Variability in face recognition ability: insights from social motivation, early perception and neural correlates

Papasavva, Michael (2022) Variability in face recognition ability: insights from social motivation, early perception and neural correlates. PhD thesis, Birkbeck, University of London.

Abstract

Face perception is arguably the most important aspect of non-verbal social cognition. Over the last decade, it has become increasingly apparent that between those with prosopagnosia and so called ‘super recognisers’, exists a normally distributed continuum of face recognition ability, within the wider neurotypical population. This thesis explores potential drivers of this variability and associated neural underpinnings. Chapter One investigates the theoretical background of these claims, in addition to discussing the current literature that motivates subsequent empirical Chapters. In Chapter Two, the assumption that socially motivated behaviour, which in turn, mediates increased experience and therefore, expertise with faces, is explored. Across two experiments, a behavioural economics style paradigm is utilised; first, in a large cohort, at the London Science Museum and subsequently repeated (AsPredicted#11359), in a smaller, but highly controlled laboratory setting. Here, the assumption that increased social motivation is associated with better face recognition performance is tested. In Chapter Three, the claim that early perceptual filtering mechanisms are related to face recognition ability is investigated. Here, a breaking Continuous Flash Suppression paradigm is utilised, to explore whether better face recognisers experience preferential access to awareness when compared with worse face recognisers. Findings are then replicated and extended in a second experiment. In Chapter Four, I review recent evidence that suggests that a greater reliance on the eye-region is a predictor of face recognition ability. Across two experiments, I again utilise a behavioural economic style key-pressing paradigm, this time employed to investigate motivated viewing behaviour, for individual face features. Again, a replication (AsPredicted#17150) and extension is attempted using more robust control conditions. In Chapter Five, I review the literature pertaining to possible neural correlates associated with face recognition variability. Here, across two electroencephalography experiments, I combine traditional Event Related Potential techniques (targeting the N170 & P100), with sophisticated multivariate pattern analysis machine learning. In experiment one, classifiers are trained to decode face orientation, independently, in better and worse face recognisers. In a second experiment, I again utilise machine learning to train classifiers to decode between face-parts and corresponding house parts, again, across face recognition groups. Chapter Six summarises the experiments presented in this thesis, identifies common themes and outlines their implications regarding potential contributory factors of face recognition variability as well as potential limitations and future research directions.

Metadata

Statistics

DownloadsShow export options

Export as XMLJSONCSV

Activity Overview

6 month trend

220Downloads

6 month trend

239Hits

Additional statistics are available via IRStats2.

Archive Staff Only (login required)

Edit/View Item