Faces are highly complex visual stimuli that playa disproportionate role in social interactions. Based on visual information from faces alone, humans can judge an individual's race, gender, age, identity, mood and intentions. The first aim of this thesis is to investigate aspects of the mechanisms that underlie the extraordinary sensitivity of human face perception. In the first study, sensitivity to various face features was quantified. Evidence was found of enhanced sensitivity to external (head-shape, hairline), relative to internal (nose, mouth, eyes, eyebrows) features, with the lowest sensitivity for features most affected by face dynamics. Discrimination thresholds were significantly higher when features were embedded within an average face context, relative to isolated presentation. Although inversion significantly reduced sensitivity, the external feature advantage remained. These results have implications on the nature of holistic processing mechanisms for unfamiliar faces. In a different study, face perception was investigated in peripheral vision. Discrimination thresholds for full-faces, internal features and external features were lower in the left visual field, relative to all other visual field locations, consistent with the cortical lateralisation of the face-sensitive fusiform face area. Sensitivity to headshapes, on the other hand, was greatest in the lower visual field. This is consistent with the premise of non-overlapping processing mechanisms for the general shape of objects and specific object categories, such as faces. The aim of a further study was to investigate the neuronal representation of faces by determining the effect of adaptation on face discrimination. Face adaptation significantly reduced unfamiliar face discrimination sensitivity. This effect was, however, highly identity-specific; adaptation did not transfer between different face identities. The magnitude of this adaptation effect increased monotonically with adapting face distinctiveness (how much an individual face differs from an average/mean face), but only up to a limit. These results suggest that individual face identities are encoded by discrete neuronal populations that respond with increasing magnitude as faces become progressively more different from an average face but exclude more extreme face caricatures. Based on these results, the final part of this thesis outlined the development of a new clinical test of face perception. This new test is designed to overcome the limitations of existing face tests such as restricted testing ranges and confounding factors (e.g. memory, familiarity). The range of normal face sensitivity was determined in a large group of young adults. Older adults showed preserved face discrimination sensitivity. A case study of a patient with suspected developmental prosopagnosia demonstrates that the test has the potential to accurately identify impairments of face perception enjoying a substantially higher sensitivity than standard face tests.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:676476 |
| Date | January 2014 |
| Creators | Logan, Andrew J. |
| Publisher | Glasgow Caledonian University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0123 seconds