P SY CH OL OG I C AL S CIE N CE
Research Article
Accuracy and Biases in
Newlyweds’ Perceptions of
Each Other
Not Mutually Exclusive but Mutually Beneficial
Shanhong Luo1 and Anthony G. Snider2
1
Department of Psychology and 2Department of Environmental Studies, University of North Carolina at Wilmington
ABSTRACT—There
has been a long-standing debate about
whether having accurate self-perceptions or holding positive illusions of self is more adaptive. This debate has recently expanded to consider the role of accuracy and bias
of partner perceptions in romantic relationships. In the
present study, we hypothesized that because accuracy,
positivity bias, and similarity bias are likely to serve
distinct functions in relationships, they should all make
independent contributions to the prediction of marital
satisfaction. In a sample of 288 newlywed couples, we
tested this hypothesis by simultaneously modeling the actor
effects and partner effects of accuracy, positivity bias,
and similarity bias in predicting husbands’ and wives’
satisfaction. Findings across several perceptual domains
suggest that all three perceptual indices independently
predicted the perceiver’s satisfaction. Accuracy and similarity bias, but not positivity bias, made unique contributions to the target’s satisfaction. No sex differences were
found.
There has been a long-standing debate about whether having
accurate self-perceptions or holding positive illusions of self is
more adaptive (e.g., Colvin, Block, & Funder, 1995; Taylor &
Brown, 1988). Accuracy and bias are considered mutually exclusive perceptual properties, such that a person can be either
accurate or biased, but cannot be both. Additionally, it has been
assumed that only one property can have adaptive value: If
accuracy is beneficial, then bias is not, and vice versa. In the
Address correspondence to Shanhong Luo, Department of Psychology, Social Behavioral Science Building, University of North Carolina at Wilmington, 109 C, SBS, Wilmington, NC 28403, e-mail:
[email protected].
1332
past two decades, this debate has expanded to consider the role
of accuracy and bias of partner perceptions (one’s perception of
one’s partner) in romantic relationships, with a focus on perceptions of personality (for reviews, see Fletcher, Simpson, &
Boyes, 2006; Gagne & Lydon, 2004). Some researchers argue
that maintaining accurate perceptions of the partner is critical to
both the perceiver’s (Kobak & Hazan, 1991) and the target’s
(Swann, De La Ronde, & Hixon, 1994) satisfaction, whereas
others reason that engaging in a leap of faith regarding a partner
by seeing the partner in a positive light is important for relationship functioning (e.g., Murray, Griffin, & Holmes, 1996;
Murray, Holmes, Bellavia, Griffin, & Dolderman, 2002).
Recent theorizing regarding the role of accuracy and bias in
partner perceptions includes a fundamental shift from the simplified either-or approach to a more integrative, dialectical approach. Specifically, it has been suggested that accuracy and
bias in partner perceptions are not necessarily mutually exclusive, but can coexist (e.g., Kenny & Acitelli, 2001; Murray
et al., 1996). Moreover, accuracy and bias can both be beneficial, although in different ways, because they are subject to
different motives, goals, and situations. For example, Gagne and
Lydon (2004) argued that accuracy is more relevant in information-driven relationship judgments, whereas positivity bias is
more important in esteem-related judgments. Fletcher et al.
(2006) took an evolutionary approach to delineate under what
conditions (e.g., short- vs. long-term relationships) and for what
attributes (e.g., attractiveness vs. status) people are motivated to
be accurate or positively biased. Finally, Neff and Karney (2005)
proposed that partners involved in happier relationships tend to
hold more positive biases on global characteristics, but to perceive specific attributes accurately.
According to these proposals, accuracy and bias each benefit
relationships by playing a leading role in different kinds of
Copyright r 2009 Association for Psychological Science
Volume 20—Number 11
Shanhong Luo and Anthony G. Snider
perceptions (i.e., perceiving different attributes or making
different judgments). However, previous research has shown
that accuracy and biases coexist in the same set of perceptions of
partner personality traits (e.g., Kenny & Acitelli, 2001; Murray
et al., 1996). Taking advantage of new methodologies, the current research disentangled accuracy and biases in the same set
of perceptions and directly evaluated their relative contributions in the prediction of a satisfying relationship. Because accuracy and biases are likely to be driven by different motives
and to serve different functions in relationships, we hypothesized that they would make independent contributions to marital
satisfaction.
