Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.

Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

A Trait Approach to Eyewitness Research: Do Stable Accurate and Inaccurate Eyewitnesses Exist and Can They be Postdictively Identified?

Jessica L Snowden, University of Nebraska - Lincoln


Eyewitness errors are a leading cause of erroneous convictions. This three-session study explored the stability of eyewitness performance, both within and across eyewitness identification and event memory reporting tasks for five videos, and measured individual differences previously linked with eyewitness performance. The four witness performance measures included lineup accuracy, independent item event accuracy, consistent information item event accuracy, and suggestibility (susceptibility to misleading information items). Ultimately, there was some support, albeit mixed, for stable witness performance, both within and across tasks. Consistent with hypothesized witness stability, the performance measures were significantly correlated with each other, except that consistent information accuracy and suggestibility were not related and that individual lineup decisions were not correlated with one another. Within witness tasks, a minority of participants consistently displayed poor performance and a small proportion of participants performed consistently well, although no participant got all lineups correct. Despite significant correlations among cross-task witness performance measures, very few participants performed consistently across both the identification task and all three event reporting task measures. Individual difference variables overall displayed minimal predictive utility for witness performance in binomial regression models relative to the other witness performance measures themselves. Future research, for example, employing a substantially larger sample and with videos with longer exposure times for different types of crimes, might afford greater opportunity to determine whether the relatively low rates of stable witness performance from the present study are replicable, as well as whether witness performance levels might be postdicted. In the interim, these low rates of stable performance support great caution in relying on uncorroborated eyewitness testimony.

Subject Area

Social psychology

Recommended Citation

Snowden, Jessica L, "A Trait Approach to Eyewitness Research: Do Stable Accurate and Inaccurate Eyewitnesses Exist and Can They be Postdictively Identified?" (2017). ETD collection for University of Nebraska-Lincoln. AAI10270162.