Most humans are experts in recognizing faces of familiar individuals, but are poor at individuating unfamiliar faces. The discrepancy between these two types of face recognition suggests qualitative differences in the perceptual encoding and memory storage of familiar and unfamiliar faces, yet little is known about the neural basis of these differences. In the present thesis, behavioural and event-related brain potential (ERP) measures were combined to investigate the mechanisms that underlie the perception and recognition of unfamiliar faces. The first two experiments investigated whether memory traces for unfamiliar faces are based on low-level view-dependent or more high-level view-independent codes. Results provide strong evidence for a qualitative change from strictly view-dependent to view-independent representations in visual face memory as initially novel unfamiliar faces become more familiar. A second series of three experiments examined whether identity-specific cues are processed in an optional or obligatory fashion. Findings suggest that the perception of facial identity is strongly task-dependent (i.e., optional), even for famous faces, but can also show a degree of mandatory processing when identity is task-irrelevant. The sixth study examined the persistence of perceptual memories of unfamiliar faces and revealed a substantial weakening of face representations in visual working memory over short periods of time. The final experiment investigated the neural basis of developmental prosopagnosia (DP). Results demonstrated spared identity-sensitive processing in DP participants, indicating that their face recognition deficits do not always result from severely disrupted visual face recognition processes. However, despite evidence for perceptual learning of invariant aspects of face structure, these processes seemed to be inefficient in individuals with DP. Taken together, this thesis explored how we perceive and remember individual unfamiliar faces. Results indicate that unfamiliar face recognition is mediated by fast and flexible (i.e., strongly task-dependent) identity-specific visual processes, which rapidly become view-invariant during face learning.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:589670 |
| Date | January 2013 |
| Creators | Zimmermann, Friederike Gisela Sophie |
| Publisher | Birkbeck (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://bbktheses.da.ulcc.ac.uk/36/ |
Page generated in 0.0223 seconds