A Theory and Computational Model of Auditory Selective Attention

Research Student: Stuart N Wrigley Supervisor: Guy J Brown

This project concentrates on the role that attention plays in auditory perception. In common usage, attention usually refers to both selectivity and capacity limitation. It is widely accepted that conscious perception is selective and that perception encompasses only a small fraction of the information impinging upon the senses. The second phenomenon - that of capacity limitation - can be illustrated by the fact that two tasks when performed individually pose no problem; however, when they are attempted simultaneously, they become very difficult. This occurs even when the two tasks are not physically incompatible such as reading a book and listening to the radio. It is this that leads to the common conclusion that attention is a finite resource.

When producing a model of auditory attention, a number of experimental findings need to be borne in mind:

  • Attentional allocation: multiple frequency regions can be attended to simultaneously
  • Attentional shape: the spotlight hypothesis describing abrupt cutoffs between attended and unattended regions is no longer consistent with the data. Instead, allocation of attention must be considered to be on a gradient basis with the maximum at the attended region and decaying with distance from that point.
  • Attentional set has a dramatic effect on the outcome of binaural streaming tasks (eg Carlyon et al., 2001).

Current work is focusing on the incorporation of an attentional model into a neural oscillator based model of stream segregation. Such networks have been successfully employed for the stream segregation task (eg Wang and Brown, 1999) but have made little attempt to address the role attention plays in streaming.

A copy of the final thesis is available (see references below).


  • Carlyon, RP, Cusack, R, Foxton, JM and Robertson, IH (2001). Effects of attention and unilateral neglect on auditory stream segregation. Journal of Experimental Psychology: Human Perception and Performance 27(1) 115-127.
  • Wang, DL and Brown, GJ (1999). Separation of speech from interfering sounds based on oscillatory correlation. IEEE Transactions on Neural Networks 10 684-697.
  • Wrigley, SN (2002). A theory and computational model of auditory selective attention. PhD Thesis, Department of Computer Science, University of Sheffield, UK. (doubled sided pdf | single sided pdf)