The current research was also designed to extend previous
work by overcoming two other limitations. First, previous discussions and examinations of bias effects on relationship outcomes have primarily focused on positivity bias and largely
ignored the role of similarity bias—the belief that a partner is
more similar to oneself than is true. Second, most research in
this area has focused on actor effects and has not rigorously
considered partner effects. In other words, satisfaction is typically predicted by the perceiver’s own perceptions (actor effects)
rather than his or her spouse’s perceptions (partner effects). In
the current research, we proposed a general model to simultaneously model the actor effects and partner effects of all three
perceptual processes (i.e., accuracy, positivity bias, and similarity bias) in predicting husbands’ and wives’ satisfaction. We
tested this general model in several personality domains in a
sample of 288 newlywed couples.
ACCURACY, POSITIVITY BIAS, AND SIMILARITY BIAS
IN PARTNER PERCEPTIONS
Accuracy
There are several reasons why obtaining accurate perceptions of
each other is likely to be beneficial to both parties in a relationship. An accurate assessment of partner attributes (a) enables perceivers to correctly evaluate their partners’ needs and
anticipate their behaviors, thus fostering a sense of control,
predictability, and security on the part of the perceivers (e.g.,
Kenny & Acitelli, 2001; Swann et al., 1994); (b) helps partners
coordinate activities and reconcile conflicting goals, thereby
leading to more harmonious interactions (e.g., Kobak & Hazan,
1991; Neff & Karney, 2005); and (c) is important to targets because accurate perceptions provide a feeling of being validated,
which is a crucial requirement for intimacy (Reis & Shaver,
1988). Indeed, previous research has found consistent evidence
that individuals’ ratings of their partners show substantial
agreement with their partners’ self-ratings (e.g., Kenny & Acitelli, 2001; Watson, Hubbard, & Wiese, 2000b). Moreover, accuracy in perceptions, in general, is associated with positive
relationship outcomes (Kobak & Hazan, 1991; Murray et al.,
1996; Neff & Karney, 2005; for an exception, see Murray et al.,
2002).
Volume 20—Number 11
Positivity Bias
The perception of one’s partner in an overly positive light has
been documented as one of the most pervasive biases in romantic relationships (e.g., Murray et al., 1996; Rusbult, Van
Lange, Wildschut, Yovetich, & Verette, 2000). These positive
illusions may be driven by the fundamental need to feel good
about the self, which, by extension, includes one’s immediate
network, such as romantic partners (see Taylor & Brown, 1988).
Moderately positively skewed partner perceptions are thought to
be adaptive because they enable perceivers to justify the belief
that their partner is the ‘‘right’’ one and to counteract the attractiveness of potential alternative partners (e.g., Murray et al.,
1996). A positivity bias is also likely to make targets feel valued
and trusted, particularly when their partners give them the
benefit of the doubt in stressful or uncertain situations (e.g.,
Brickman, 1987; Rusbult et al., 2000). Previous research has
provided robust support that individuals who perceive their
partners and relationships positively tend to be more satisfied in
their relationships (e.g., Fowers, Lyons, & Montel, 1996; Murray
et al., 1996; Rusbult et al., 2000).
Similarity Bias
The tendency for romantic partners to perceive each other as
more similar than they actually are is another pervasive bias in
romantic relationships (e.g., Kenny & Acitelli, 2001; Watson,
Hubbard, & Wiese, 2000). Perceptions of exaggerated similarity
can benefit perceivers by fostering feelings of closeness and
intimacy (Aron, Aron, Tuder, & Nelson, 1991). A similarity bias
is likely to lead targets to feel understood (e.g., Murray et al.,
2002), and to help targets feel more confident of their partner’s
love (Byrne & Blaylock, 1963; Condon & Crano, 1988). Indeed,
recent research has shown that similarity bias concerning values, personality attributes, and day-to-day feelings is associated
with greater marital satisfaction (Murray et al., 2002).
CONSIDERING ACCURACY, POSITIVITY BIAS, AND
SIMILARITY BIAS SIMULTANEOUSLY
Given that each of the three perceptual processes is likely to
bring distinct benefits to perceivers and targets, we hypothesized that accurate perceptions, positivity bias, and similarity
bias would all make independent contributions to perceivers’
and targets’ satisfaction. Because we simultaneously modeled
the actor and partner effects of all three perceptual processes,
the current research is well poised to answer several questions
that previous research has been unable to address:
Do accuracy, positivity bias, and similarity bias have beneficial effects independently of each other, or are their effects
mutually exclusive?
