There are several issues involved in the effort to relate laboratory performance tasks to questionnaire-based trait measures. First, personality traits and the behaviors measured by laboratory tasks are far from isomorphic. The former refer to stable characteristic individual differences in ways of perceiving the world and responding to it; in all likelihood, traits reflect combinations of affective and cognitive processes. In contrast, laboratory tasks typically refer to relatively specific cognitive processes. Thus, it is not clear that measures of the two types of processes should necessarily relate strongly. Along these lines, Reynolds et al. (2006) found that self-report measures of impulsivity did not correlate strongly with behavioral tasks assessing aspects of impulsive behavior.
