A person-by-situation approach to predicting behavior with implicit measures

点击翻译按钮获取中文摘要

The ability to accurately predict human behavior is a signature challenge of social science. Attitudes encapsulate specific thoughts and feelings that presumably should predict behavior. However, often people are unwilling or unable to provide accurate reports of their attitudes, which frequently is the case when their attitudes reflect socially controversial beliefs. It has now been roughly two decades since the invention and innovation of a class of attitude measures termed "implicit measures", which were designed to access mental content that a person is either unwilling or unable to report. This important methodological advance has had an immense impact on both scientific theory and public discourse. However, the evidence is mixed with respect to the utility of these measures in predicting behavior. The current project reframes decades of research to suggest that the mental content picked up by these measures varies in predictive ability 1) across people who differ from one another in important ways, and 2) as those people find themselves in a variety of different situations. The appropriate question may not be "Do implicit measures predict behavior?". Rather they are "Do implicit measures predict behavior better for some people rather than others and in some situations more than in others?" and "Do these two levels of inquiry interact so that prediction works particularly well for some people in some situations?"The project provides the most comprehensive test to date of factors affecting the utility of implicit measures. The 12 studies test predictions using three implicit measures across five attitude domains and examine eight individual differences within four different eliciting contexts. Specific methods test person factors (attitudinal accessibility, consistency) and situational factors (activation, control). The breadth of inquiry and large samples are essential to make confident and generalizable claims. The findings of this research have critical implications regarding the validity of attitudes that are difficult to measure, including those that contribute to the adverse treatment of others. Only by accurately assessing attitudes across people and contexts can their downstream consequences on societal problems be better understood and addressed.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.