How do accurate and biased perceptions impact perceivers’
and targets’ satisfaction?
Are there any sex differences in these effects?
1333
Accuracy, Bias, and Satisfaction
We followed Kenny and his colleagues’ actor-partner interdependence model (APIM; see Kenny, 1996; Kenny, Kashy, &
Cook, 2006) to design the general model. As demonstrated in
Figure 1, three actor effects predict husbands’ satisfaction—
husbands’ own accuracy (path a), similarity bias (path b), and
positivity bias (path c). Three partner effects also predict husbands’ satisfaction—their wives’ accuracy (path x), similarity
bias (path y), and positivity bias (path z). The corresponding
paths for wives’ satisfaction are marked as a 0 , b 0 , c 0 , x 0 , y 0, and z 0 .
The predictors were allowed to correlate with each other. We also
allowed the error terms (i.e., residuals) of husbands’ and wives’
satisfaction to be correlated with each other because there is
consistent evidence in the literature that one spouse’s satisfaction is strongly related to the other’s (e.g., Watson, Hubbard, &
Wiese, 2000a). Given that very few systematic, replicable
gender differences have been identified for relationship processes (see Karney & Bradbury, 1995), we did not hypothesize
any gender differences. We tested the general model in several
domains, including Big Five personality traits, attachment,
affectivity, and emotional expressivity, because each captures
different central aspects of personality. In addition, self and
partner perceptions in these domains have been shown to have
important implications for relationship functioning (e.g., Luo &
Klohnen, 2005; Watson et al., 2000).
SD 5 6.2). The couples had been married an average of 154 days
(range 5 25–452 days). The majority of the participants were
Caucasian (90.9%) and had a college education or more advanced degree (83%). Most participants (82%) reported this was
their first marriage, and 75% had no children. All participants
completed questionnaires in small-group sessions comprising a
maximum of 3 couples each. Participants rated themselves and
their spouses on all key measures. The couples were compensated $120 each for their participation.
Measures
Demographic Questionnaire
The demographic questionnaire included questions about gender, age, ethnicity, education, length of acquaintanceship with
the partner, and duration of the marriage.
Big Five Inventory
The Big Five Inventory contains five subscales: Neuroticism (8
items), Extraversion (8 items), Agreeableness (9 items), Conscientiousness (9 items), and Openness (10 items; John & Srivastava, 1999). Participants responded to each item using a 5point scale ranging from disagree strongly to strongly agree.
Alpha reliabilities in this sample ranged from .79 to .88 for the
subscales.
METHOD
Participants and Procedure
The sample consisted of 288 newlywed couples recruited
through mail in two Midwestern cities (mean age 5 27.8 years,
Husband
Accuracy
a
x′
Husband
Similarity Bias
b
y′
Husband
Positivity Bias
c
z′
Wife
Accuracy
Wife
Similarity Bias
Wife
Positivity Bias
The Positive and Negative Affect Schedule (PANAS)
Participants completed the trait form of the PANAS (Watson,
Clark, & Tellegen, 1988). The PANAS has two 10-item subscales
assessing positive affect (e.g., enthusiastic, active, interested) and
Errorh
Husband
Satisfaction
Errorw
x
a′
y
Wife
Satisfaction
b′
z
c′
Fig. 1. The general model predicting husbands’ and wives’ marital satisfaction from accuracy,
similarity bias, and positivity bias.
1334
Volume 20—Number 11
Shanhong Luo and Anthony G. Snider
negative affect (e.g., nervous, upset, irritable), respectively. For
each item, participants were asked to indicate ‘‘to what extent
you generally feel this way’’ and ‘‘to what extent your spouse
generally feels this way,’’ using a 5-point scale ranging from very
slightly or not at all to extremely. Alpha reliabilities in this
sample were .86 and .89 for the two subscales.
Emotional Expressivity
Participants used a 5-point scale (ranging from not at all to very
strongly) to indicate the extent to which they and their spouse
typically expressed 15 emotions (Gross & John, 1998), including
6 positive emotions (e.g., joy, love, excitement) and 9 negative
emotions (e.g., anger, fear, sadness, shame). Alpha reliabilities in
this sample were .83 and .89 for the two subscales.
