Association for Behavior Analysis International

The Implicit Relational Assessment Procedure (IRAP) was created to measure derived relational responding (i.e., response bias) especially toward socially-charged stimuli. The IRAP measures latency between a) various stimulus pairing onsets and b) responses to those pairings via rules given to participants. IRAP response allocations are particularly interesting when they do not match self-report verbal descriptions of preference; such differences may indicate inconsistencies between implicit and explicit biases. Drake et al. (2015) examined response-time biases for the 2012 presidential election and found IRAP scores correlated with self-reports. However, unbalanced representation of political affiliation was a limitation regarding the results. The studies presented herein pursued a more balanced representation of political affiliation, while examining three different stimulus pairings that were potentially relevant to the 2016 election: presidential candidate names (Donald Trump and Hillary Clinton), socioeconomic status (rich people and poor people), and gender (man and woman). These stimuli were paired with a good or bad evaluative word during each trial. Correlations between IRAP and self-report data were examined. Implications for basic research with the IRAP, as well as the social significance of the findings, will be discussed.

The Implicit Relational Assessment Procedure (IRAP) is a latency-based behavioral measure of “natural verbal relations” (Barnes-Holmes, Hayden, & Barnes-Holmes, 2008). These natural verbal relations include what some call implicit biases – quick responses which are influenced by one’s conditioning history and yet may differ from one’s own explicit report. The IRAP bears an advantage over other implicit measures in that it may assess more specific relations among stimulus combinations than other measures. For example, the IRAP has been used to demonstrate that negative evaluations of candidates and not just positive evaluations factor into voting intentions (Drake et al., 2015). The 2016 presidential election is noteworthy due to substantially negative views of each candidate, even among voters from their respective political parties. Furthermore, past studies of political attitudes with implicit measures have shown a high degree of convergence with self-report. The current study utilized data provided by college undergraduates from two universities in the mid-western and southern United States. Although self-reports revealed negative and/or neutral sentiment toward both candidates, the IRAP generally revealed positive and/or neutral effects and little evidence of convergence with self-report. These data will be discussed in light of methodological considerations as well as future directions in IRAP research.