, SD ⫽ 1.39) or decrease (M ⫽ 2.48,
SD ⫽ 1.55) their tendency to have disagreements with their part-
Postdisagreement Commitment ner, and that, on average, they were with their partner 4 –7 hr per
Commitment after disagreement was assessed on each diary day.
record using an abbreviated form of the Rusbult et al. (1991) Preliminary analyses also examined whether need fulfillment
measure described in Study 1. Three items were selected on the was associated with the nature of events recorded. Overall need
basis of their relevance to daily interactions and were modified fulfillment was not significantly associated with number of dis-
with the phrase “right now.” The three commitment items were agreements recorded or discussion length (rs ⬍ .17). Those who
“Right now, for how much longer do you want your relationship to had greater need fulfillment were less likely to have discussed the
last?”, “Right now, do you feel committed to maintaining your issue previously (r ⫽ ⫺.33, p ⬍ .001) and were more likely to
relationship with your partner?”, and “Right now, do you feel perceive that disagreements had been resolved (r ⫽ .28, p ⬍ .01).
attached to your relationship with your partner?” As with the This same pattern of findings emerged for the autonomy–
abbreviated satisfaction measure, these abbreviated diary items competence composite and for autonomy, competence, and relat-
were averaged (on each record) such that higher scores reflected edness needs individually.
higher commitment after disagreement. Internal reliability (Cron- Finally, before conducting multilevel modeling analyses, we
bach’s ␣) in this sample was .94. examined the pattern of correlations among the various predictor
and outcome variables. For each participant, aggregate postdis-
agreement satisfaction and postdisagreement commitment scores
Follow-Up Questionnaire
were computed by averaging across all events. Correlations among
At the end of the 10-day period, participants completed a these variables and with the measures of need fulfillment and
follow-up questionnaire designed to examine factors such as par- relationship motivation are provided in Table 6. As shown, there
ticipant’s perceived accuracy, difficulty of the recording proce- were strong positive correlations between each of the need fulfill-
dure, and how the diary recording procedure may have affected ment variables. Autonomy was significantly correlated with both
Table 6
Study 3: Correlations Among Need Fulfillment, Relationship Motivation, and Aggregated Event
Variables
Characteristic 1a 1b 1c 1d 2 3a 3b
1. Need fulfillment
1a. Overall —
1b. Autonomy .86*** —
1c. Competence .88*** .74*** —
1d. Relatedness .79*** .47*** .52*** —
2. Relationship motivation .62*** .43*** .45*** .70*** —
3. Postdisagreement (aggregated)
3a. Satisfaction .42*** .30*** .30*** .46*** .51*** —
3b. Commitment .42*** .30*** .25** .50*** .64*** .78*** —
Note. Ns ranged from 119 to 120 depending on completeness of data.

