Date of Graduation

12-2016

Document Type

Thesis

Degree Name

Master of Arts in Psychology (MA)

Degree Level

Graduate

Department

Psychological Science

Advisor

James M. Lampinen

Committee Member

Denise Beike

Second Committee Member

Bill Levine

Abstract

It is necessary to better serve justice to understand the mechanisms behind eyewitness identification and reports of confidence. The material contained within attempt to fit eyewitness identification to a diffusion model of processing, RTCON (Ratcliff & Starns, 2009). Participants saw eight mock crime videos and were then tasked with using eight showups or eight lineups to identify the suspects within the video. Half of the presentations were target present and half were target absent. Additionally, participants were either presented with biased or unbiased instructions. Strangely, unbiased lineups led to higher hit rates which is contrary to most findings in the field. The key elements were comparing the ROC curves of the collected data with the ROC curves of simulated data from the RTCON model. The variables manipulated in the model included mu, a variable of memory strength; confidence criteria, a variable of response bias; and decision criteria, a second variable of response bias. The ROC curves derived from the simulated data were poor matches to the collected data. This finding leads to questions on the applicability of this model of diffusion processing to eyewitness research.

Share

COinS