Individual differences in face and voice matching abilities: The relationship between accuracy and consistency

Abstract

Deciding whether two different face photographs or voice samples are from the same person represent fundamental challenges within applied settings. To date, most research has focussed on average performance in these tests, failing to consider individual differences and withinperson consistency in responses. Here, participants completed the same face (Experiment 1) or voice matching test (Experiment 2) on two separate occasions, allowing comparison of overall accuracy across the two timepoints as well as consistency in trial-level responses. In both experiments, participants were highly consistent in their performances. In addition, we demonstrated a large association between consistency and accuracy, with the most accurate participants also tending to be the most consistent. This is an important result for applied settings in which organisational groups of super-matchers are deployed in real-world contexts. Being able to reliably identify these high performers based upon only a single test informs regarding recruitment for law enforcement agencies worldwide.

Psychophysical profiles in super-recognizers

“…Given the paucity of psychophysical studies of SRs’ abilities, we propose that future studies adopt such systematic approaches to provide a more detailed description of their unique skill(s). While we find consistency to be predictive of superior ability (as also reported recently in typical cohorts 47 ), we would advocate deploying measures of consistency in addition to more traditional performance metrics and not as a substitute for them. That is, they ought to be integrated to extend the granularity of current 8 multifactorial frameworks of SR diagnostic tests, as originally proposed 6 .…”

Wisdom of the inner crowd benefits both face and voice matching

“…Previous research has shown that inner crowd benefits tended to be greater with a delay between estimates (Steegen et al, 2014; van Dolder & van den Assem, 2018; Vul & Pashler, 2008). By analysing data from the current experiments, along with a reanalysis of previously collected data (Kramer et al, 2021), we found that inner crowds performed better than individual responses both without a delay and with a one‐week delay. While the sample sizes involved here did not allow for a direct comparison between these conditions (which would require 64 participants in each sample to detect a medium‐sized effect with 80% power), it may be that the length of the matching tests used (40 face trials and 80 voice trials) prevented a delay from providing additional benefits.…”

“…Rather than making explicit, binary judgements about whether face pairs were the ‘same’ or ‘different’ people, participants rated the likelihood that the two images were of the same person (e.g., Kramer et al, 2021; O'Toole et al, 2007). This approach meant that representational and decisional components could be separated, with the focus being placed on the former.…”

“…Participants did not receive any feedback during or after each test, and so there is perhaps no reason to predict an improvement across sessions. Previous research has shown a high level of consistency in responses when repeating the same test twice (Kramer et al, 2021). This could be explained by participants either remembering and reproducing their responses, or simply giving the same responses when faced with the same stimuli.…”

“…On each trial, two face photographs were displayed onscreen and participants were instructed to decide whether they thought these faces were the same person or two different people. Following previous research (Kramer et al, 2021; O'Toole et al, 2007), responses were provided using a labelled rating scale: (1) sure they are the same; (2) think they are the same; (3) don't know; (4) think they are not the same; (5) sure they are not the same. Trial order was randomised for each participant, no time limits were imposed upon responses, and no feedback was given at any stage.…”