Lineup fairness: propitious heterogeneity and the diagnostic feature-detection hypothesis

Abstract

Researchers have argued that simultaneous lineups should follow the principle of propitious heterogeneity, based on the idea that if the fillers are too similar to the perpetrator even an eyewitness with a good memory could fail to correctly identify him. A similar prediction can be derived from the diagnostic feature-detection (DFD) hypothesis, such that discriminability will decrease if too few features are present that can distinguish between innocent and guilty suspects. Our first experiment tested these predictions by controlling similarity with artificial faces, and our second experiment utilized a more ecologically valid eyewitness identification paradigm. Our results support propitious heterogeneity and the DFD hypothesis by showing that: 1) as the facial features in lineups become increasingly homogenous, empirical discriminability decreases; and 2) lineups with description-matched fillers generally yield higher empirical discriminability than those with suspect-matched fillers.

Keywords: Diagnostic feature-detection hypothesis; Eyewitness identification; Lineup fairness; Propitious heterogeneity; Simultaneous lineup.

Fig. 1

Example lineups from Experiment 1 composed of facial stimuli from FACES 4.0. Only the eyes vary in the top left, the eyes and nose vary in the top right, and eyes, nose, and mouth vary in the bottom

Fig. 2

ROC data from Experiment 1. The curves drawn through the empirical data points are not based on model fits, but rather are simple trendlines drawn in Excel. The correct ID rate on the y axis is the proportion of targets chosen from the total number of target-present lineups in a given condition. The false ID rate on the x axis is the proportion of all filler identifications from the total number of target-absent lineups in a given condition (as we had no designated innocent suspects), divided by the nominal lineup size (six) to provide an estimated innocent suspect ID rate

Fig. 3

ROC data (with trendlines) from Experiment 2 collapsed over the different description-matched and suspect-matched lineups. The false ID rate on the x axis is the proportion of innocent suspect identifications from the total number of target-absent lineups in a given condition

Fig. 4

ROC data (with trendlines) for all description-matched and suspect-matched lineups from Experiment 2

Fig. 5

CAC data from Experiment 2. The bars represent standard errors. Proportion correct on the y axis is #correct IDs/(#correct IDs + #false IDs)