**
p ⬍ .01. *** p ⬍ .001.
competence and relatedness, and relatedness and competence were ment and postdisagreement relationship quality may be due, in
also significantly correlated with each other. In addition, and more part, to the role that need fulfillment plays in facilitating intrinsic
germane to the goals of Study 3, each of the need fulfillment relationship motivation. Thus, we proposed a model whereby the
variables was significantly and positively associated with the ag- association between need fulfillment and postdisagreement rela-
gregated postdisagreement satisfaction and commitment variables. tionship quality was mediated by relationship motivation. In test-
Each of the need fulfillment variables was also significantly and ing mediation, we followed Kenny, Kashy, and Bolger’s (1998)
positively associated with relationship motivation. Relationship criteria, which suggest mediation when (a) the predictor (need
motivation was also significantly and positively associated with fulfillment) significantly predicts the criterion (postdisagreement
the aggregated postdisagreement satisfaction and commitment satisfaction, commitment); (b) the predictor significantly predicts
variables. the mediator (relationship motivation); (c) the mediator signifi-
cantly predicts the criterion controlling for the predictor; and (d)
Analytic Strategy the association between the predictor and the criterion, controlling
for the mediator, is substantially reduced or is zero. We further
The structure of the data was such that disagreements were examined the magnitude of the reduction in Step 4 with a modified
nested within persons. Level 1 variables were event variables (i.e., version of the Sobel test (Baron & Kenny, 1986). This modified
postdisagreement satisfaction and commitment on each diary formula includes the addition of the product of the standard errors
record) and were nested within Level 2 person variables (i.e., need of the relevant paths. Thus, a significant Sobel z suggests that the
fulfillment, relationship motivation). All analyses for this study reduction in the association between the predictor (need fulfill-
involved only Level 2 predictors. For a detailed description and ment) and the criterion (postdisagreement satisfaction, commit-
examples of this approach using event-contingent diary data, see ment) with and without controlling for the mediator (relationship
Nezlek (2001). motivation) is reliable. Analyses were conducted for the overall
need fulfillment score and for each of the need fulfillment sub-
Need Satisfaction and Relationship Quality as a Function scales entered simultaneously. In addition, analyses were con-
of Disagreements in Daily Life ducted separately for postdisagreement satisfaction and commit-
ment. Thus, two sets of analyses were conducted for each
Need fulfillment was measured at the trait level, whereas post- postdisagreement outcome.
disagreement satisfaction and commitment were each measured at Need fulfillment and relationship motivation were measured at
the event level. The model for these analyses thus included the the trait level, whereas postdisagreement satisfaction and commit-
fixed effects for the intercept and the slope(s) for need fulfillment ment were each measured at the event level. The model for Step 1
and one random effect for the intercept predicting postdisagree- of the analyses thus included the fixed effects for the intercept and
ment satisfaction and commitment separately. Overall, need ful- the slope(s) for need fulfillment and one random effect for the
fillment significantly predicted postdisagreement satisfaction, F(1, intercept predicting postdisagreement satisfaction and commit-
118) ⫽ 27.50, p ⬍ .0001, ␤ ⫽ .42, pr ⫽ .43, and postdisagreement ment, as described in the previous section. The model for Step 2
commitment, F(1, 118) ⫽ 24.64, p ⬍ .0001, ␤ ⫽ .41, pr ⫽ .42. included the fixed effects for the intercept and the slope(s) for need
Thus, those who had greater need fulfillment felt more satisfied fulfillment predicting relationship motivation and no random ef-
and more committed to their relationship after disagreements, fects. The model for Steps 3 and 4 included the fixed effects for the
relative to other participants. The above analyses were repeated, intercept and the slopes for need fulfillment and relationship
replacing overall need satisfaction with autonomy, competence, motivation and one random effect for the intercept.
and relatedness subscales. Relatedness was the only unique pre-
dictor of postdisagreement satisfaction, F(1, 116) ⫽ 17.55, p ⬍ Overall Need Fulfillment
.0001, ␤ ⫽ .39, pr ⫽ .36, and postdisagreement commitment, F(1,
116) ⫽ 24.55, p ⬍ .0001, ␤ ⫽ .46, pr ⫽ .42. Results were the same Figure 4 presents the path models that summarize the mediation
when the autonomy– competence composite was used in place of analyses for overall need fulfillment. The top portion of the figure
the individual autonomy and competence subscales. Using the Z summarizes analyses for postdisagreement satisfaction, whereas
test procedure for nonindependent correlation coefficients (Meng the bottom portion of the figure summarizes analyses for postdis-
et al., 1992), we compared the coefficients for autonomy versus agreement commitment. Analyses proceeded according to the
relatedness-need fulfillment and for competence versus steps described above (Kenny et al., 1998). First, as described
relatedness-need fulfillment. Once again, relatedness was the above, need fulfillment significantly predicted postdisagreement
strongest predictor of both satisfaction (autonomy, Z ⫽ 1.80, p ⬍ satisfaction and commitment such that those who had greater need
.05; competence, Z ⫽ 2.08, p ⬍ .05) and commitment (autonomy, fulfillment felt more satisfied and more committed to their rela-
Z ⫽ 1.83, p ⬍ .05; competence, Z ⫽ 2.12, p ⬍ .05). tionship after disagreements, relative to other participants. Second,
need fulfillment significantly predicted relationship motivation,
Path Analyses: The Role of Relationship Motivation F(1, 118) ⫽ 75.88, p ⬍ .0001, ␤ ⫽ .62, pr ⫽ .63. Those who had
greater need fulfillment were more intrinsically motivated to be in
We were also interested in examining the potential processes the relationship. Third, relationship motivation significantly pre-
through which need fulfillment is related to postdisagreement dicted postdisagreement satisfaction, F(1, 116) ⫽ 19.84, p ⬍
relationship quality. On the basis of the conceptualization of need .0001, ␤ ⫽ .43, pr ⫽ .38, and postdisagreement commitment, F(1,
satisfaction as a precursor to more intrinsic regulation (Ryan, 116) ⫽ 46.56, p ⬍ .0001, ␤ ⫽ .57, pr ⫽ .54, when need fulfillment
1995), we hypothesized that the association between need fulfill- was included in the equation. Fourth, need fulfillment no longer
Figure 4. Mediation model for need fulfillment predicting postdisagreement satisfaction (a) and commitment
(b). Path coefficients are standardized. ***p ⬍ .001.
significantly predicted postdisagreement satisfaction and commit- satisfaction and commitment. Fourth, none of the needs predicted
ment after controlling for relationship motivation ( ps ⬎ .10). postdisagreement satisfaction or commitment after we controlled
Finally, the magnitude of the reduction in how need fulfillment for relationship motivation ( ps ⬎ .17). Finally, the magnitude of
predicts postdisagreement satisfaction both with (␤ ⫽ .15) and the reduction in how need fulfillment, specifically relatedness,
without (␤ ⫽ .42) relationship motivation in the model was sig- predicts postdisagreement satisfaction both with (␤ ⫽ .15) and
nificant (Sobel z ⫽ 4.33, p ⬍ .0001). A similar pattern emerged without (␤ ⫽ .39) relationship motivation in the model was sig-
with postdisagreement commitment such that the reduction in how nificant (Sobel z ⫽ 3.58, p ⬍ .0001). A similar pattern emerged
need fulfillment predicts postdisagreement commitment both with with postdisagreement commitment such that the reduction in how
(␤ ⫽ .02) and without (␤ ⫽ .41) relationship motivation in the relatedness-need fulfillment predicts postdisagreement commit-
model was significant (Sobel z ⫽ 6.26, p ⬍ .0001). As was the ment both with (␤ ⫽ .10) and without (␤ ⫽ .46) relationship
case in Studies 1 and 2, this pattern of findings also emerged when motivation in the model was significant (Sobel z ⫽ 6.03, p ⬍
the autonomy and competence composite was used in place of .0001). Finally, as in Studies 1 and 2, analyses testing the unique
overall need fulfillment. Thus, general support was found for the contribution of each need to these outcomes were repeated, replac-
notion that people who have greater need fulfillment tend to feel ing the individual autonomy and competence subscales with a
more intrinsically motivated to be in the relationship and in turn composite of the two. Predictors thus included the autonomy–
feel more satisfied and more committed (relative to others) fol- competence composite and the relatedness-need fulfillment sub-
lowing disagreements.
scale. Results for the mediation analyses were the same as those
reported above. These results suggest that, of the three needs,
Need Fulfillment Subscales relatedness-need fulfillment may be particularly important in the
The above analyses were repeated, replacing overall need ful- context of romantic relationships. More specifically, those who
fillment with autonomy, competence, and relatedness subscales. experience more relatedness with their partner have more intrinsic
All three subscales were entered into the equation as simultaneous reasons for being in the relationship, which, in turn, predicts better
predictors. Table 7 provides the coefficients summarizing the relationship well-being (relative to others) in terms of satisfaction
mediation analyses testing the unique contribution of each need in and commitment after disagreements.
predicting postdisagreement satisfaction and commitment. First, We attempted to address several alternative explanations that
relatedness uniquely predicted postdisagreement satisfaction and could account for these findings. First, as noted in the preliminary
commitment. Second, relatedness uniquely predicted relationship analyses section, there was a significant association between need
motivation. Third, after controlling for each of the three needs, fulfillment and perceived resolution of conflict. Thus, it could be
relationship motivation significantly predicted postdisagreement argued that individuals with greater need fulfillment experienced
Table 7
Study 3: Need Fulfillment and Relationship Motivation Predicting Postdisagreement Satisfaction
and Commitment in Daily Life
Satisfaction Commitment
Characteristic B SE B ␤ pr B SE B ␤ pr
Step 1: Need fulfillment predicting postdisagreement
Autonomy 0.12 0.16 .09 .07 0.28 0.22 .16 .12

Competence 0.05 0.17 .04 .03 ⫺0.20 0.23 ⫺.11 ⫺.08
Relatedness 0.50 0.12 .39*** .36 0.80 0.16 .46*** .42
Step 2: Need fulfillment predicting relationship motivation
Autonomy 1.06 1.12 .09 .09 1.06 1.12 .09 .09