Adult Attachment
Participants completed a 16-item short version of Brennan,
Clark, and Shaver’s (1998) 36-item attachment measure, which
yields scores on the dimensions of anxiety and avoidance. Participants used a 7-point scale (ranging from strongly disagree to
strongly agree) to indicate how they and their spouse typically
felt and acted in their relationship. Alpha reliabilities in this
sample were .75 and .85 for the two subscales.
Marital Satisfaction
We used three measures that assessed different aspects of relationship satisfaction. Participants completed the Locke-Wallace Marital Adjustment Test (Locke & Wallace, 1959), which is
a 15-item self-report measure of marital satisfaction and couple
agreement on a number of issues (e.g., finances, recreation,
affection, friends, sex, and conflict resolution). The rating scales
varied across items. Sexual satisfaction was assessed using 10
items from the Pinney Sexual Satisfaction Inventory (Pinney,
Gerrard, & Denney, 1987). Participants responded using a 5point scale ranging from strongly disagree to strongly agree.
Finally, participants completed the 3-item Conflict subscale
from the Relationship Assessment Questionnaire (Simms &
Watson, 2006), which measures the frequency of conflict in the
relationship (e.g., ‘‘How often do you and your spouse quarrel?’’). Items were rated on a 5-point scale ranging from never to
once or more a day. A composite index of relationship satisfaction was created by averaging the three standardized measures.
Alpha reliability for the composite satisfaction score in this
sample was .70 for both husbands and wives.
Computing Accuracy and Bias Indices
Accuracy
We followed the procedure used by Murray et al. (2002) to
compute the person-centered indices. We use a fictional couple—Jason and Mary—to illustrate the computational procedure. To obtain the accuracy index for Jason, we computed the
profile correlation between Jason’s perceptions of Mary and
Volume 20—Number 11
Mary’s self-ratings across all items for each of the four domains
(Big Five, PANAS, emotional expressivity, and attachment).
Profile correlations can range from 1.0 to 1.0, and higher
correlations indicate greater accuracy.
Similarity Bias
To obtain an index of similarity bias for Jason, we computed a
regular partial correlation. Specifically, we correlated Jason’s
self-ratings with his perceptions of Mary and then partialed out
Mary’s self-ratings. This partial correlation captured the degree
to which Jason’s perception of Mary as similar to himself exceeded their actual similarity.
Positivity Bias
To index the portion of positivity of partner perceptions that went
beyond the actual positivity of the partner’s self-ratings, we first
needed to establish the positivity of each item on the measures
we used. To accomplish this, we followed Block’s (1961/1978)
expert-based prototype approach. The first author and seven
advanced research assistants judged the social desirability of
each item on a 6-point scale ranging from very socially undesirable to very socially desirable. Average interrater agreement
was .87. We averaged the ratings for each item across all eight
judges and used these averages as a positivity prototype. We
computed the correlation between Jason’s perceptions of Mary
and the positivity prototype, with Mary’s self-ratings being
partialed out. This index thus captured the extent to which Jason’s positivity in his perceptions of Mary went beyond the actual positivity of Mary’s self-ratings.
RESULTS
Preliminary Analyses
Before conducting our primary analyses, we examined husbands’ and wives’ mean scores on the three perceptual indices.
Table 1 shows the means and standard deviations for the three
indices for spouses in the four domains. Wives consistently had
higher scores in almost every domain for every perceptual index,
although not all gender differences reached statistical significance. These results indicate that wives tended to be both more
accurate and more biased in their perceptions of their spouses
than husbands were.
Model Fit for the Proposed Model
We used a structural equation modeling program in LISREL
(Joreskog & Sorbom, 1993) to test the proposed model (Fig. 1).
