Andrew Karpinski

The single category implicit association test as a measure of implicit social cognition

The Single Category Implicit Association Test (SC-IAT) is a modification of the Implicit Association Test that measures the strength of evaluative associations with a single attitude object. Across 3 different attitude domains--soda brand preferences, self-esteem, and racial attitudes--the authors found evidence that the SC-IAT is internally consistent and makes unique contributions in the ability to understand implicit social cognition. In a 4th study, the authors investigated the susceptibility of the SC-IAT to faking or self-presentational concerns. Once participants with high error rates were removed, no significant self-presentation effect was observed. These results provide initial evidence for the reliability and validity of the SC-IAT as an individual difference measure of implicit social cognition.

When grades determine self-worth: Consequences of contingent self-worth for male and female engineering and psychology majors

The impact of grades on daily self-esteem, affect, and identification with major was examined in a sample of 122 male and female students majoring in engineering and psychology. Self-esteem, affect, and identification with major increased on days students received good grades and decreased on days they received poor grades; basing self-esteem on academic competence moderated the effect of bad grades. Bad grades led to greater drops in self-esteem but not more disidentification with the major for women in engineering. Instability of self-esteem predicted increases in depressive symptoms for students initially more depressed.

Measuring Self-Esteem using the Implicit Association Test: The Role of the Other

In two experiments, the use of a self-other Implicit Association Test (IAT) as a measure of implicit self-esteem was examined. In Study 1, participants completed two self-other IATs: one in which the other was unspecified and one in which the other was specified to be a close friend. Esteem-IAT scores were significantly higher for the unspecified other-IAT (d =. 82), indicating that the content of the other has a considerable influence on the over-all esteem-IAT scores. Study 2 provided a conceptual replication of these results and provided some initial evidence that the valence of an unspecified other is negative. Across both studies, the nature of the mental representation of the self, as measured by an esteem-IAT, changed as a function of the mental representation of the other. The crucial role of the other-associations in the esteem-IAT calls into question its use and interpretation as a measure of self-esteem.

Attitude Importance as a Moderator of the Relationship Between Implicit and Explicit Attitude Measures

The authors examined attitude importance as a moderator of the relationship between the Implicit Association Test (IAT) and explicit attitude measures. In Study 1 (N = 194), as ratings of attitude importance regarding the 2000 presidential election increased, the strength of the relationship between a Bush-Gore IAT and explicit attitude measures also increased. Study 2 provided a conceptual replication of these results using attitudes toward Coke and Pepsi (N = 112). In addition, across both studies, explicit attitude measures were better predictors of deliberative behaviors than IAT scores. In Study 3 (N = 77), the authors examined the role of elaboration as a mechanism by which attitude importance may moderate IAT-explicit attitude correlations. As predicted, increased elaboration resulted in stronger IAT-explicit attitude correlations. Other possible mechanisms by which attitude importance may moderate the IAT-explicit attitude relationship also are discussed.

Implicit and Explicit Attitudes Toward Gay Males and Lesbians Among Heterosexual Males and Females

ABSTRACT In this article, we examined explicit and implicit attitudes toward gay males and lesbians using the Single Category IAT (SC-IAT). In Study 1, we examined attitudes toward gay people in general. Participants reported positive explicit attitudes and neutral implicit attitudes toward gay people. In Study 2, we examined implicit and explicit attitudes toward gay men and lesbians separately. Participants rated gay men and lesbians positively on explicit attitude measures. Analysis of SC-IAT scores revealed neutral associations with gay men and positive associations with lesbians. As a secondary goal, we also tested the Balanced Identity Theory in both studies and did not find evidence of balance between implicit sexual orientation attitudes, implicit sexual identity, and implicit self-esteem using the SC-IAT.

Thinking about interracial interactions

White people who have difficulty implicitly pairing black names with positive words also tend to be impaired on tasks requiring cognitive control after interacting with a black experimenter. A new functional imaging study finds that such subjects also show more activity in brain regions associated with cognitive control when looking at black faces that are irrelevant to their task.

Body weight contingency of self-worth

The body weight contingency of self-worth (body weight CSW) is the tendency for people to base their self-esteem on body weight. In two studies, the body weight contingency was evaluated against a general appearance contingency of self-worth (appearance CSW) to investigate whether or not individuals may base self-esteem on body weight. Compared to the appearance CSW, the body weight CSW predicted unique variance in self-esteem, instability of self-esteem, body shape anxiety, and several aspects of psychological dysfunction. Additionally, the body weight CSW moderated relationships between body weight and relevant outcome variables, whereas the appearance CSW did not. Gender and ethnic differences in the tendency to base self-worth on body weight were also explored.

Assessing dependency using self-report and indirect measures: examining the significance of discrepancies.

The present study addressed convergence between self-report and indirect approaches to assessing dependency. We were moderately successful in validating an implicit measure, which was found to be reliable, orthogonal to 2 self-report instruments, and predictive of external criteria. This study also examined discrepancies between scores on self-report and implicit measures, and has implications for their significance. The possibility that discrepancies themselves are pathological was not supported, although discrepancies were associated with particular personality profiles. Finally, this study offered additional evidence for the relation between dependency and depressive symptomatology and identified implicit dependency as contributing unique variance in predicting past major depression.