Competence 0.65 1.18 .06 .05 0.65 1.18 .06 .05
Relatedness 6.83 0.84 .62*** .60 6.83 0.84 .62*** .60
Steps 3 and 4: Mediation
Autonomy 0.08 0.15 .06 .05 0.20 0.19 .11 .10

Competence 0.01 0.16 .01 .00 ⫺0.28 0.20 ⫺.15 ⫺.13
Relatedness 0.19 0.14 .15 .13 0.18 0.18 .10 .10
Relationship motivation 0.04 0.01 .34*** .33 0.09 0.01 .57*** .48
***
p ⬍ .001.
better postdisagreement relationship quality simply because of lapse as a covariate and (separately) testing the moderating effect
their tendency to believe that the conflict had been resolved. In of lapse. Results remained unchanged when controlling for lapse.
addition, there was a significant association between need fulfill- There were only two significant interactions with lapse: relation-
ment and being less likely to have discussed the issue previously. ship motivation by lapse predicting postdisagreement satisfaction,
Thus, a case could be made that the issues discussed by partici- t(765) ⫽ ⫺2.43, p ⬍ .05, pr ⫽ ⫺.09, and commitment, t(770) ⫽
pants with greater need fulfillment were less serious or less threat- ⫺2.05, p ⬍ .05, pr ⫽ ⫺.07. Tests of simple slopes revealed that
ening as they were less likely to discuss issues that had come up the association between relationship motivation and postdisagree-
repeatedly in their relationship. To address these alternative ex- ment satisfaction was somewhat stronger when less time had
planations, we repeated all of the above analyses, controlling for passed between event occurrence and event recording, t(116) ⫽
these two variables. Results were unchanged. 5.09, p ⬍ .0001, pr ⫽ .43, relative to when more time had passed,
Other possible alternative explanations have to do with the t(116) ⫽ 4.05, p ⬍ .0001, pr ⫽ .35. The same pattern emerged in
method used and the broader experience involved with the record- predicting postdisagreement commitment.
ing procedure. First, those who experienced greater need fulfill- Finally, as in Study 2, we were also interested in testing whether
ment may have had fewer disagreements or may have been less the findings for the unique role of relatedness needs in this medi-
likely to record their disagreements, and thus they were more ation model could be accounted for by perceived closeness. Thus,
likely to remain satisfied and committed. It was also possible that closeness, as assessed by the IOS Scale (Aron et al., 1992), was
those who experienced greater need fulfillment had shorter and included in each step of the analyses. Results were unchanged,
presumably less involved or less serious disagreements and that providing further evidence of the unique role of relatedness-need
this was related to postdisagreement satisfaction and commitment. fulfillment in predicting relationship motivation and, subsequently,
Preliminary analyses suggested no statistically significant associ- more adaptive responses to relationship conflicts.
ation between need fulfillment and these two variables. Nonethe-
less, all of the analyses were repeated, controlling for these two GENERAL DISCUSSION
variables, and all results remained. As noted in the preliminary
analyses section, there was considerable lag between the time that Together, the findings from these three studies highlight the
events occurred and the time that they were reported (M ⫽ 140 importance of need fulfillment in relationship functioning and
min), and there was also substantial variation in the amount of time well-being and the particular relevance of relatedness to these
that passed between the time that events occurred and the time that outcomes. In each study, overall need fulfillment and each indi-
they were recorded (SD ⫽ 184 min; hereafter referred to as vidual need were associated with various aspects of relationship
“lapse”). It is possible that those who experienced greater need functioning and well-being. However, when each need was entered
fulfillment may have had different lapses between event occur- into the equation simultaneously (Studies 1 and 3), relatedness
rence and event recording. It is also possible that the results consistently emerged as the strongest and, in many cases, as the
reported differed on the basis of the amount of time that had passed only unique predictor of these outcomes. Study 1 found that need
between the time of event occurrence and the time of event fulfillment was associated with greater individual well-being (i.e.,
recording. Thus, we repeated the Study 3 analyses, controlling for higher self-esteem, more positive affect, less negative affect, more
vitality), more secure attachment (i.e., less avoidant and less anx- a 10-day period. Thus, we examined the role of need fulfillment in
ious), better relationship quality (i.e., higher satisfaction and com- responses to conflict in daily life, as indicated by postdisagreement
mitment), less perceived conflict, and more adaptive responses to satisfaction and commitment. More important, in Study 3, we
conflict (i.e., more understanding and less defensive). In examin- tested a mediation model to further examine the process by which
ing the unique contribution of each need to these outcomes, a need fulfillment is associated with relationship well-being. Results
consistent pattern emerged suggesting that, although each of the showed that individuals who had greater need fulfillment had more
needs contributed in different ways to indicators of individual autonomous or intrinsic reasons for being in their relationship
well-being, relatedness was the strongest unique predictor of rela- (which were in turn associated with higher satisfaction and com-
tionship functioning and well-being. mitment after disagreements) relative to other participants. When
In Study 2, we included a sample of couples and examined how autonomy, competence, and relatedness needs were entered into
each partner’s need fulfillment contributes to relationship out- the equation simultaneously, relatedness once again emerged as
comes. The findings of Study 2 suggest that the benefits of need the strongest predictor of these outcomes.
fulfillment, specifically as they pertain to relationship functioning The mediation model tested in Study 3, especially the associa-
and well-being, are not limited to one’s own need fulfillment but tion between need fulfillment and intrinsic relationship motivation,
carry over to one’s partner as well. For overall need fulfillment as is particularly interesting in light of much of the theorizing and
well as for competence and autonomy needs, there were significant research dealing with the association between intrinsic motivation
actor and partner effects in predicting satisfaction, perceived con- and relatedness. Traditionally, autonomy and competence needs
flict, and defensive responses to conflict. For relatedness, actor have been shown to be the most powerful influences on intrinsic
need fulfillment predicted satisfaction, commitment, perceived motivation. However, SDT hypothesizes that secure relatedness is
conflict, understanding responses to conflict, and defensive re- needed for intrinsic motivation to flourish (Ryan & La Guardia,
sponses to conflict; the only significant partner effect was in 2000), and there is some evidence to suggest that relatedness
predicting satisfaction. More important, analyses involving the supports integration and intrinsic motivation. For example, when
combination of partners’ need fulfillment (i.e., Actor ⫻ Partner students perceive their teachers as warm and caring, they evidence
interactions) showed that when both partners experienced greater greater intrinsic motivation toward academics (Ryan & Grolnick,
relatedness-need fulfillment, they experienced greater relationship 1986; Ryan et al., 1994). In the current research, relatedness was
satisfaction, perceived less conflict, and reported less defensive the strongest predictor of intrinsic relationship motivation. Thus,
responses to conflict. Thus, when both partners experienced although there may be some situations in which relatedness is less
greater relatedness-need fulfillment, they experienced better rela- central to intrinsic motivation, it appears that, in the context of
tionship functioning and well-being. close relationships and relationship motivation, relatedness may be
The findings of Study 2 are particularly interesting when con- particularly important.
sidering both theoretical conjecture and empirical evidence within Several other theories in the literature on close relationships
the broader SDT framework regarding the benefits of both giving have addressed many of the issues raised here such as the dynam-
and receiving autonomy support (Deci et al., 2006). Theoretically, ics of motivations and filling needs in relationships. For example,
the benefits of autonomy support have been assumed to accumu- self-expansion theory (Aron & Aron, 1996; Aron, Aron, Tudor, &
late because giving and receiving autonomy support helps to meet Nelson, 1991) states that people are motivated to expand their
one’s basic psychological needs. Recent research has confirmed resources, perspectives, and characteristics by including the other
this theoretical supposition and has demonstrated that, among person within the self. In relation to SDT, a few points are worth
close friends, perceiving autonomy support within the relationship noting. First, self-expansion and the sense of closeness that derives
was associated with better relationship quality, including greater from “including another within one’s self” seem to be largely
perceived need fulfillment. More important, the extent to which about SDT’s need for relatedness. However, self-expansion theory
there was mutuality in perceived autonomy support—that is, the does not explicitly address needs for autonomy and competence.
extent to which an individual perceived both that he or she and his Further, although in the current set of findings relatedness and
or her friend experienced autonomy support from the friendship— closeness (as indicated by a measure of including the other in the
accounted for much of the variance in the association between self) were moderately correlated, relatedness provided a unique
perceived autonomy support and relationship quality. The current contribution to the outcomes of interest across studies. This sug-
