The Challenge of Diagnostic Inferences From Implicit Measures: The Case of Non-Evaluative Influences in the Evaluative Priming Paradigm

Abstract

Implicit measures are diagnostic tools to assess attitudes and evaluations that people cannot or may not want to report. Diagnostic inferences from such tools are subject to asymmetries. We argue that (causal) conditional probabilities p(AM+|A+) of implicitly measured attitudes AM+ given the causal influence of existing attitudes A+ is typically higher than the reverse (diagnostic) conditional probability p(A+|AM+), due to non-evaluative influences on implicit measures. We substantiate this argument with evidence for non-evaluative influences on evaluative priming—specifically, similarity effects reflecting the higher similarity of positive than negative prime-target pairs; integrativity effects based on primes and targets’ potential to form meaningful semantic compounds; and congruity proportion effects that originate in individuals’ decisional strategies. We also cursorily discuss non-evaluative influences in the Implicit Association Test (IAT). These influences not only have implications for the evaluative priming paradigm in particular, but also highlight the intricacies of diagnostic inferences from implicit measures in general.

The effect of group membership and individuating information on automatic and deliberate evaluation of well-known people.

“…The consistency of the results across different conceptual replications is a constraint on any alternative account that would attribute the results to constructs other than those that we attempted to test (i.e., deliberate and automatic judgment). On the other hand, the theoretical and empirical work that has doubted the usefulness of umbrella term automatic judgment and the validity of the measures that we used here (Brown-Iannuzzi et al, 2019; Greenwald & Nosek, 2001; McFarland & Crouch, 2002; Mierke & Klauer, 2003; Olson & Fazio, 2004; Steffens, 2004; Teige-Mocigemba & Klauer, 2008; Uhlmann et al, 2006; Unkelbach & Fiedler, 2020) provide grounds to doubt our interpretation of the present findings.…”

“…Research has also found that the EPT is a better predictor of less controlled behaviors, relative to self-report (e.g., Dovidio et al, 1997Dovidio et al, , 2002Spruyt et al, 2015). On the other hand, there is also evidence that the EPT can sometimes reflect non-automatic judgment (Teige-Mocigemba & Klauer, 2008;Unkelbach & Fiedler, 2020), and is sensitive to factors that are unrelated to judgment (see review in Unkelbach & Fiedler, 2020).…”

The effect of practice on automatic evaluation: A registered replication

“…Because there is evidence that practice increases automaticity in research about non-evaluative responses (Carlson & Lundy, 1992;Kramer et al, 1990;Schneider & Fisk, 1984;Schneider & Shiffrin, 1977), an effect of practice of evaluation on the priming effect in the EPT would support the contention that the EPT captures automatic evaluation. This seems particularly relevant presently, when the validity of reaction-time based measures of evaluation, such as the EPT, is being frequently challenged (e.g., Greenwald & Nosek, 2001;McFarland & Crouch, 2002;Mierke & Klauer, 2003;Steffens, 2004;Teige-Mocigemba & Klauer, 2008;Unkelbach & Fiedler, 2020).…”

Is the automatic evaluation of individual group members inherently biased by their group membership?

“…Yet, indirect measures of evaluation are not perfectly valid because they sometimes capture processes that might be unrelated to judgment (Brown-Iannuzzi et al, 2019;Olson & Fazio, 2004;Uhlmann et al, 2006) or to automaticity (Greenwald & Nosek, 2001;McFarland & Crouch, 2002;Mierke & Klauer, 2003;Steffens, 2004;Teige-Mocigemba & Klauer, 2008;Unkelbach & Fiedler, 2020). Therefore, evidence collected with such indirect measures is only suggestive, especially if not replicated with various, different indirect measures.…”