The proposed model simultaneously tested the unique contributions of the actor and partner effects of accuracy, similarity
bias, and positivity bias to marital satisfaction for both husbands
and wives. We allowed all six predictors to be correlated with
each other and the two error terms of satisfaction to be correlated. Because no substantial gender difference was expected,
equality constraints were imposed on all paths concerning
1335
Accuracy, Bias, and Satisfaction
TABLE 1
Mean Values of the Three Perceptual Indices by Domain
Accuracy
Similarity bias
Positivity bias
Domain
Husband
Wife
Husband
Wife
Big Five
Attachment
Affectivity
Emotional expressivity
.57 (.28)
.65 (.42)
.72 (.45)
.63 (.39)
.58 (.27)
.63 (.41)
.78 (.43)nn
.70 (.36)nn
.23 (.36)
.35 (.49)
.48 (.42)
.49 (.49)
.30 (.34)
.43 (.51)
.57 (.43)nn
.56 (.46)
nn
Husband
Wife
.32 (.41)
.25 (.40)
.48 (.44)
.42 (.49)
.41 (.39)nn
.32 (.42)
.56 (.44)
.55 (.48)nn
Note. Standard deviations are given in parentheses. Asterisks indicate a significant difference between husbands and wives
(p < .01, two-tailed). N 5 288.
gender (i.e., a 5 a 0 , b 5 b 0 , c 5 c 0 , and so on). Lack of statistical
significance in the constrained model would suggest that these
constraints did not significantly worsen model fit—in other
words, that there was no significant difference between
the model with and without gender difference. If that was the
case, then the model without gender difference would be
preferable for its parsimoniousness.
To evaluate the goodness of fit of the constrained model, we
considered four fit indices: chi-square, the goodness-of-fit index
(GFI), the comparative fit index (CFI), and the point estimate and
90% confidence interval of the root-mean-square error of approximation (RMSEA). Among these four indices, chi-square
and GFI are more affected by sample size, whereas CFI and
RMSEA are less affected by sample size (Fan, Thompson, &
Wang, 1999). We tested this model separately in each of the four
domains. Table 2 reports the four fit indices for all four domains.
The chi-square values ranged from 1.43 to 7.49, with none being
statistically significant despite the large sample size (N 5 288),
suggesting that forcing the paths to be equal across gender did
not significantly worsen the model fit. All GFIs and CFIs were
equal to or above .99, indicating an excellent fit. The point estimates of RMSEAs ranged from .00 to .03, which meets the
criterion for a good model fit (RMSEA .06; Hu & Bentler,
1999). The lower limit of the 90% confidence interval of
RMSEAs was .00 for all domains, and the upper limit ranged
from .00 to .09. These intervals all fell within the range for a wellfitting model (Hu & Bentler, 1999). Taken together, these results
TABLE 2
Model-Fit Indices for Each Domain
RMSEA
Domain
w2(6)
GFI
CFI
Point
estimate
90% CI
Big Five
Attachment
Affectivity
Emotional expressivity
1.43
3.26
7.49
6.86
1.00
1.00
.99
.99
1.00
.99
.99
1.00
.00
.00
.03
.02
.00–.00
.00–.05
.00–.09
.00–.08
Note. N 5 288. GFI 5 goodness-of-fit index; CFI 5 comparative fit index;
RMSEA 5 root-mean-square error of approximation; 90% CI 5 90% confidence interval.
1336
provide consistent and strong evidence for an excellent model fit,
suggesting that there was no substantial gender difference.
Predicting Marital Satisfaction From Accuracy, Similarity
Bias, and Positivity Bias
Next, we examined whether the path coefficients were consistent
with our predictions. Table 3 provides a summary of standardized path coefficients for actor and partner effects. Husbands
and wives had the same path coefficients because no significant
gender difference was detected. For actor effects, all three
perceptual indices made independent contributions to the prediction of satisfaction; the only exception was that similarity
bias did not contribute to the prediction of affectivity. Positivity
bias consistently had the strongest path coefficients. This pattern was well replicated across the four domains. In terms of
partner effects, accuracy was the most robust predictor of satisfaction; it was consistently significant and had the strongest
path coefficients. Similarity bias made independent contributions to satisfaction in all domains except for affectivity. Positivity bias, however, reached statistical significance only in the
case of the Big Five. Accuracy and similarity bias seem to be
uniquely important to both the self’s and the spouse’s marital
satisfaction, whereas positivity bias is important to the self’s but
not to the partner’s marital satisfaction.