Study 2 extends these findings to consider both partners’ need gests that, although relatedness and self-expansion have some
fulfillment—a benefit of autonomy support—in relationship func- conceptual similarity, the relatedness need identified by SDT in-
tioning and well-being. In Study 2, we examined perceived need cludes additional aspects of closeness and intimacy not tapped by
fulfillment from the perspective of both partners and demonstrated including the other in the self. Second, to these authors’ knowl-
a pattern of findings consistent with the findings of Deci et al. edge, self-expansion theory does not discuss different qualities of
(2006). When both partners experienced greater need fulfillment self-expansion, whereas SDT explicitly acknowledges and dis-
from their partner (particularly greater fulfillment of their need for cusses healthier, more adaptive ways to expand the self via its
relatedness), individuals experienced better relationships in terms notion of the continuum of integration (and the continuum of
of greater satisfaction, less perceived conflict, and less defensive extrinsic to intrinsic motivations for expanding the self). In other
responses to conflict. words, self-expansion theory suggests that people are motivated to
In Study 3, we sought to examine the mechanisms through expand their resources, perspectives, and characteristics but does
which need fulfillment contributes to relationship functioning and not distinguish more intrinsic versus more extrinsic reasons for
well-being and incorporated a diary procedure in which partici- doing so. Perhaps not all motivations for relating and expanding
pants recorded every disagreement they had with their partner over one’s self are equal. Seeking closeness from a partner to acquire
resources (e.g., fame, approval from others, monetary gains) could an argument or to gain the approval of one’s partner. Reasons for
be a different form of motivation than seeking to learn new enacting prorelationship behaviors may also impact how these
perspectives and grow. This may be particularly important in light behaviors are perceived by one’s partner. If one’s partner perceives
of the mediation model presented in Study 3, which demonstrated that one is enacting these behaviors to “make points” or to gain
that one of the reasons that need fulfillment is associated with something from the partner, prorelationship behaviors may not be
better postdisagreement relationship quality (relative to others) is as beneficial to the relationship, as they are perceived as being
because need fulfillment is associated with being more intrinsi- motivated by one’s own interests and not by one’s focus on the
cally motivated to be in the relationship. Thus, in the context of partner or the relationship. In addition, the SDT perspective would
romantic relationships, feeling that one’s partner meets one’s re- suggest that, to the extent that one’s autonomy, competence, and
latedness needs is beneficial primarily because it may facilitate an relatedness needs are fulfilled, it would be easier and more natural
intrinsic motivation for being in the relationship. As mentioned to take one’s partner and the relationship into account and behave
previously, intrinsic relationship motivation has been associated more interdependently.
with a variety of positive relationship outcomes (Blais et al., 1990; Communal– exchange perspectives (Mills & Clark, 2001) dis-
Knee et al., 2005, 2002). Thus, although SDT in general and the tinguish between a communal orientation and an exchange orien-
relatedness need in particular are consistent with the notion of tation. The former carries the expectation of immediate repayment
self-expansion, the SDT perspective goes beyond what has been for benefits given. The latter concerns the expectation of mutual
proposed by self-expansion theory to suggest that some ways of responsiveness to the others’ needs more generally and with a
seeking self-expansion may lead to more authentic and profound more long-term sense of equity. The motivation implied within the
forms of self-expansion and relatedness than others. communal– exchange framework is important and relates to SDT
Interdependence theory (Kelley & Thibaut, 1978) offers another in that an exchange orientation could seem more likely to go along
perspective from which to consider the current set of findings. with extrinsic motivation, whereas a communal orientation could
Interdependence theory describes how outcomes are negotiated seem more likely to go along with intrinsic motivation. Thus, when
within the interpersonal structure of dyadic situations. According one is expecting relatively immediate “repayment” from one’s
to this perspective, individuals are motivated to maximize personal partner, one is in the mindset of behaving for those more imme-
and relational rewards within the context of relationship decisions diate reciprocal rewards rather than out of true interest in devel-
and behaviors. In so doing, partners transform the decisions that oping the relationship and enjoying it for its own sake. Similarly,
they would initially make (and ways that they would initially when one is motivated to be responsive to the partner’s needs in a
behave) that do not consider the partner’s desires into different more long-term fashion, one seems more motivated by apprecia-
ways of deciding and behaving that do take into account the tion and genuine desire to relate rather than by guilt or obligation.
partner’s desires. As with self-expansion theory, interdependence As mentioned above with reference to interdependence theory,
as a concept seems most fundamentally about the need for relat- when one’s psychological needs are fulfilled, it may be easier and
edness, although it could also be argued that interdependence is more natural to behave out of a communal orientation. Further, a
really about the negotiation between individual outcomes and communal orientation would seem to promote trust and a deeper
relational outcomes. However, it is important to note that individ- sense of intimacy and relatedness. It would seem difficult to be
ual interests and motivations in interdependence theory are not truly responsive to a partner’s needs if one felt that one was being
equivalent to SDT’s notion of autonomy (as described previously forced to do so out of guilt, obligation, or expected repayment
in the introduction). The SDT conceptualization of autonomy is (lack of autonomy fulfillment); if one felt incompetent or inade-
not the same as selfishly pursuing one’s own interests. quate in the relationship (lack of competence fulfillment); or if one
Much of the more recent research involving interdependence felt disconnected from one’s partner or felt as if one’s partner did
and the transformation of motivation in close relationships has not understand something important about oneself (lack of relat-
focused on two prorelationship behaviors: accommodation (choos- edness fulfillment).
ing not to retaliate in the face of a partner’s transgression) and Reis, Clark, and Holmes (2004) proposed “partner responsive-
willingness to sacrifice (foregoing one’s own immediate interests ness” as an important overarching concept in the study of personal
to promote the well-being of one’s partner or relationship). Pro- relationships in general and in closeness and intimacy within those
relationship behaviors have been associated with dyadic adjust- relationships in particular. According to Reis et al. (2004), partner
ment and with a greater probability of couple persistence (Van responsiveness is a process by which people come to believe that
Lange et al., 1997), and when partners perceive prorelationship their romantic partner is attentive to and supportive of core ele-
behaviors, they come to trust each other and rely on the relation- ments of the self and, through this process, partners become closer
ship more (Weiselquist, Rusbult, Foster, & Agnew, 1999). Inter- or more intimate. It is important to note that partner responsiveness
dependence theory in general and the focus on prorelationship may occur as a result of somewhat objective behavioral responses
behaviors in particular do not acknowledge the possibility that not (e.g., social responses of one partner to another that communicate
all transformation of motivation or reasons for enacting prorela- that the partner understands and values the other’s needs, often
tionship behaviors are equal. SDT contends that one’s reasons for assessed through observational research; Gottman, 1979, 1994)
being in the relationship and one’s reasons for engaging in prore- and more subjective perceptions of a partner’s attentiveness to
lationship behaviors have important implications for how benefi- one’s needs. This concept of partner responsiveness is evident in
cial these behaviors may be, both for the relationship as a whole as each of the perspectives outlined above, and the notion of partner
well as for the individuals who compose it. It may be particularly responsiveness is particularly relevant to need fulfillment in close
beneficial to the relationship and partner when people engage in relationships and the current set of findings. Consistent with an
these behaviors because they truly want to and not simply to avoid extensive body of research— both within the close relationships
literature in general (see Stroebe & Stroebe, 1996, for review) and uniquely to relationship functioning and well-being independent of
within the SDT literature in particular (see Deci & Ryan, 2000, for the role of perceived closeness in these outcomes.
review)—the current studies point to the importance of perceiving The present studies used a variety of methods and data analytic
support from one’s social network for greater well-being. The approaches to examine how need fulfillment is associated with
current research focused specifically on the importance of having relationship functioning and well-being. These studies are, of
one’s psychological needs met in the context of one’s romantic course, not without limitations. First, need fulfillment was not
relationship and found that the extent to which one perceived that assessed longitudinally in any of the studies. Previous research has
