Hulme, OJ; (2006) Visibility, invisibility and reportability. Doctoral thesis , UCL (University College London).

Preview

Text Hulme_thesis.Redactec.pdf Download (35MB) | Preview

Abstract

This thesis attempts to investigate the relationship between neural activity and visual consciousness. It attacks two problems central to this endeavour. The first problem is that of visibility and invisibility. This is addressed by looking at the neural determinants of whether something can be seen (or not). The relationship between neural activity and object perception has received considerable attention using sophisticated stimulus manipulations such as masking and dichoptic presentation. Here we investigated this relationship simply by occluding objects with an opaque screen that dissociated the direct perception of the object from the awareness of its presence. We used fMRI to measure brain activity when subjects viewed objects (faces and houses) that were occluded and found that the response of object selective areas is invariant to whether the object is visible or occluded. This suggests that when objects are directly viewed, the response of object-selective regions is due to the awareness of the object's presence, not just the direct perception of it. The second problem is that of reportability. This is addressed by looking at the neural determinants of how visual stimuli are reported. Understanding the neural underpinnings of reportability is critical since experiments on consciousness invariably employ reportability as an 'index' of consciousness. This involves operationally defining whether the subject was conscious of something by simply asking 'did you see it'. It can be argued that this operationalisation conflates two processes perception and its reportability whose neural underpinnings are potentially dissociable. In experiment 2 I used a 'partial report methodology' to dissociate the neural correlates of these two processes in an fMRI study of object detection. The task involved presenting a visual array of objects, and then presenting a spatial cue after a variable delay. Using a signal detection theoretic approach we were able to decouple 4 stages of stimulus reportability, stimulus processing, attentional selection, perceptual decision and motoric report. We show that retinotopically specific responses in early visual cortex reflect stimulus processing, a network of parietal and temporal regions reflects perceptual decisions, fronto-parietal-pulvinar networks reflect attentional selection and motor regions reflect report. These findings provide a tentative neural framework with which to evaluate the relationship between 'consciousness' and reportability. By consideration of both findings I attempt to synthesise a speculative theoretical perspective that unites the neural correlates of visibility, invisibility and reportability into a consistent framework.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item