Use this URL to cite or link to this record in EThOS:
Title: Decoding high level influences on facial expression recognition
Author: Adams, Vicky Samantha
ISNI:       0000 0004 7962 7438
Awarding Body: University of East Anglia
Current Institution: University of East Anglia
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
The present thesis explores the neural mechanisms underlying the recognition of emotion; the effect of high-level influences such as prior knowledge, task goals and the possible contribution of embodied simulation in facial expression recognition. The initial experiments (Chapters 2 & 3) investigate high-level processing that occurs when facial features are occluded in the recognition of facial expressions (visual route of recognition). This research examines the information about occluded facial features in early visual (V1-V3), face and emotion sensitive areas with fMRI, as well as the temporal dynamics of posterior brain regions in processing occluded facial features with EEG. MVPA reveals similar patterns of decoding across nonoverlapping samples of face information, suggesting the involvement of contextual influences beyond low-level processing (Chapter 2), as well as reliable decoding of facial expression (happy, fear and disgust) in conditions missing feature information (Chapters 2 & 3). This decoding, found 50-700ms, has three decoding phases, which potentially reveal the presence of feedforward and feedback processes (Chapter 3). These chapters also investigate the influence of task constraints, finding decoding differences between implicit and explicit processing conditions. Overall, this research helps understand how the brain deals with occluded stimuli; in keeping with accounts implying the rich role for top-down influences, such as predictive coding. The following experiment (Chapter 4) investigates embodiment in emotion recognition with fMRI; exploring shared representation in the perception and production of facial expression. MVPA reveals reliable expression decoding in the premotor brain regions across perception and production, demonstrating representational overlap across the sensory perception and motor production of expression. This tentatively supports strongly embodied simulation-based (nonvisual) theories. Collectively, the present research contributes to our knowledge of high-level influences in facial expression recognition, supporting the involvement of visual and non-visual routes to recognition, as well as providing further directions for future research.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available