one’s needs were being fulfilled by one’s partner determined a suggested that daily need fulfillment is associated with daily
feeling of greater (or lesser) relationship functioning and well- fluctuations in well-being (Reis et al., 2000; Sheldon et al., 1996).
being. Study 2 spoke to this issue more directly, as it focused on Although this previous research has focused on individual indica-
perceived need fulfillment among both partners and on how each
tors of well-being, it may very well be that daily fluctuations in
partner’s need fulfillment contributes to one’s own relationship
need fulfillment are also associated with daily fluctuations in
functioning and well-being as well as on how the combination of
relationship well-being such that individuals experience better
partner’s need fulfillment contributes to substantially greater rela-
“relationship days” on days when they experience greater need
tionship functioning and well-being. One of the unique contribu-
tions of Study 2 is its focus on both partners’ perceptions of how fulfillment in their relationship. Future research is needed to ex-
responsive the other was to fulfilling one’s psychological needs. amine this question. Another limitation is that Studies 1 and 3 were
Study 2 demonstrated that, although perceiving that one’s partner composed primarily of women. Study 2 included heterosexual
was meeting one’s own needs predicted important outcomes for couples and thus equal numbers of men and women, which alle-
oneself, one’s partner’s perceptions that those needs were being viates this concern somewhat, but additional research is needed to
met also uniquely predicted these outcomes for oneself. This is one enhance the generalizability of the findings.4 Another potential
of the key characteristics of close relationships and is perhaps one limitation to these studies is that all data were collected on college
of the more interesting indicators of partner connectedness: That students, who may not be representative of other adult relation-
one’s partner’s perceptions can impact how one feels about one’s ships. However, at least in terms of relationship length, the par-
relationship. In addition, Study 2 demonstrated that relationships ticipants in these studies were in relatively stable, long-term rela-
benefit even more from both partners feeling that the other is tionships for college student samples. Nonetheless, it is important
responsive to his or her needs, particularly the need for relatedness. for future research to examine these processes in nonstudent sam-
Although most relationship research focuses on individual percep- ples. In addition, all measures used in these studies were based on
tions, the current Study 2 emphasized the importance of perceived self-report, so there is no indication of how fulfillment of these
need fulfillment from both members of the dyad, thus contributing needs translates into behavior within one’s relationship, particu-
to our broader understanding of how both partner’s perceptions of larly in terms of responding to conflict. Finally, all of the studies
the other’s responsiveness contribute to relationship functioning were correlational, thus precluding causal inferences about the role
and well-being. In addition, Study 3 further expanded our under- of need fulfillment in these outcomes.
standing of why partner responsiveness to one’s needs is so ben-
Despite these limitations, the present set of studies adds to our
eficial by highlighting one mechanism through which such respon-
understanding of the importance of need fulfillment not only as it
siveness benefits relationships: Need fulfillment increases intrinsic
pertains to optimal individual functioning but also as it pertains to
relationship motivation.
optimal relationship functioning. Because need fulfillment is
In the current set of studies, it is also important to keep the
findings regarding relatedness in perspective. Study 1 illustrated largely a function of the social context, studying need fulfillment
the role of each of the needs in individual well-being but also within specific relationships and studying how such need fulfill-
demonstrated that relatedness was the strongest unique predictor of ment is associated with relationship experiences may help us to
relationship functioning and well-being. In addition, results for develop a clearer conceptualization of the ways in which needs are
overall need fulfillment were not simply a function of relatedness
needs. Across all three studies, secondary analyses revealed that 4
Across the eight samples used for Study 1 analyses, there were nearly
results for overall need fulfillment and those for a composite 400 men. We thus repeated analyses for this study using only the men in
consisting of only autonomy and competence needs were the same. the samples. In the first set of analyses involving the meta-analytically
Thus, these results were not simply driven by the role of related- combined correlations between need fulfillment and indicators of individ-
ness needs in relationship outcomes. In Studies 2 and 3, we were ual and relationship well-being, results were the same as those reported for
able to test the unique role of relatedness-need fulfillment relative the full sample. In the second set of analyses involving the meta-
to that of perceived closeness in the relationship. Because related- analytically combined prs between need fulfillment and these outcomes,
ness needs involve feeling connected to and understood by others, results were largely the same for relatedness-need fulfillment (with the
there is obvious conceptual overlap with relationship closeness. In exception of a nonsignificant pr for negative affect). However, results for
Study 2, we found that, when controlling for actor and partner autonomy and competence needs were somewhat different. Notably, for
men, there were no significant associations between autonomy and nega-
perceived relationship closeness, all actor and partner main effects
tive affect, avoidant attachment, anxious attachment, satisfaction, commit-
remained significant. More important, the interaction between ment, and defensive responses to conflict. Regarding competence, there
actor and partner perceived closeness did not account for the were no significant associations for positive affect, satisfaction, perceived
significant Actor ⫻ Partner interactions for relatedness-need ful- conflict, and understanding responses to conflict. However, there was a
fillment, which predicted satisfaction, perceived conflict, and de- significant pr between competence and commitment that did not emerge
fensiveness. Thus, relatedness-need fulfillment contributes for the full sample.
fulfilled and of how these needs contribute to optimal human Integration in personality. In R. Diestbeir (Ed.), Nebraska symposium on
functioning in a variety of contexts. motivation: Vol. 38. Perspectives on motivation (pp. 237–288). Lincoln:
University of Nebraska Press.
Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits:
References Human needs and the self-determination of behavior. Psychological
Inquiry, 11, 227–268.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and Deci, E. L., Ryan, R. M., Gagne, M., Leone, D. R., Usunov, J., &
interpreting interactions. Thousand Oaks, CA: Sage Publications Kornazheva, B. P. (2001). Need satisfaction, motivation, and well-being
Aron, A., & Aron, E. N. (1996). Self and self-expansion in relationships. in the work organizations of a former Eastern bloc country: A cross-
In G. O. Fletcher and J. Fitness (Eds.), Knowledge structures in close cultural study of self-determination. Personality and Social Psychology
relationships: A social psychological approach, (pp. 325–344). Hills- Bulletin, 27, 930 –942.
dale, NJ: Erlbaum. Drigotas, S. M., & Rusbult, C. E. (1992). Should I stay or should I go? An
Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of Other in the Self interdependence model of break-ups. Journal of Personality and Social
Scale and the structure of interpersonal closeness. Journal of Personality Psychology, 62, 62– 87.
and Social Psychology, 63, 596 – 612. Gottman, J. M. (1979). Marital interaction: Experimental investigations.
Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships New York: Academic Press.
as including other in the self. Journal of Personality and Social Psy- Gottman, J. M. (1994). What predicts divorce? The relationship between
chology, 60, 241–253. marital processes and marital outcomes. Hillsdale, NJ: Erlbaum.
Baard, P. P., Deci, E. L., & Ryan, R. M. (2000). Intrinsic need satisfaction Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis.
as a motivational basis of performance and well-being at work: An
Orlando, FL: Academic Press.
application of cognitive evaluation theory. Unpublished manuscript,
Hodgins, H. S., Koestner, R., & Duncan, N. (1996). On the compatibility
Fordham University, New York.
of autonomy and relatedness. Personality and Social Psychology Bulle-
Bakan, D. (1966). The duality of human existence. Chicago: Rand Mc-
tin, 22, 227–237.
Nally.
Hull, C. L. (1943). Principles of behavior: An introduction to behavior
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral
theory. New York: Appleton-Century-Crofts.
change. Psychological Review, 84, 191–215.
Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis: Cor-
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
recting error and bias in research findings. Newbury Park, CA: Sage.
distinction in social psychological research: Conceptual, strategic, and
Hunter, J. E., & Schmidt, F. L. (1994). Estimation of sampling error
statistical considerations. Journal of Personality and Social Psychology,
variance in the meta-analysis of correlations: Use of average correlation
51, 1173–1182.
in the homogeneous case. Journal of Applied Psychology, 79, 171–177.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for
Ilardi, B. C., Leone, D., Kasser, T., & Ryan, R. M. (1993). Employee and
interpersonal attachments as a fundamental human motivation. Psycho-
supervisor ratings of motivation: Main effects and discrepancies asso-
logical Bulletin, 117, 497–529.
ciated with job satisfaction and adjustment in a factory setting. Journal
