Atypically heterogeneous vertical first fixations to faces in a case series of people with developmental prosopagnosia

Abstract

When people recognise faces, they normally move their eyes so that their first fixation is in the optimal location for efficient perceptual processing. This location is found just below the centrepoint between the eyes. This type of attentional bias could be partly innate, but also an inevitable developmental process that aids our ability to recognise faces. We investigated whether a group of people with developmental prosopagnosia would also demonstrate neurotypical first fixation locations when recognising faces during an eye tracking task. We found evidence that adults with prosopagnosia had atypically heterogeneous first fixations in comparison to controls. However, differences were limited to the vertical, but not horizontal, plane of the face. We interpret these findings by suggesting that subtle changes to face-based eye movement patterns in developmental prosopagnosia may underpin their face recognition impairments, and suggest future work is still needed to address this possibility.

Face processing predicts reading ability: Evidence from prosopagnosia

“…Thus atypicalities in the WLE in prosopagnosia may support this expertise hypothesis. Burns, Tree, Chan, & Xu, 2017;Bate et al, 2019;Childs et al, 2021), and is commonly used to diagnose DP (Bate et al, 2014;Bate & Tree, 2016;Biotti & Cook, 2016;Biotti, Gray, & Cook, 2017;Burns, Martin, Chan, & Xu, 2017;Burns, Tree, & Weidemann, 2014;Wilcockson, Burns, Xia, Tree, & Crawford, 2020). Using norms from the original CFMT paper (Duchaine & Nakayama, 2006) and one-tailed Crawford's single case ttests (Crawford & Howell, 1998), we confirmed all 28 DP cases were abnormally impaired in face memory ability.…”

A new approach to diagnosing and researching developmental prosopagnosia: Excluded cases are impaired too

“…Low power is not just an issue when researchers are using a single test, such as the CFMT, to diagnose prosopagnosia. Many authors require significant impairments at the single-case level on two (e.g., CFMT and FFT, Fisher et al, 2020 ; Pertzov et al, 2020 ; Stumps et al, 2020 ; Wilcockson et al, 2020 ) and sometimes three (CFMT, CFPT and a prosopagnosia questionnaire, Mishra et al, 2020 ) measures before they will confirm a diagnosis. We know that roughly 50% of potential DP cases will fail to meet diagnostic criteria when assessed using the CFMT (Bate et al, 2019a ; Burns et al, 2014a ; Murray & Bate, 2020 ).…”

“…If researchers need impairment on one specific task (e.g., the CFMT: Gerlach et al, 2020 ), then they will be able to include 50% of potential DP cases in their studies, as on average, 50% of participants are sufficiently impaired at the single-case level to attain a diagnosis via the CFMT. However, if researchers need impairment on two specific tasks (e.g., the CFMT and FFT, Stumps et al, 2020 ; Wilcockson et al, 2020 ), or three (e.g., Fry et al, 2020 ), then these authors will test a much smaller sample of the most severely impaired cases who can meet all these criteria.…”

Looking at faces in the wild

“…For example, many lab-based studies have found associations between face-processing ability and face information sampling patterns (e.g. 4,[66][67][68][69][70][71] for a review see 72 ). Relatedly, developmental disorders have been associated with abonomal social attention in laboratory-based studies (e.g.…”