Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Supporting face familiarization using perceptual and engineering frameworks

Jones, Scott Phillip 2013. Supporting face familiarization using perceptual and engineering frameworks. PhD Thesis, Cardiff University.
Item availability restricted.

[thumbnail of jones.pdf] PDF - Supplemental Material
Restricted to Repository staff only

Download (374kB)
[thumbnail of 2013jonespdf.pdf]
Preview
PDF - Accepted Post-Print Version
Download (1MB) | Preview

Abstract

The identification of unfamiliar faces is known to be inferior to the recognition of faces with which we are familiar. This can lead to undesirable consequences such as misidentification. However, there is some evidence to suggest that a brief period of familiarisation can dramatically improve our ability to recognise an unfamiliar individual. Chapter 1 outlines the previous research that has aimed to understand the mechanisms of face processing, and to improve the recognition of unfamiliar faces. Three areas that require further investigation are identified and the experimental work reported in the three empirical chapters addresses these issues. Chapter 2 reports five experiments, using photographs of faces as stimuli, which examined whether a short training exposure promoting stimulus comparison can facilitate recognition of unfamiliar faces (c.f. Dwyer & Vladeanu, 2009). The results revealed that, contrary to expectation, any beneficial effects of comparison do not extend to improving discrimination between targets and nonexposed stimuli. The results of Chapter 2 required a return to the mechanisms of perceptual learning thought to underpin the comparison effect. Numerous attempts to unpack this process have relied on experiments that have examined the content, but not the location, of the unique features of a stimulus (e.g., Hall, 2003; Mitchell, Nash, & Hall, 2008; Mundy, Honey, & Dwyer, 2007). Chapter 3 used checkerboards as stimuli, manipulating the placement of the unique feature, as a way of breaking the perfect correlation between content and location and assess their relative contributions to perceptual learning. The findings indicated that discrimination between similar stimuli on the basis of exposure can be explained entirely by learning where to look, with no independent effect of learning about particular stimulus features. 11 Chapter 4 returned to the issue of potential methods to improve recognition, and examined the possibility that training using synthesised faces created from a single view and presented at multiple yaw rotations can aid face recognition (Liu, Chai, Shan, Honma, & Osada, 2009). The findings of three experiments; strengthen the claim that identifying an individual can be improved using multiple synthesised views generated from a single front view of a face, and suggest that this improvement may be affected by the quality of synthesised material. In summary, while the results reported within this thesis indicate that comparison between similar faces does not produce an effective way of supporting the recognition of unfamiliar faces, they do indicate that experience with a face and/or artificial faces may be a practical means of facilitating identification.

Item Type: Thesis (PhD)
Status: Unpublished
Schools: Psychology
Subjects: B Philosophy. Psychology. Religion > BF Psychology
Funders: EPSRC
Date of First Compliant Deposit: 30 March 2016
Last Modified: 19 Mar 2016 23:20
URI: https://orca.cardiff.ac.uk/id/eprint/47952

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

View more statistics