RICHARD W. ROBINS
Department of
Psychology
University
of California
Davis, CA
95616-8686
(530)
754-8299
rwrobins@ucdavis.edu
My research program involves three lines of investigation. First, I study how individual differences in self-esteem develop across the lifespan, play out in particular social interactions, and influence important life outcomes. Second, I study the implications of personality for psychological functioning, particularly during adolescence and young adulthood. Third, I address meta-level issues concerning psychology as a science, and use empirical methods to document historical trends in the field. Several themes are common to my research, and illustrate my general approach to the study of self-esteem and personality: The use of both experimental and longitudinal designs; the use of multiple measures and data sources; the study of persons in naturalistic interactions, contexts, and relationships; an emphasis on individual differences as well as general processes; and an attempt to understand development across the lifespan.
I.
Self-Esteem Processes and Development
A. Biases in self-perception. My research in this area focuses on how people process information about themselves and others. I draw on social-cognitive models of information processing and psychodynamic models of affect regulation and defense. I am particularly interested in the affective and motivational processes that underlie biased self-evaluations. For example, one study showed that inducing a state of self-focused attention accentuates self-perception bias in some individuals but reduces bias in other individuals (Robins & John, 1997, Psychological Science; this article was summarized in Science News and in a number of popular magazines). In another study, we found that although individuals' recollections of their behavior are fairly accurate, they tend to remember engaging in more positive behaviors and fewer negative behaviors than actually occurred (as documented by a team of coders who reviewed videotapes of the interaction) (Gosling, Craik, John, & Robins, 1998, Journal of Personality and Social Psychology). Interestingly, some individuals were particularly inclined to over-report engaging in desirable behaviors, and these individuals were characterized by a distinctive set of psychological traits. This study also extended my previous research showing that properties of the behavior being judged (e.g., observability, content domain, evaluativeness) moderate how much bias an individual shows (John & Robins, 1993, Journal of Personality).
My collaborators and I have conducted several studies showing that people's explanations for their own behavior often diverge from the explanations given by others, although the overall pattern of self-other discrepancies led us to question the validity of the widely accepted "actor-observer attribution effect" (Robins, Spranca, & Mendelsohn, 1996, Journal of Personality and Social Psychology). In our current research, we are testing whether people's causal explanations for their behavior vary across communication media (i.e., face-to-face vs. phone vs. computer-mediated). I recently received an NIH grant (start date: September, 2000) to pursue this line of research.
In addition to examining the factors that underlie self-perception bias, I am also interested in studying their consequences. A longstanding debate in psychology concerns whether it is adaptive to have illusions about one's characteristics and capabilities: Are individuals who have biased self-perceptions happier and more successful than those who have accurate self-perceptions? Studies by my research group suggest that self-perception biases are a mixed blessing, promoting short-term well-being at the cost of long-term adjustment (Robins & Beer, in press, Journal of Personality and Social Psychology; Robins & Paulhus, in press). For example, our laboratory research shows that self-perception bias promotes a temporary rise in positive affect; however, our longitudinal research shows that over the course of college self-perception bias is associated with declines in well-being and self-esteem, disengagement from the academic context, and a tendency to drop out of college.
B. The development of self-esteem across the lifespan. After almost a century of research on self-esteem, we still know little about its normative development across the lifespan. Most studies focus on development within particular age periods, primarily adolescence. To redress this gap, we recently charted age differences across the entire lifespan using cross-sectional data on a very large and diverse sample of individuals ranging in age from 9 to 90 (N = 326,000) (Robins, Trzesniewski, Tracy, Gosling, & Potter, in press, Psychology and Aging). The findings demonstrated a number of important age differences: self-esteem levels were high in childhood, dropped during adolescence, rose gradually throughout adulthood, and declined sharply in old age. On average, men had higher self-esteem than women did. This gender gap emerged in adolescence and persisted throughout adulthood, but by old age men and women again reported similar levels of self-esteem. We are currently working on a meta-analysis of the self-esteem literature to confirm the developmental trends observed in this study (Trzesniewski, Donnellan, & Robins, in preparation).
In addition to documenting normative age differences in self-esteem, I am also trying to understand the processes that explain why one person declines in self-esteem while another increases. In one study, we used longitudinal data to examine the factors that promote self-esteem during adolescence (Block & Robins, 1993, Child Development; summarized in Science News). Consistent with the large-scale cross-sectional study, we found a gender divergence in self-esteem emerging during adolescence. Further analyses showed that interpersonal qualities such as warmth and nurturance promote self-esteem in women whereas the absence of anxious or ruminating thoughts promotes self-esteem in men. In another line of research on the processes involved in self-esteem development, we found that found that individuals who believe their intelligence is a fixed quantity that cannot be increased tend to decline in self-esteem over time (Robins & Pals, in press, Self and Identity).
My research on self-esteem development also involves attention to measurement issues. In a series of four studies, we demonstrated the reliability and validity of two measures of global self-esteem, and showed that self-esteem can be measured by a single item. However, our findings suggest that this single-item measure is not valid for young children and that its validity is moderated by response latency (Robins, Hendin, & Trzesniewski, in press, Personality and Social Psychology Bulletin). In other research, we are testing whether self-esteem can be assessed using a reaction time measure of automatic (i.e., implicit) associations to stimuli that are self-relevant or non-self-relevant (Robins & Hendin, 1998, APA presentation).