Blais, M. R., Sabourin, S., Boucher, C., & Vallerand, R. (1990). Toward a
of Applied Social Psychology, 23, 1789 –1805.
motivational model of couple happiness. Journal of Personality and
Social Psychology, 59, 1021–1031. Kashy, D. A., & Kenny, D. A. (2000). The analysis of data from dyads and
Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and groups. In H. T. Reis & C. M. Judd (Eds.), Handbook of research
anger. (2nd ed.). New York: Basic Books. methods in social and personality psychology (pp. 451– 477). Cam-
Bowlby, J. (1980). Attachment and loss: Vol. 3: Loss. New York: Basic bridge, England: Cambridge University Press.
Books. Kasser, V., & Ryan, R. M. (1999). The relation of psychological needs for
Brennan, K., Clark, C., & Shaver, P. (1998). Self-report measurement of autonomy and relatedness to vitality, well-being, and mortality in a
adult attachment: An integrative overview. In J. A. Simpson & W. S. nursing home. Journal of Applied Social Psychology, 29, 935–954.
Rholes (Eds.) Attachment theory and close relationships, (pp. 46 –76). Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: A theory
New York: Guilford Press. of interdependence. New York: Wiley.
Campbell, L., & Kashy, D. A. (2002). Estimating actor, partner, and Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social
interaction effects for dyadic data using PROC MIXED and HLM: A psychology. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), Handbook of
user-friendly guide. Personal Relationships, 9, 327–342. social psychology (Vol. 1, 4th ed., pp. 233–265). Boston: McGraw-Hill.
Carver, C. S., & Scheier, M. F. (1990). Principles of self-regulation: Action Kernis, M. H. (2000). Substitute needs and the distinction between fragile
and emotion. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of and secure high self-esteem. Psychological Inquiry, 11, 298 –300.
motivation and cognition: Foundations of social behavior (Vol. 2, pp. Knee, C. R., Lonsbary, C., Canevello, A., & Patrick, H. (2005). Self-
3–52). New York: Guilford Press. determination and conflict in romantic relationships. Journal of Person-
Carver, C. S., & Scheier, M. F. (2000). Autonomy and self-regulation. ality and Social Psychology, 89, 997–1009.
Psychological Inquiry, 284 –291. Knee, C. R., Patrick, H., Vietor, N. A., & Neighbors, C. (2004). Implicit
Collins, N. L., & Miller, L. C. (1994). Self-disclosure and liking: A theories of relationships: Moderators of the link between conflict and
meta-analytic review. Psychological Bulletin, 116, 457– 475. commitment. Personality and Social Psychology Bulletin, 30, 617– 628.
Crittenden, P. M. (1990). Internal representational models of attachment Knee, C. R., Patrick, H., Vietor, N. A., Neighbors, C. T., & Nanayakkara,
relationships. Infant Mental Health Journal, 11, 259 –277. A. (2002). Self-determination as growth motivation in romantic relation-
de Charms, R. (1968). Personal causation. New York: Academic. ships. Personality and Social Psychology Bulletin, 28, 609 – 619.
Deci, E. L., La Guardia, J. G., Moller, A. C., Scheiner, M. J., & Ryan, Kobak, R. (1994). Adult attachment: A personality or relationship con-
R. M. (2006). On the benefits of giving as well as receiving autonomy struct? Psychological Inquiry, 5, 42– 44.
support: Mutuality in close friendships. Personality and Social Psychol- La Guardia, J., Ryan, R. M., Couchman, C., & Deci, E. L. (2000).
ogy Bulletin, 32, 313–327. Within-person variation in security of attachment: A self-determination
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self- theory perspective on attachment, need fulfillment, and well-being.
determination in human behavior. New York: Plenum. Journal of Personality and Social Psychology, 79, 367–384.
Deci, E. L., & Ryan, R. M. (1991). A motivational approach to self: Le, B., & Agnew, C. R. (2001). Need fulfillment and emotional experience
in interdependent romantic relationships. Journal of Social and Personal children’s perceptions. Journal of Personality and Social Psychology,
Relationships, 18, 423– 440. 50, 550 –558.
Meng, X.-L., Rosenthal, R., & Rubin, D. B. (1992). Comparing correlated Ryan, R. M., & La Guardia, J. G. (2000). What is being optimized over
correlation coefficients. Psychological Bulletin, 111, 172–175. development? A self-determination theory perspective on basic psycho-
Mills, J., & Clark, M. S. (2001). Viewing close romantic relationships as logical needs across the life span. In S. Qualls & R. Abeles (Eds.),
communal relationships: Implications for maintenance and enhance- Dialogues on psychology and aging (pp. 145–172). Washington, DC:
ment. In J. Harvey & A. Henzel (Eds.), Close romantic relationships: American Psychological Association.
Maintenance and enhancement (pp. 13–25). Mahwah, NJ: Erlbaum. Ryan, R. M., La Guardia, J. G., Solky-Butzel, J., Chirkov, V., & Kim, Y.
Murray, H. A. (1938). Explorations in personality. New York: Oxford (2005). On the interpersonal regulation of emotions: Emotional reliance
University Press. across gender, relationships, and cultures. Personal Relationships, 12,
Nezlek, J. B. (2001). Multilevel random coefficient analyses of event- and 145–163.
interval-contingent data in social and personality psychology research. Ryan, R. M., & Lynch, J. H. (1989). Emotional autonomy versus detach-
Personality and Social Psychology Bulletin, 27, 771–785. ment: Revisiting the vicissitudes of adolescence and young adulthood.
Norton, R. (1983). Measuring marital quality: A critical look at the depen- Child Development, 60, 340 –356.
dent variable. Journal of Marriage and the Family, 45, 141–151. Ryan, R. M., Stiller, J., & Lynch, J. H. (1994). Representations of rela-
Prager, K. J., & Buhrmester, D. (1998). Intimacy and need fulfillment in tionships to teachers, parents, and friends as predictors of academic
couple relationships. Journal of Social and Personal Relationships, 15, motivation and self-esteem. Journal of Early Adolescence, 14, 226 –249.
435– 469. Shaver, P. R., Collins, N. L., & Clark, C. L. (1996). Attachment styles and
Reis, H. T., Clark, M. S., & Holmes, J. G. (2004). Perceived partner internal working models of self and relationship partners. In G. J. O.
responsiveness as an organizing construct in the study of intimacy and Fletcher & J. Fitness (Eds.), Knowledge structures in close relation-
closeness. In D. J. Mashek & A. Aron (Eds.), The handbook of closeness
ships: A social psychological approach (pp. 25– 61). Hillsdale, NJ:
and intimacy (pp. 201–228). Mahwah, NJ: Erlbaum.
Erlbaum.
Reis, H. T., & Patrick, B. P. (1996). Attachment and intimacy: Component
Sheldon, K. M., Ryan, R. M., & Reis, H. T. (1996). What makes for a good
processes. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychol-
day? Competence and autonomy in the day and in the person. Person-
ogy: Handbook of basic principles (pp. 523–563). New York: Guilford.
ality and Social Psychology Bulletin, 22, 1270 –1279.
Reis, H. T., & Shaver, P. (1988). Intimacy as an interpersonal process. In
Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for
S. Duck, D. F. Hay, S. E. Hobfoll, & W. Ickes (Eds.) Handbook of
assessing the quality of marriage and similar dyads. Journal of Marriage
personal relationships: Theory, research, and interventions (pp. 367–
and the Family, 38, 15–28.
389). Oxford, United Kingdom: Wiley.
Stroebe, W., & Stroebe, M. (1996). The social psychology of social
Reis, H. T., Sheldon, K. M., Gable, S. L., Roscoe, J., & Ryan, R. M.
support. In A. Kruglanski & E. T. Higgins (Eds.), Social psychology:
(2000). Daily well-being: The role of autonomy, competence, and relat-
The handbook of basic principles (pp. 597– 621).
edness. Personality and Social Psychology Bulletin, 26, 419 – 435.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, Swann, W. B., De La Ronde, C., & Hixon, J. G. (1994). Authenticity and
NJ: Princeton University Press. positivity strivings in marriage and courtship. Journal of Personality and
Rosenthal, R. (1994). Statistically describing and combining studies. In H. Social Psychology, 66, 857– 869.
Cooper & L. V. Hedges (Eds.), The handbook of research synthesis (pp. Vallerand, R. J., & O’Connor, B. P. (1989). Motivation in the elderly: A
231–244). New York: Russell Sage Foundation. theoretical framework and some promising findings. Canadian Psychol-
Rosenthal, R., & Rosnow, R. L. (1991). Essentials of behavioral research: ogy, 30, 538 –550.
Methods and data analysis. New York: McGraw-Hill. Van Lange, P. A., Rusbult, C. E., Drigotas, S. M., Arriaga, X. A., Witcher,
Rusbult, C. E., Verette, J., Whitney, G. A., Slovik, L. F., & Lipkus, I. B. S., & Cox, C. L. (1997). Willingness to sacrifice in close relation-
(1991). Accommodation processes in close relationships: Theory and ships. Journal of Personality and Social Psychology, 72, 1373–1395.
preliminary empirical evidence. Journal of Personality and Social Psy- Watson, D. (1988). Intraindividual and interindividual analyses of positive
chology, 60, 53–78. and negative affect: Their relation to health complaints, perceived stress,
Ryan, R. M. (1995). Psychological needs and the facilitation of integrative and daily activities. Journal of Personality and Social Psychology, 54,
processes. Journal of Personality, 63, 397– 427. 1020 –1030.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
facilitation of intrinsic motivation, social development, and well-being. dation of brief measures of positive and negative affect: The PANAS
American Psychologist, 55, 68 –78. scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Ryan, R. M., & Frederick, C. M. (1997). On energy, personality and health: Wieselquist, J., Rusbult, C. E., Foster, C. A., & Agnew, C. R. (1999).
Subjective vitality as a dynamic reflection of well-being. Journal of Commitment, pro-relationship behavior, and trust in close relationships.
Personality, 65, 529 –565. Journal of Personality and Social Psychology, 77, 942–966.
Ryan, R. M., & Grolnick, W. S. (1986). Origins and pawns in the class- White, R. W. (1959). Motivation reconsidered: The concept of compe-
room: Self-report and projective assessments of individual differences in tence. Psychological Review, 66, 297–333.
Appendix
Study 1: Range of Partial Correlations and Standard Error of the Means

