Academic journal article Canadian Journal of Experimental Psychology

The Implicit Association Test as a General Measure of Similarity

Academic journal article Canadian Journal of Experimental Psychology

The Implicit Association Test as a General Measure of Similarity

Article excerpt

Abstract

The Implicit Association Test (IAT) is widely used as a measure of semantic similarity (i.e., associations in semantic memory). The results of previous research and of a new study show that IAT effects can, however, also be based on other types of similarity between stimuli. We therefore put forward the hypothesis that the IAT provides a general measure of similarity. Given that similarity is highly dynamic and context-dependent, our view that the IAT measures similarity is compatible with existing evidence showing that IAT effects are highly malleable. We provide further evidence for this in a new study in which the outcome of an IAT depended on whether the perceptual or functional characteristics of the stimuli were made salient.

Within a few years time, the Implicit Association Test (IAT) has become a widely used tool to study attitudes and other associations in memory. This task, which was introduced by Greenwald, McGhee, and Schwartz (1998), is based on the idea that it should be easier to respond to stimuli that represent a particular concept when the same response is also assigned to exemplars of a second associated concept than when the same response is also required for exemplars of a second unassociated concept. For instance, participants might be asked to press one of two keys on the basis of whether the presented word is the name of a flower, the name of an insect, a positive adjective, or a negative adjective. Results show that performance is better when flower names and positive words are assigned to one key and insect names and negative words are assigned to the second key (compatible task) than when pressing the first key is correct for insect names and positive words and pressing the second key is correct for flower names and negative words (incompatible task) (e.g., Greenwald et al., 1998, Experiment 1). This pattern of results is in line with the a priori assumption that, in semantic memory, the target concept "flower" is associated more strongly with the attribute concept "positive" than with the attribute concept "negative" whereas the reverse is true for the target concept "insect."

Results such as these suggest that the IAT can be used as a tool to measure and study associations in semantic memory. As an illustration, we will briefly explain how the IAT has been used to measure selfesteem (defined as the relative strength of the associations between the target concept "self and the attribute concepts "positive" and "negative," see Greenwald et al., 2002). In a self-esteem IAT, words are presented that represent the concept "self (e.g, ME), the concept "not-self" or "other" (e.g., THEM), the concept "positive" (e.g., FRIENDLY), and the concept "negative" (e.g., HOSTILE). Results typically show that performance is better in a task where self and positive items are assigned to one key and not-self and negative items are assigned to a second key than in a task where the assignments are reversed (i.e., first key for self and negative items; second key for not-self and negative items; e.g., Greenwald & Farnham, 2000). Such a pattern of results is in line with the idea that people on average possess positive self-esteem. Interindividual differences in the self-esteem IAT effect have also been interpreted as a reliable and valid indicator of actual differences in (implicit) self-esteem (e.g., Greenwald & Farnham, 2000).

Given the potential practical benefits of the IAT, it is important to have a good theoretical understanding of the task. Theoretical research on the IAT has been directed toward two questions: the what-question arid the how-question. The first question is about what the IAT measures. That is, what aspects of the stimuli or knowledge about the stimuli can determine the direction and magnitude of IAT effects? As can be inferred from the above, the most widely accepted view is that IAT effects reflect associations between concepts in semantic memory (e. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.