Psychophysical profiles in super-recognizers

Abstract

Facial identity matching ability varies widely, ranging from prosopagnosic individuals (who exhibit profound impairments in face cognition/processing) to so-called super-recognizers (SRs), possessing exceptional capacities. Yet, despite the often consequential nature of face matching decisions-such as identity verification in security critical settings-ability assessments tendentially rely on simple performance metrics on a handful of heterogeneously related subprocesses, or in some cases only a single measured subprocess. Unfortunately, methodologies of this ilk leave contributions of stimulus information to observed variations in ability largely un(der)specified. Moreover, they are inadequate for addressing the qualitative or quantitative nature of differences between SRs' abilities and those of the general population. Here, therefore, we sought to investigate individual differences-among SRs identified using a novel conservative diagnostic framework, and neurotypical controls-by systematically varying retinal availability, bandwidth, and orientation of faces' spatial frequency content in two face matching experiments. Psychophysical evaluations of these parameters' contributions to ability reveal that SRs more consistently exploit the same spatial frequency information, rather than suggesting qualitatively different profiles between control observers and SRs. These findings stress the importance of optimizing procedures for SR identification, for example by including measures quantifying the consistency of individuals' behavior.

Figure 1

Examples of stimuli used in Experiments 1 and 2. (a) In Experiment 1 image size varied logarithmically from 512 pixels in width/height (images 1 and 1′) to 8 pixels (images 7 and 7′). The top row of a. shows the effect of the laplacian pyramid on SF con tent, and the bottom row displays the actual stimulus size. (b) In Experiment 2, images were bandpass filtered to preserve horizontal (top row of (b)) or vertical (bottom row of (b)) information in 15° steps from 0° to 180° (with every second step shown here). (b) was reproduced from Pachai et al..

Figure 2

Logistic function fits to trial-level correct responding as a function of filter orientation in Experiment 2. Grand averaged data (left) are plotted alongside Model 2’s fit, with gray tracings displaying individual observers’ fits for comparison. Model 3’s fits to the orientation-split averaged data are plotted separately for images retaining horizontal (middle) and vertical (right) SF information, across filter bandwidth conditions. Error bars represent ± 1SEM; dotted lines represent chance performance.

Figure 3

Logistic approximations of the psychometric function fitted to trial-level correct responding, accounting for both Group and Orientation as factors. Averaged data are grouped by column (control observers, left column; SRs, right column), with gray tracings in each displaying individual observers’ model fits for comparison. Averaged data were further split by filter orientation, and are plotted separately for images retaining horizontal (top row) and vertical (bottom row) SF information, across filter bandwidth conditions. Error bars represent ± 1SEM; dotted lines represent chance performance.

Figure 4

Logistic fits correct responding accounting for Group. Grand averaged data (left) from Model 3 are plotted alongside Model 5’s fits (with gray tracings displaying individual fits to observers’ data) to control observers (middle) and SRs (right) for comparison. Error bars represent ± 1SEM; dotted lines represent chance performance.

Figure 5

Logistic approximation of the psychometric function fitted to trial-level correct responding for Experiment 1. Grand averaged data (left) are plotted alongside Model 2’s fit, with gray tracings displaying individual observers’ model fits for comparison. Model 3’s fits to the group-averaged data are plotted separately for control observers (middle) and SRs (right) across image size conditions. Error bars represent ± 1SEM; dotted horizontal lines represent chance performance.