Voluntary controllability of the implicit association test (IAT)

Greenwald, McGhee, and Schwartz showed that white participants indicated a more positive evaluative association with whites than with blacks in the Implicit Association Test (IAT), and were being neutral on explicit measures. Their results suggested that the IAT might resist self-presentational forces which can mask personally or socially undesirable racial attitudes. In the current study, two experiments tested whether participants could voluntarily suppress the tendency to appear (1) more favorable to flowers than to insects on the IAT of those attitudes, or (2) pro-white on the racial IAT of whites and blacks. Both experiments found that participants could not fake the IAT effectively when merely asked to do so; they could produce a faked implicit attitude only when they were instructed to respond slowly to a subset of the stimuli. Overall, participants did not spontaneously discover the apparently controllable strategy for faking the IAT; they had to be taught how to implement it.