Finally, we examined the percentage of variance in satisfaction that could be accounted for by the predictors. Table 4 shows
squared multiple correlations between satisfaction and the set of
the six predictors (individual and spousal accuracy, similarity
bias, and positivity bias). These predictors accounted for 24% to
37% of the variance in husbands’ satisfaction, explaining an
average of 29% of the variance. For wives, the explained variance ranged from 22% to 31%, with an average of 27%. These
percentages are highly impressive considering the fact that we
only used partner perceptions in one personality domain to
predict overall marital satisfaction, which can be affected by
thousands of factors.
DISCUSSION
The current findings shed new light on the long-standing debate
concerning whether accurate or biased perceptions are more
Volume 20—Number 11
Shanhong Luo and Anthony G. Snider
TABLE 3
Standardized Path Coefficients Predicting Marital Satisfaction From Accuracy, Positivity Bias, and
Similarity Bias
Actor effects
Domain
Similarity bias
Positivity bias
Accuracy
Similarity bias
Positivity bias
.12nn
.19nn
.10nn
.16nn
.14
.10n
.19nn
.07
.14nn
.13
.44nn
.23nn
.32nn
.35nn
.34
.15nn
.13nn
.15nn
.13nn
.14
.12nn
.12nn
.08
.12nn
.11
.09n
.00
.01
.08
.04
Big Five
Attachment
Affectivity
Emotional expressivity
Average
Note. N 5 288.
n
p < .05 (two-tailed).
Partner effects
Accuracy
nn
p < .01 (two-tailed).
adaptive. For a long time, this basic question has been framed in
an either-or fashion; that is, either accurate or biased perceptions was considered to be adaptive (e.g., Colvin et al., 1995;
Taylor & Brown, 1988). Recently, relationship researchers have
started to reconcile these two seemingly contradictory views and
to accept the notion that both perspectives may be correct—that
both accurate and positively biased perceptions have adaptive
consequences (e.g., Gagne & Lydon, 2004). However, empirical
testing of this important theoretical development has been very
limited. The current study is the first to examine accuracy,
positivity bias, and similarity bias in the same study and to show
that they all make independent contributions to predicting
husbands’ and wives’ satisfaction, and particularly to predicting
perceivers’ own satisfaction. This suggests that individuals tend
to have a happier marriage if they perceive their spouse with a
combination of accuracy, positivity bias, and similarity bias than
if they are only accurate or only biased.
When the three perceptual indices were used simultaneously
to predict targets’ satisfaction, accuracy and similarity bias
continued to make independent contributions, whereas positivity bias failed to be a significant predictor. This finding suggests that individuals function better when the feedback they
receive from their spouse is consistent with their own self-perceptions, and when their spouses overestimate how much they
TABLE 4
Squared Multiple Correlations (R2) Between Satisfaction and the
Perceptual Indices in Each Domain
Domain
Big Five
Attachment
Affectivity
Emotional expressivity
Average
Note. N 5 288.
nn
p < .01 (two-tailed).
Volume 20—Number 11
Husband’s satisfaction
Wife’s satisfaction
.37
.28nn
.24nn
.26nn
.29
.31nn
.30nn
.23nn
.22nn
.27
nn
have in common. Positive spousal feedback that goes beyond
one’s self-ratings does not seem to make one feel better about
one’s marriage. It seems that, at least in this particular context
(i.e., newlyweds’ partner perceptions in the domains we investigated), our finding is more consistent with the verification
theory (e.g., Swann et al., 1994). It is important to note that
positivity bias is not destructive—we did not find any significant
negative link between positivity bias and target satisfaction—
and may not show its beneficial effects on the target until later in
the marriage.
These findings strongly suggest that accuracy and biases
should not be deemed as mutually exclusive. Instead, they are
likely to complement each other by serving uniquely adaptive
functions in relationships. Previous research has shown that
accuracy and positivity bias can have independent beneficial
effects only by working on different types of perceptions (see
Fletcher et al., 2006; Gagne & Lydon, 2004; Neff & Karney,
2005). The current findings constitute an important extension by
highlighting the fact that accurate perceptions and different
biases can coexist in the same set of perceptions, and that it is
possible to tease them apart. In other words, an individual’s
perception of his or her partner is a mixture of at least three
components: accuracy, positivity bias, and similarity bias.
(There are likely other biases and random errors.) In the current
research, these components all made independent contributions
to predicting marital satisfaction.