Autonomy Competence Relatedness
Characteristic k N pr SE pr SE pr SE
Self-esteem 8 1,918 ⫺.02 to .20 .03 .13 to .50 .03 ⫺.08 to .10 .03
Positive affect 4 955 ⫺.04 to .03 .04 .15 to .27 .04 .10 to .39 .04
Negative affect 4 955 ⫺.30 to .10 .04 ⫺.03 to ⫺.40 .04 ⫺.05 to ⫺.24 .04
Vitality 2 484 ⫺.02 to .01 .06 .14 to .20 .06 .14 to .23 .05
Attachment
Avoidant 4 1,345 ⫺.06 to ⫺.19 .03 ⫺.03 to .01 .03 ⫺.25 to ⫺.49 .03
Anxious 4 1,345 ⫺.01 to ⫺.26 .03 ⫺.05 to ⫺.21 .03 ⫺.16 to ⫺.25 .03
Satisfaction 8 1,918 ⫺.03 to .16 .03 ⫺.01 to .19 .03 .37 to .70 .03
Commitment 7 1,561 ⫺.17 to .11 .03 ⫺.17 to .29 .03 .13 to .86 .03
Perceived conflict 5 1,354 ⫺.03 to ⫺.33 .04 ⫺.16 to .00 .03 ⫺.24 to ⫺.38 .03
Understanding 5 1,467 .01 to .09 .03 ⫺.10 to .20 .03 .08 to .33 .03
Defensive 5 1,467 ⫺.30 to .02 .04 ⫺.03 to ⫺.18 .04 ⫺.06 to ⫺.26 .03
Note. k ⫽ number of samples; N ⫽ total number of observations; pr ⫽ partial correlation coefficient.
Received April 4, 2006

