Relationships between the race implicit association test and other measures of implicit and explicit social cognition

Abstract

Background: The race-based Implicit Association Test (IAT) was proposed to measure individual differences in implicit racial bias subsumed within social cognition. In recent years, researchers have debated the theoretical tenets underpinning the IAT, questioning whether performance on this task: (1) measures implicit attitudes that operate automatically outside of conscious awareness; (2) reflects individual differences in social cognition; and (3) can predict social behavior. One way to better address these research questions is to assess whether the race-IAT correlates with other implicit processes that are subsumed within social cognition.

Aims: The current study assessed whether the race-IAT was related to other commonly used individual difference measures of implicit (and explicit) social cognition. Experiment 1 assessed whether dissociable patterns of performance on the race-IAT were related to measures of implicit imitative tendencies, emotion recognition and perspective taking toward White task actors, as well as explicit measures of trait and state affective empathy and racial bias. Overcoming limitations of task conceptual correspondence, Experiment 2 assessed whether these latter tasks were sensitive in detecting racial biases by using both White and Black task actors and again examined their relationships with the race-IAT.

Method: In two lab-based experiments, 226 and 237 participants completed the race-IAT followed by an extensive battery of social cognition measures.

Results: Across both experiments, pro-White/anti-Black bias on the race-IAT was positively related to a pro-White bias on explicit measures of positive affective empathy. However, relationships between the race-IAT and implicit imitative tendencies, perspective taking, emotion recognition, and explicit trait and negative state affective empathy were statistically equivalent.

Conclusion: The race-IAT was consistently related to explicit measures of positive state affective empathy but not to other individual difference measures of implicit social cognition. These findings are discussed with regards to the theoretical underpinnings of the race-IAT as an individual difference measure of implicit social cognition, as well as alternative explanations relating to the reliability of social cognition measures and the various combinations of general-purpose (social and non-social) executive processes that underpin performance on these tasks.

Keywords: implicit measures; individual differences; race IAT; race implicit association test; social cognition; theory.

Figure 1

Procedural overview of experimental tasks employed in Experiment 1 (White actor only) and Experiment 2 (White and Black actors). Stimuli from the IAT, SRC, eGNG, and VPT are publicly available for research use and represent actual exemplars. Stimuli from the AET were selected from the IAPS and are not presented here due to copyright restrictions; for this figure we have replaced these with publicly available alternatives. Written informed consent was obtained from the individuals for the publication of any potentially identifiable images included in this article.

Figure 2

Violin plots illustrating the frequency distributions of IAT scores within each cluster. The red cross indicates cluster means and standard deviations (horizontal and vertical lines, respectively).