Although our study was not designed to test the unique
functions of each perceptual process, we use a hypothetical case
to illustrate how the three perceptions could work independently
and simultaneously to enhance relationship satisfaction. Jason
and Mary are a newlywed couple, and recently Jason lost his job
due to the global economic recession. It is likely that Mary’s
accurate perceptions of Jason would be beneficial to his job
search because such perceptions may help Jason reevaluate
himself and assess the situation more correctly. Additionally,
her positivity bias for Jason would enable her to continue or even
strengthen her faith in love despite the apparent temporary
1337
Accuracy, Bias, and Satisfaction
difficulties Jason is facing. Finally, believing that she and her
spouse are similar to each other should allow Mary to better
appreciate Jason’s frustrations, and this empathy would be extremely helpful to him. Although this example is fictitious, it
illustrates how accuracy, positivity bias, and similarity bias can
all play a distinct adaptive role in relationship maintenance.
An important next step would be to specify and test the
functions of each process and how these processes work in
concert to enhance perceivers’ and targets’ satisfaction. It is
crucial to examine the extent to which the findings generalize to
different stages of relationship development (other than the
newlywed stage), samples of different ages and regions, and
other dimensions (e.g., perceived abilities). It is highly possible
that in different specific circumstances, different perceptual
processes play a leading role in relationship functioning. For
example, when an individual is initially attracted to someone,
positivity bias may be the driving force underlying the individual’s relationship behaviors, but as a relationship continues
to develop and moves into the more stable and committed phase,
accuracy may become increasingly important. Finally, although
we hypothesized that accuracy and the two biases have unique
benefits, and that all three processes help enhance relationship
satisfaction, it is possible that marital satisfaction drives perception rather than the other way around. It will be very important for future research to give more attention to the
directionality of the causal link between perceptions and relationship outcomes.
Acknowledgments—The authors would like to thank David
Watson, Diane Berry, and Eva Klohnen for their collaboration on
the Iowa Marital Assessment Project (IMAP) and all of the IMAP
staff for their help in the data collection for this project. This research was supported by National Institute of Mental Health Grant
1-R01-MH61804-01 to Diane Berry and by National Institute of
Mental Health Grant 1-R03-MH068395-01 to Eva C. Klohnen.
REFERENCES
Aron, A., Aron, E.N., Tudor, M., & Nelson, G. (1991). Close relationships as including other in the self. Journal of Personality and
Social Psychology, 60, 241–253.
Block, J. (1978). The Q-sort method in personality assessment and
psychiatric research. Palo Alto, CA: Consulting Psychologists
Press. (Original work published 1961)
Brennan, K.A., Clark, C.L., & Shaver, P.R. (1998). Self-report measurement of adult attachment: An integrative overview. In J.A.
Simpson & W.S. Rholes (Eds.), Attachment theory and close relationships (pp. 47–76). New York: Guilford Press.
Brickman, P. (Ed.). (1987). Commitment, conflict, and caring. Englewood Cliffs, NJ: Prentice Hall.
Byrne, D., & Blaylock, B. (1963). Similarity and assumed similarity of
attitudes between husbands and wives. Journal of Abnormal and
Social Psychology, 67, 636–640.
1338
Colvin, C.R., Block, J., & Funder, D.C. (1995). Overly positive selfevaluations and personality: Negative implications for mental
health. Journal of Personality and Social Psychology, 68, 1152–
1162.
Condon, J.W., & Crano, W.D. (1988). Inferred evaluation and the relation between attitude similarity and interpersonal attraction.
Journal of Personality and Social Psychology, 54, 789–797.
Fan, X., Thompson, B., & Wang, L. (1999). Effects of sample size,
estimation method, and model specification on structural equation
modeling fit indexes. Structural Equation Modeling, 6, 56–83.
Fletcher, G.J.O., Simpson, J.A., & Boyes, A.D. (2006). Accuracy and
bias in romantic relationships: An evolutionary and social psychological analysis. In M. Schaller, J.A. Simpson, & D.T. Kenrick
(Eds.), Evolution and social psychology (pp. 189–210). New York:
Psychology Press.
Fowers, B.J., Lyons, E.M., & Montel, K.H. (1996). Positive marital
illusions: Self-enhancement or relationship enhancement? Journal of Family Psychology, 10, 192–208.
Gagne, F.M., & Lydon, J.E. (2004). Bias and accuracy in close relationships: An integrative review. Personality and Social Psychology Review, 8, 322–338.
