Self, Matthew William; (2003) A study of integration and binding in the human visual brain. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text A_study_of_integration_and_bin.pdf Download (15MB) | Preview

Abstract

This thesis is concerned with the visual system's ability to integrate various components of the visual scene into a unified percept within a fraction of a second. It takes as its starting point the theory of functional specialisation in the human visual system and attempts to answer the question raised by this theory. If the visual system is divided up into specialised modules that process separate parts of the visual world (i.e. colour, motion, depth etc), then how does the brain integrate these various components. Firstly, using psychophysical techniques, we addressed whether the brain is able to integrate the colour and motion cues that define objects. We found that objects defined by more than one cue are more easily and rapidly discriminated than those defined by one cue alone, suggesting that the brain can integrate different cues. Using functional magnetic resonance imaging (fMRI) we searched for the cortical site(s) for this process. We found that the posterior part of the lateral occipital complex (LO) was more active when subjects viewed objects defined by both colour and motion cues compared to those defined by either alone. LO also had an adaptation profile consistent with the presence of neurons responsive to both colour and motion defined objects. This suggests that LO is the site of the integrative process evident from the psychophysical studies. Generalising this finding to face stimuli, we found that the site of integration for faces was not LO, but the area specialised for processing faces (the fusiform face area). Taking all results together, we propose that cue integration is a rapid, feed-forward process that is executed by the area specialised for processing the relevant stimulus and is crucial to the rapid and efficient recognition of visual objects by the human visual system.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item