Reflections on the Difference Between Implicit Bias and Bias on Implicit Measures

Abstract

We are pleased about the considerable interest in our target article and that there is overwhelming agreement with our central thesis that, if the term implicit is understood as unconscious in reference to bias, implicit bias (IB) should not be equated with bias on implicit measures (BIM) (Cesario, this issue; Corneille & B ena, this issue; Cyrus-Lai et al., this issue; De Houwer & Boddez, this issue; Dovidio & Kunst, this issue; Melnikoff & Kurdi, this issue; Norman & Chen, this issue; Olson & Gill, this issue; Schmader et al., this issue; but see Krajbich, this issue; Ratliff & Smith, this issue). We are also grateful for the insightful commentaries, which continue to advance the field’s thinking on this topic. The comments inspired us to think further about the relation between IB and BIM as well as the implications of a clear distinction between the two. In the current reply, we build on these comments, respond to some critical questions, and clarify some arguments that were insufficiently clear in our target article. Before doing so, we would like to express our appreciation for the extreme thoughtfulness of the commentaries, every single one of which deserves their own detailed response. For the purpose of this reply, we will focus on recurring themes and individual points that we deem most important for moving forward. We start our reply with basic questions about the concept of bias, including the difference between behavioral effects and explanatory mental constructs, the role of social context, goals, and values in evaluating instances of bias, and issues pertaining to the role of social category cues in biased behavior. Expanding on the analysis of the bias construct, the next sections address questions related to the implicitness of bias, including the presumed unconsciousness of BIM, methodological difficulties of studying unconscious effects, and the implications of a broader interpretation of implicit as automatic. The next sections again build on the discussions in the preceding sections, addressing questions about the presumed significance of IB research for understanding societal disparities and the value of BIM research if IB is treated as distinct from BIM. The final section presents our general conclusions from the conversation about our target article and several suggestions on how to move forward. Reflections on Bias