C. Theoretical perspective on the self. In addition to my empirical research on self-esteem processes and development, I've written several theoretical papers on the self. In a chapter published in the Handbook of Personality Psychology, I developed a conceptual framework that organizes the self-concept literature into four different theoretical perspectives, and highlights the unique set of cognitive, affective, and motivational processes associated with each perspective (Robins & John, 1997). In a more recent chapter, I further developed this framework and argued that scientific research on the self should be: (a) central to any theory of personality functioning; (b) organized around evolutionary, functionalist explanations; (c) informed by comparative, cross-species research; and (d) focused on basic psychological processes such as attention, memory, and emotion, and associated neural mechanisms (Robins, Norem, & Cheek, 1999, Handbook of Personality: Theory and Research).
II. Personality and its Consequences for
Psychological Functioning
A. Personality and relationship outcomes. My research in this area examines how stable
personality traits shape the quality of romantic relationships. In 1998, I received a fellowship to conduct
research on personality and relationship outcomes at the Social, Genetic, and
Developmental Psychiatry Research Group of the University of London. This research uses data from the Dunedin
Longitudinal Study, a 30-year longitudinal study that has repeatedly assessed a
representative birth cohort since age 3.
In our first study, we tested six models of the independent and
interactive effects of personality on each partner’s reports of relationship
satisfaction and quality (Robins, Caspi, & Moffitt, 2000, Journal of
Personality and Social Psychology). We
found that a woman’s relationship happiness is predicted by her partner’s low
Negative Emotionality, high Positive Emotionality, and high Constraint, whereas
a man’s relationship happiness is predicted only by his partner’s low Negative
Emotionality. We also found evidence of
additive but not interactive effects: Each partner’s personality contributed
independently to relationship outcomes but not in a synergistic way. In a second paper examining the link between
personality and physical abuse in relationships, we found that: (1)
perpetrators’ characteristics (specifically negative emotionality) predict
their abusive behavior, (2) victims’ characteristics (specifically negative
emotionality) predict their victimization, (3) both partners’ characteristics
independently and jointly predict couples’ reciprocal abuse, supporting a
dyadic process model (Moffitt, Robins, & Caspi, 2000, in press, Crime and
Justice Research). Together, these two
papers show that the most important ingredient of a happy and non-abusive
relationship is that both partners are low in their tendency to experience and
express negative emotions. The tendency
to experience positive emotions and the ability to effectively regulate one's
impulses also influence relationship outcomes, but their effects tend to be
weaker and less consistent.
I received an NIMH grant to continue this
line of research with Drs. Caspi and Moffitt.
Our most recent paper examines the effects of personality on romantic
relationships using longitudinal data from the Dunedin study. We found that antecedent personality traits,
assessed at age 18, predicted relationship satisfaction at age 26 as well as
change in relationship satisfaction from age 21 to 26. The findings led us to question the
prevailing view that relationship outcomes are generated through dyadic
processes unique to each couple; instead, we concluded that stable personality
traits contribute to an individual's general level of satisfaction (or
dissatisfaction) across relationships.
A manuscript describing this research will appear in a special issue of
the Journal of Personality on "personality and relationships"
(Robins, Caspi, & Moffitt, in press).
B. Personality and behavioral development during adolescence. My research in this area relates personality characteristics to developmentally significant outcomes in adolescence, including juvenile delinquency, school performance, and psychopathology. In one study, we identified three replicable personality types in a sample of high- and low-risk adolescent boys (Robins, John, Caspi, Moffitt, & Stouthamer-Loeber, 1996, Journal of Personality and Social Psychology; Robins, John, & Caspi, 1998). The types (labeled Resilient, Overcontrolled, and Undercontrolled) differed in the prevalence of internalizing and externalizing disorders, delinquency, school performance, and intellectual ability. Over the past several years, this research has received a considerable amount of attention and several other research groups have published cross-cultural replications of our findings.
C. Personality development during young adulthood. Several of my current research projects use data from my own ongoing longitudinal study of personality and self-esteem development during young adulthood. In one study, we used a variety of statistical procedures to examine different forms of continuity and change in personality over four years of college (Robins, Fraley, Roberts, & Trzesniewski, 2001, Journal of Personality). In another study using the longitudinal data, we found that personality traits predict the particular set of goals that a person chooses to pursue in life, and we identified personality characteristics that determine whether a person prioritizes achievement (i.e. work) goals over relationship (i.e., social) goals (Roberts & Robins, 2000, Personality and Social Psychology). We also recently completed a chapter reviewing current research on personality development for the Handbook of the Life Course (Roberts, Robins, Caspi, & Trzesniewski, in press).
III. An Empirical Approach to the History of Psychology
In a third line of research, I take a meta-perspective on the field of psychology, drawing on theoretical perspectives in the philosophy and sociology of science. For example, one recent study tracked trends in psychology over the past 5 decades using three empirical indicators of scientific prominence (Robins, Gosling, & Craik, 1999, American Psychologist). This article has generated a great deal of interest; the American Psychologist subsequently published six comments on the article along with our rejoinder (Robins, Gosling, & Craik, 2000). Moreover, figures from this article have been (or will be) reprinted in several textbooks. These recent studies on trends in psychology build on my earlier interest in developing empirical tests of how ideas and findings become prominent in psychological science (Robins & Craik, 1993), as well as my more general interest in the philosophy of science (Robins & Craik, 1994).
Return to the UC Davis Department of Psychology Home Page