How not to enhance the confidence-accuracy relation: The detrimental effects of attention to the identification process

This study (N = 218) examines potential moderators of the confidence-accuracy (CA) relation in eyewitness identifications The design included five experimental interventions (accountability, context reinstatement, retrospective narration, hypothesis disconfirmation, and hypothesis generation), as well as one trait measure (public self-consciousness). Although the interventions were hypothesized to enhance the CA relation relative to a control condition, they did not do so. In fact, quite contrary to our initial expectations, only participants in the control condition displayed unqualified insight into their identification accuracy. We conclude that attempts to enhance awareness of the thoughts and reasoning process involved in an identification decision may frequently have minimal, or even counterproductive, effects. Instead, results of decision process measures suggest that the CA relation might be more effectively enhanced by increasing the salience of relatively automatic decision criteria (e.g., "did his face 'pop out' at me?") and decreasing the salience of algorithmic identification strategies (e.g., "was I thorough in making my decision?").