Robust associations between the 20-item prosopagnosia index and the Cambridge Face Memory Test in the general population

Abstract

Developmental prosopagnosia (DP) is a neurodevelopmental condition, characterized by lifelong face recognition deficits. Leading research groups diagnose the condition using complementary computer-based tasks and self-report measures. In an attempt to standardize the reporting of self-report evidence, we recently developed the 20-item prosopagnosia index (PI20), a short questionnaire measure of prosopagnosic traits suitable for screening adult samples for DP. Strong correlations between scores on the PI20 and performance on the Cambridge Face Memory Test (CFMT) appeared to confirm that individuals possess sufficient insight into their face recognition ability to complete a self-report measure of prosopagnosic traits. However, the extent to which people have insight into their face recognition abilities remains contentious. A lingering concern is that feedback from formal testing, received prior to administration of the PI20, may have augmented the self-insight of some respondents in the original validation study. To determine whether the significant correlation with the CFMT was an artefact of previously delivered feedback, we sought to replicate the validation study in individuals with no history of formal testing. We report highly significant correlations in two independent samples drawn from the general population, confirming: (i) that a significant relationship exists between PI20 scores and performance on the CFMT, and (ii) that this is not dependent on the inclusion of individuals who have previously received feedback. These findings support the view that people have sufficient insight into their face recognition abilities to complete a self-report measure of prosopagnosic traits.

Self-ratings of face recognition ability are influenced by gender but not prosopagnosia severity.

“…These factors potentially explain why some individuals who meet Self-ratings of prosopagnosia severity 10 objective criteria for DP return typical scores on self-report measures (e.g. Bowles et al, 2009;Gray et al, 2017;Palermo et al, 2017). Although this did not occur in the current study (all 47 DPs scored above the recommended cut-off on the PI20), it is important to note that they had all received feedback on their face recognition skills (i.e.…”

Searching for faces in crowd chokepoint videos

“…The relationship between confidence and performance is not well understood. Evidence has shown that people's self-rated abilities with face matching and memory may be predictive of their actual abilities, although this association may only be modest (Bate & Dudfield, 2019;Bobak, Mileva, & Hancock, 2019;Bobak, Pampoulov, & Bate, 2016;Gray, Bird, & Cook, 2017;Palermo et al, 2017;Shah, Sowden, Gaule, Catmur, & Bird, 2015). However, research has shown that confidence at the level of individual responses does appear to reflect accuracy.…”

Normal composite face effects in developmental prosopagnosia

“…None of the DPs had a history of brain injury or psychiatric disorder (e.g., Schizophrenia, Autism Spectrum Disorder). Diagnostic evidence for the presence of DP was collected using the Cambridge Face Memory Test (CFMT; Duchaine & Nakayama, 2006a) the Twenty-Item Prosopagnosia Index (PI20; Gray, Bird, & Cook, 2017;Shah, Gaule, Sowden, Bird, & Cook, 2015), and a Famous Face Test suitable for use with UK residents (FFT UK ). Scores on the CFMT were compared against data from 50 typical observers reported by Duchaine & Nakayama (2006a).…”