Neil Roach - Research

My research focuses on the neural mechanisms of perception. Ultimately I want to understand how sensory information is represented in the brain and how it is interpreted when we form decisions and interact with the environment. To address these questions I conduct psychophysical experiments and use the results to constrain and test physiologically-motivated computational models of sensory encoding and decoding.

I have a range of interests spanning both normal and abnormal sensory processing and am interested in hearing from prospective PhD students or postdoctoral fellows who are interested in this approach. Examples of current research topics include:

1. Visual forward modelling

The spatial structure of natural images is correlated over time, due to self-motion of the observer and motion of objects in the environment. Can the visual system exploit this spatiotemporal regularity to predict the likely future pattern of input it will receive? If so, how are the forward predictions formed? How do they affect subsequent visual processing and what functional role does this predictive processing fulfill? For an example of recent work on these questions, see Roach, McGraw & Johnston 2011, Current Biology

2. Multisensory processing

What are the rules governing the integration and segregation of auditory-visual information? How does the brain code the timing of different sensory signals? How can we explain distortions of perceived timing? For an example of recent work on this topic, see Roach, et. al. 2011, Proc Royal Soc B

3. Visual adaptation

The statistics of recent sensory input impacts upon all levels of visual processing, from the response profiles of single neurons to the judgements made by human observers. As a result, adaptation provides a fertile testbed for theories aiming to bridge the gap between perception and its underlying neurophysiological mechanisms. How well can we reconcile the changes occurring at these different levels of analysis? In doing so, what can we learn about how the system works? For an example of recent work on this topic, see Roach, et. al. 2009, J.Neurophysiol