PhD position: Analytic and computational models of auditory processing

[Chris Sumner <Chris@xxxxxxxxxxxxx>]

Dear all,

please pass this on to any potential candidates or other interested parties.

Apologies if you recieve this twice.

cheers,
Chris Sumner

-------------
Analytic and computational models of auditory processing

Dr Steve Coombes, School of Mathematical Sciences, University of Nottingham.
Dr Chris Sumner, MRC Institute of Hearing Research.

The brain is able to construct a view of the outside world from sound 
alone: it lets us listen to music and locate each musician in space; 
pick out a bassoon in an orchestra or follow the grand themes of a 
symphony; follow one person speaking amidst the cacophony of a room full 
of conversations. Yet the brain receives only two one-dimensional 
inputs: one from each ear. And even with one ear, we still perceive a 
world of discrete, separate sound sources. This project will investigate 
with analytical and computational models, the neural processing that 
might underlie this remarkable feat of hearing.

The range of possible projects is broad. However, the motivation will be 
to bring abstract models of neural systems to real problems in auditory 
neuroscience. Analytical analogs will be sought between different modes 
of processing in order to describe canonically the neural processing. 
For example, what is the functional value of shunting versus subtractive 
inhibition in an auditory neuron? Analytical expansions of biophysical 
equations can show this. Other problems include how the dynamic 
properties of highly artificial networks relate to networks of 
biophysically realistic neurons with realistic auditory inputs. 
Traditional models of neural processing have been based on the idea that 
information is conveyed in neurons by the rate at which they fire. 
However, in the auditory system much information is conveyed about sound 
by the timing of single spikes. Recently, mathematicians have started to 
study the properties of spiking neurons, their dynamics and behaviour in 
networks such as oscillations and synchronisation. Another interesting 
problem is how, when individual neurons are so non-linear, 
representations in the cortex can be quite linear. Analytic studies will 
be complemented by simulations to confirm them, or inform when 
analytical solutions are not possible.

A suitable candidate will probably have a good first degree in maths, 
physics or engineering. Some experience in programming is very 
desirable. They should have an interest in applying their quantitative, 
analytical skills to scientific questions. They will interact with 
neuroscientists, psychologists and mathematicians. There will also be 
opportunities for becoming involved in experimental work on hearing. 
This will lead to an interdisciplinary PhD with mature skills in 
computational neuroscience and mathematics.

This position is funded by a Marie Curie Early Stage Training 
Fellowship, as part of the programme for Research Training in 
Mathematical Medicine and Biology at the University of Nottingham 
(MMBNOTT). Marie-Curie Fellowships are available for this project from 1 
September 2006 and 1 June 2007 and are open to non-UK nationals. There 
is a monthly living allowance (gross), plus travel allowance, mobility 
allowance, career exploratory allowance, all in excess of 30K euros  per 
year. For full details of the training programme, eligibility 
requirements and application procedures please see 
www.maths.nottingham.ac.uk/mmbnott. For informal enquires please contact 
Chris Sumner (chris@xxxxxxxxxxxxx) or Steve Coombes 
(stephen.coombes@xxxxxxxxxxxxxxxx).

--
Dr. Christian J. Sumner
MRC Institute of Hearing Research
Tel: 0115 922 3431
Fax: 0115 951 8503


This message has been checked for viruses but the contents of an attachment
may still contain software viruses, which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.