A Structural and Process Analysis of the Implicit Association Test

Abstract

The Implicit Association Test (IAT) is based on the observation that participants find it easier to respond in the same way to exemplars of two concepts when these concepts are similar (e.g., "positive" and "flower") compared to when the concepts are dissimilar (e.g., "positive" and "insect"). In the first part of this article, I argue that the IAT is structurally similar to stimulus-response compatibility tasks. On the basis of this analogy, I then present two response conflict accounts of IAT effects. The data of an experiment that was designed to test these accounts showed that IAT effects reflect attitudes toward the target concepts rather than attitudes toward the individual exemplars of those concepts. The results shed light on the processes that underlie IAT effects, suggest that automatic attitude activation may depend on the construal of the object that is fostered by the context, and clarify the relation between different indirect measures of attitudes. © 2001 Academic Press Greenwald, McGhee, and Schwartz (1998) recently introduced the Implicit Association Test (IAT). The simple but ingenious idea behind the IAT is that it should be easier to map two concepts onto a single response when those concepts are somehow similar or associated in memory than when the concepts are unrelated or dissimilar. To test this idea, Greenwald et al. (1998, Experiment 1) presented names of flowers (e.g., TULIP), names of insects (e.g., SPIDER), positive words (e.g, LOVE), and negative words (e.g., UGLY) on a computer screen. Participants were asked to categorize these words by pressing one of two keys. It can be assumed on a priori grounds that the concept "flower" and the concept "positive" are associated in memory, whereas the concept "insect" is associated with the concept "negative." Therefore, when "flower" and "positive" are both assigned to one key and "insect" and "negative" are both assigned to a second key, responses should be fast because the response assignments are compatible with existing associations in memory. When response assignments are incompatible with existing associations (e.g., press left for "insect" and "positive"; press right for "flower" and "negative"), responses should be slower. The results clearly confirmed that reaction times were faster with compatible than with incompatible response assignments. A STRUCTURAL ANALYSIS OF THE IATIn this section, I argue that the IAT is structurally similar to stimulus-response compatibility tasks. It has long been know that responses can be emitted more quickly and accurately if the responses are somehow similar to the stimuli to which these responses need to be made than when the responses and stimuli are dissimilar (e.g., Fitts & Seeger, 1953;Kornblum & Lee, 1995). For instance, if participants are asked to press a left key in response to a stimulus presented on the left side of a screen and to press a right key in response to a stimulus presented on the right side of the screen, responses are faster and more accurate than when the response assignmen...

Associative and propositional processes in evaluation: An integrative review of implicit and explicit attitude change.

“…Whereas single presentations of the CS typically reduce conditioning effects in PC, such single presentations leave EC effects unaffected (e.g., Baeyens, Crombez, Van den Bergh, & Eelen, 1988). Taken together, these results suggest that, even though PC effects may be driven by higher order propositional processes (e.g., Holyoak et al, 1989;Lovibond, 2003;Williams, 1995), EC effects are better explained by a qualitatively different, associative mechanism (De Houwer et al, 2001).…”

“…For instance, whereas affective priming tasks seem to be primarily influenced by automatic evaluations of the particular exemplars used as prime stimuli (e.g., Livingston & Brewer, 2002;, the Implicit Association Test seems to be influenced by the valence of both the individual exemplars (e.g., Blüzmke & Friese, 2006;Govan & Williams, 2004; and the 715 IMPLICIT AND EXPLICIT ATTITUDE CHANGE particular categories that are applied to these exemplars within the task (e.g., De Houwer, 2001;. Thus, when testing hypotheses derived from the APE model, it is important to consider whether the attitude object in question is represented on the exemplar level or on the level of general categories and whether the evaluation level in the implicit measure corresponds to the one implied in the explicit measure.…”

“…Whereas single presentations of the CS typically reduce conditioning effects in PC, such single presentations leave EC effects unaffected (e.g., Baeyens, Crombez, Van den Bergh, & Eelen, 1988). Taken together, these results suggest that, even though PC effects may be driven by higher order propositional processes (e.g., Holyoak et al, 1989;Lovibond, 2003;Williams, 1995), EC effects are better explained by a qualitatively different, associative mechanism (De Houwer et al, 2001).With regard to the role of propositional processes in human reasoning, we generally agree with the argument that many processes that have been described as associative may actually be propositional (e.g., Erb et al, 2003;Osman, 2004). In fact, we claim that any kind of reasoning process is inherently propositional, given that reasoning is concerned with validation and the assessment of truth values (Deutsch & Strack, in press).…”

“…This research has been reviewed extensively elsewhere (De Houwer et al, 2001Walther et al, 2005), and thus we refrain from an elaborate discussion of the relevant evidence. From the perspective of the APE model, however, it is important to note that processes of EC should not directly influence evaluative judgments.…”

Implicit Cognitive Distortions and Sexual Offending

“…Crucially, implicit cognition comes with a method for testing some of its assumptions and for measuring implicit associations. Greenwald et al's (1998) method of measuring implicit associations is accumulating evidence at a rapid pace in a number of areas (e.g., Ashburn-Nardo, Voils, & Monteith, 2001;Dasgupta, McGhee, Greenwald, & Banaji, 2000;De Houwer, 2001;Greenwald & Farnham, 2000;Milne & Grafman, 2001;Rudman, Greenwald, Mellott, & Schwartz, 1999;Teachman, Gregg, & Woody, 2001). Greenwald suggests that people learn implicit associations between social targets (e.g., woman or man) and social attributes (e.g., desirable) and that these can be measured using a latency-response time paradigm.…”