A Diffusion Model Approach for Understanding the Impact of 17 Interventions on the Race Implicit Association Test

Abstract

Performance on implicit measures reflects construct-specific and nonconstruct-specific processes. This creates an interpretive issue for understanding interventions to change implicit measures: Change in performance could reflect changes in the constructs of interest or changes in other mental processes. We reanalyzed data from six studies ( N = 23,342) to examine the process-level effects of 17 interventions and one sham intervention to change race implicit association test (IAT) performance. Diffusion models decompose overall IAT performance ( D-scores) into construct-specific (ease of decision-making) and nonconstruct-specific processes (speed–accuracy trade-offs, non-decision-related processes like motor execution). Interventions that effectively reduced D-scores changed ease of decision-making on compatible and incompatible trials. They also eliminated differences in speed–accuracy trade-offs between compatible and incompatible trials. Non-decision-related processes were affected by two interventions only. There was little evidence that interventions had any long-term effects. These findings highlight the value of diffusion modeling for understanding the mechanisms by which interventions affect implicit measure performance.

Lying on the Dissection Table: Anatomizing Faked Responses

“…But an immense body of research has shown that people are able to fake on psychological measures (e.g., Birkeland et al, 2006;Viswesvaran & Ones, 1999). Even going beyond classical tests, measures that had originally been considered to be immune against faking (e.g., Implicit Association Tests; IATs; Greenwald et al, 1998) have turned out to be fakeable (e.g., Röhner et al, 2011;Röhner & Lai, 2021). As faking results in changes in test scores and rank orders, it is Faking Detection as a Solution?…”

“…But an immense body of research has shown that people are able to fake on psychological measures (e.g., Birkeland et al, 2006 ; Viswesvaran & Ones, 1999 ). Even going beyond classical tests, measures that had originally been considered to be immune against faking (e.g., Implicit Association Tests; IATs; Greenwald et al, 1998 ) have turned out to be fakeable (e.g., Röhner et al, 2011 ; Röhner & Lai, 2021 ). As faking results in changes in test scores and rank orders, it is a serious problem that can impair the validity of tests (e.g., Salgado, 2016 ; see Ziegler et al, 2012 , for an overview), and this impairment of validity may be higher for construct validity than for criterion validity (e.g., Ones & Viswesvaran, 1998 ; Ziegler & Buehner, 2009 ).…”

Science-utility and science-trust associations and how they relate to knowledge about how science works

“…An algorithm based on the difference in mean reaction times is used as an indicator of the association between the categories [see 38 for how issues such as false responses and long latencies are dealt with]. Even though the IAT score has been criticized for containing aspects that are not process-specific, such as speed-accuracy trade-offs, it clearly captures valid and construct-specific processes [ 39 , 40 ] and has been shown to have predictive validity [e.g., 41 , 42 ]. IATs have been used to measure both attitudes and beliefs by combining the target words with either evaluative terms (e.g., good/bad) or another category (e.g., female/male).…”

Judgment of Crowds as Emotional Increases with the Proportion of Black Faces

“…In the past decade, drift diffusion models have been used to characterize instances of bias sensitivity in decisions related to perceived race (Benton & Skinner, 2015;Correll et al, 2015;Johnson et al, 2017Johnson et al, , 2021Röhner & Lai, 2021). Each of these studies suggests that the effect of perceived race on social decisions is driven by increased sensitivity, rather than bias, to Black faces.…”