In this study, researchers testing a 'not sure' instruction to reduce the harmful impact of system and estimator variables on lineup identification accuracy.
The current project tested the effectiveness of providing witnesses with an explicit 'not sure' instruction before viewing a lineup and found that while a ‘not sure’ instruction may act as a prophylactic against the harmful effects of system and estimator variables known to inflate false identifications, the ‘not sure’ option failed to improve eyewitness decision-making overall, and failed to act as a prophylactic against the harmful effects of the appearance-change instruction or of cross-race identifications. The main two studies examined whether the ‘not sure’ option would restore any loss in discriminability associated with a system variable. Results indicated that although witnesses exhibited lower discriminability when making cross-race (cf. same-race) identifications, the ‘not sure’ option failed to restore that discriminability. Although witnesses used the ‘not sure’ option when their memories were poor, as expected, providing witnesses with the ‘not sure’ option did not improve empirical eyewitness discriminability. In addition, although the appearance-change instruction increased false identifications, as expected based on past research, the ‘not sure’ option failed to protect against that harmful effect, contrary to hypotheses. The specific objective of the proposed research was to test the effectiveness of this ‘not sure’ instruction. The pilot study (n = 322) examined whether variations in the administration of the ‘not sure’ instruction (i.e., whether it was described as a ‘not sure’ option or a ‘don’t know’ option; whether the option was accompanied by a corresponding pre-lineup instruction regarding the option or not) affected identification decision outcomes; results indicated that the instruction was equally effective across these variations.