Testing a ''not sure'' instruction as a prophylactic against the harmful impact of system and estimator variables on lineup identification accuracy

The proposed project tests the effectiveness of eyewitness-identification lineup instructions that provide an explicit 'not sure' instruction to witnesses before viewing the lineup.

One of the ultimate goals of eyewitness researchers is to develop procedures that (a) minimize false identifications caused by system and estimator variables (while minimizing any decrease in correct identifications), and (b) require few resources for law enforcement to enact. The specific objective of the proposed research is to test the effectiveness of this ‘not sure’ instruction in suppressing the false identification rate associated with a system variable (the appearance of the perpetrator and changes in eyewitness identification instructions).

A series of randomized controlled experiments will involve Black and Hispanic community members watching mock crime videos and being randomly assigned (a) to receive either the ‘not sure’ instruction or not, (b) to receive the appearance- change of the perpetrator in each lineup with an instruction or not, or (c) to view a target-present or target-absent lineups.

Subsequent experiments will involve community members, either Black or Hispanic watching a mock crime video and being randomly assigned (a) to receive a ‘not sure’ instruction or not, (b) to view either a same-race or cross-race perpetrator, or (c) to view either a target-present or target-absent lineup. Data will be analyzed using Receiver Operating Characteristic (ROC) analysis, a statistical technique only recently introduced into the eyewitness identification field that has never been applied to the effectiveness of a ‘not sure’ instruction.

"Note: This project contains a research and/or development component, as defined in applicable law," and complies with Part 200 Uniform Requirements - 2 CFR 200.210(a)(14).

CA/NCF