Gross, J.J., & John, O.P. (1998). Mapping the domain of expressivity:
Multi-method evidence for a hierarchical model. Journal of
Personality and Social Psychology, 74, 170–191.
Hu, L., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new
alternatives. Structural Equation Modeling, 6, 1–55.
John, O.P., & Srivastava, S. (1999). The Big Five trait taxonomy:
History, measurement, and theoretical perspectives. In L.A.
Pervin & O.P. John (Eds.), Handbook of personality (2nd ed., pp.
102–138). New York: Guilford.
Joreskog, K.G., & Sorbom, D. (1993). LISREL 8 user’s reference guide.
Chicago: Scientific Software International.
Karney, B.R., & Bradbury, T.N. (1995). The longitudinal course of
marital quality and stability: A review of theory, methods, and
research. Psychological Bulletin, 118, 3–34.
Kenny, D.A. (1996). Models of nonindependence in dyadic research.
Journal of Social and Personal Relationships, 13, 279–294.
Kenny, D.A., & Acitelli, L.K. (2001). Accuracy and bias in the perception of the partner in a close relationship. Journal of Personality and Social Psychology, 80, 439–448.
Kenny, D.A., Kashy, D.A., & Cook, W.L. (2006). Dyadic data analysis.
New York: Guilford.
Kobak, R., & Hazan, C. (1991). Attachment in marriage: Effects of
security and accuracy of working models. Journal of Personality
and Social Psychology, 60, 861–869.
Locke, H., & Wallace, K. (1959). Short marital-adjustment and prediction tests: Their reliability and validity. Marriage and Family
Living, 21, 251–255.
Murray, S.L., Holmes, J.G., Bellavia, G., Griffin, D.W., & Dolderman,
D. (2002). Kindred spirits? The benefits of egocentrism in close
relationships. Journal of Personality and Social Psychology, 82,
563–581.
Murray, S.L., Holmes, J.G., & Griffin, D.W. (1996). The benefits of
positive illusions: Idealization and the construction of satisfaction
in close relationships. Journal of Personality and Social Psychology, 70, 79–98.
Neff, L.A., & Karney, B.R. (2005). To know you is to love you: The
implications of global adoration and specific accuracy for marital
relationships. Journal of Personality and Social Psychology, 88,
480–497.
Volume 20—Number 11
Shanhong Luo and Anthony G. Snider
Pinney, E.M., Gerrard, M., & Denney, N.W. (1987). The Pinney Sexual
Satisfaction Inventory. Journal of Sex Research, 23, 233–251.
Reis, H.T., & Shaver, P. (1988). Intimacy as an interpersonal process.
In S.W. Duck (Ed.), Handbook of personal relationships (pp. 367–
389). New York: Wiley.
Rusbult, C.E., Van Lange, P.A.M., Wildschut, T., Yovetich, N.A., &
Verette, J. (2000). Perceived superiority in close relationships:
Why it exists and persists. Journal of Personality and Social
Psychology, 79, 521–545.
Simms, E.N., & Watson, D. (2006). Interpersonal relationships: Assessment and personality correlates. Unpublished manuscript,
University of Iowa, Iowa City.
Swann, W.B., De La Ronde, C., & Hixon, J.G. (1994). Authenticity and
positivity strivings in marriage and courtship. Journal of Personality and Social Psychology, 66, 857–869.
Taylor, S.E., & Brown, J.D. (1988). Illusion and well-being: A social
psychological perspective on mental health. Psychological Bulletin, 103, 193–210.
Volume 20—Number 11
Watson, D., Clark, L.A., & Tellegen, A. (1988). Development and
validation of brief measures of positive and negative affect: The
PANAS scales. Journal of Personality and Social Psychology, 54,
1063–1070.
Watson, D., Hubbard, B., & Wiese, D. (2000a). General traits of personality and affectivity as predictors of satisfaction in intimate
relationships: Evidence from self- and partner-ratings. Journal of
Personality, 68, 413–449.
Watson, D., Hubbard, B., & Wiese, D. (2000b). Self-other agreement in
personality and affectivity: The role of acquaintanceship, trait
visibility, and assumed similarity. Journal of Personality and
Social Psychology, 78, 546–558.
(RECEIVED 11/7/08; REVISION ACCEPTED 2/16/09)
1339