Revision received August 14, 2006
Accepted August 20, 2006 䡲

2007 PatrickKneeCanevelloLonsbary PSPB

Uploaded by

Copyright:

Available Formats

2007 PatrickKneeCanevelloLonsbary PSPB

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2007 PatrickKneeCanevelloLonsbary PSPB

Uploaded by

Copyright:

Available Formats

Journal of Personality and Social Psychology Copyright 2007 by the American Psychological Association

2007, Vol. 92, No. 3, 434 – 457 0022-3514/07/$12.00 DOI: 10.1037/0022-3514.92.3.434

The Role of Need Fulfillment in Relationship Functioning and Well-Being:

Heather Patrick C. Raymond Knee

Amy Canevello Cynthia Lonsbary

Keywords: psychological needs, self-determination theory, relationships

PSYCHOLOGICAL NEEDS: THE SDT PERSPECTIVE

Mean relationship Mean number of

1 204 82 23.1 (4.9) 2.6 (2.5) 1.9 (1.0)

Characteristic k N r 95% CI CI z Range of r SE of mean

Effect Size: pr testing the unique contribution of autonomy, competence, and

Autonomy Competence Relatedness

Characteristic k N pr 95% CI CI z pr 95% CI CI z pr 95% CI CI z

Method Analytic Strategy

Overall need fulfillment

IOS interactions with their partner. Items were answered on a 7-point

Note. Ns ranged from 119 to 120 depending on completeness of data.

Step 1: Need fulfillment predicting postdisagreement

Autonomy 0.12 0.16 .09 .07 0.28 0.22 .16 .12

Step 2: Need fulfillment predicting relationship motivation

Autonomy 1.06 1.12 .09 .09 1.06 1.12 .09 .09

Steps 3 and 4: Mediation

Autonomy 0.08 0.15 .06 .05 0.20 0.19 .11 .10

Study 1: Range of Partial Correlations and Standard Error of the Means

Note. k ⫽ number of samples; N ⫽ total number of observations; pr ⫽ partial correlation coefficient.

Received April 4, 2006

You might also like