Skip to main content

Face Perception and Recognition, on the Fringe of Human Awareness.

Hajilou, Omid (2020) Face Perception and Recognition, on the Fringe of Human Awareness. Doctor of Philosophy (PhD) thesis, University of Kent,. (doi:10.22024/UniKent/01.02.85577) (KAR id:85577)

PDF
Language: English


Download (8MB) Preview
[thumbnail of 94Omid_Hajilou_-_PhD_Thesis_2020.pdf]
Preview
This file may not be suitable for users of assistive technology.
Request an accessible format
Official URL
https://doi.org/10.22024/UniKent/01.02.85577

Abstract

In Rapid Serial Visual Presentation (RSVP), we can present a large volume of information, on the fringe of awareness, whilst observing the brain's electrical signals using an Electroencephalogram (EEG). The vast majority of stimuli are not consciously perceived, but the salient ones breakthrough into awareness, enter into working memory and can be reported by the participant (Bowman, et al., 2013). Deception detection studies have successfully employed this countermeasure resistant fringe-P3 method, using letters, numbers and words (Bowman, Filetti, Alsufyani, Janssen, & Su, 2014), to differentiate between familiar and unfamiliar information. The inclusion of faces, and their application in Concealed Information Tests (CIT) have yet to be fully explored. In this thesis, we hypothesised that the fringe-P3 method could be successfully used to detect intrinsic salience of familiar faces, even when there was no task associated with the stimuli. Using experiments, we investigated the sensitivity of the ERP-based RSVP paradigm, to infer recognition of celebrity, as well as, lecturer faces, and performed statistical tests in the Time and Frequency domains, to differentiate between known and unknown faces, at group and subject levels. Furthermore, we used ground-truth data simulations to explore the viability of using online statistical tests, to focus experimental data collection efforts, on the critical stimulus with the highest significance, in order to improve statistical power (i.e. reduce the risk of Type II errors), without the inflation of Type I errors. As a result, we introduced new methods of analysis, and a two-part experimental design, where Part II's parameters are independently influenced by Part I's results, using online statistical tests. Finally, we applied our new findings in a concluding experiment, which explored a real-life scenario of revealing participants' familiarity with their lecturers, through the data captured from their brain. Our findings provide evidence that familiar faces are differentially perceived and processed by participants' brains, as compared to novel (unfamiliar) faces. Therefore, we propose our final experiment to be a workable solution for deception detection applications of crime compatriots (e.g. accomplices), using faces in RSVP-based EEG tests.

Item Type: Thesis (Doctor of Philosophy (PhD))
Thesis advisor: Bowman, Howard
DOI/Identification number: 10.22024/UniKent/01.02.85577
Uncontrolled keywords: RSVP, EEG, ERP, Deception Detection, Face recognition, Concealed Information Test, fringe-P3.
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Computing
SWORD Depositor: System Moodle
Depositing User: System Moodle
Date Deposited: 19 Jan 2021 11:10 UTC
Last Modified: 19 May 2021 13:54 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/85577 (The current URI for this page, for reference purposes)
  • Depositors only (login required):