Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Centre for Vision and Visual Cognition

Seminars

'Task-specific cortical activity becomes more stable when its information content becomes conscious'

25th May 2012, 16:30, L47, Dr Aaron Schurger

Aaron did his Ph.D. with Professor  Ann Triesman and Professor Jonathan Cohen in Princeton and he is currently in Professor Stan Dehaene's lab in Paris.

Aaron is hosted by Dr Bob Kentridge (robert.kentridge@durham.ac.uk), therefore get in touch with Bob if you would like to meet and chat with Aaron.



Although the duration and intensity of a neural response are known to be factors in determining whether or not the sensory information "reaches" awareness, it seems likely that there is at least one other factor. Theorists have referred to this using metaphors, such as a "dominant consensus" or "coherence with the dominant focus". Empirically, subjective experience has been associated with recurrent neural interactions in the cerebral cortex, perhaps mediated by the thalamo-cortical system, and we interpret the coherence or consensus metaphors in the context of a recurrent neural network. A recurrent neural network is characterized by a settling process during which the activity in the network converges toward a stable state. In sensory perception the landscape in state space is determined by learned connectivity patterns, context, and sensory input. The settling process progresses from a relatively unstable / low-certainty state, to a relatively stable / high-certainty state, consistent with evidence accumulation models of perceptual decision making. What Marcel Kinsbourne refers to as a coherent neuronal interaction might correspond to such a stable state, where one from among many possible states is selected in the form of a global perceptual decision. I will present new results obtained using novel data analysis techniques applied to combined MEG / EEG recordings aimed at measuring the relative stability of content-specific neural representations over brief intervals. The goal of this research is to test whether or not conscious representations can be distinguished from non-conscious responses based on differences in stability. Preliminary results suggest that they can. This approach may help to further our understanding of the difference between conscious and non-conscious perception and has potential applications in detecting signs of consciousness in non-communicating patients


Contact daniel.smith@dur.ac.uk for more information about this event.