[Comp-neuro] PhD in Computational and Mathematical Neuroscience (EPSRC funded, 3.5 yrs, Sept 2019 start)

James Rankin james.rankin at gmail.com
Thu Jan 3 14:02:24 CET 2019

*PhD in Computational and Mathematical Neuroscience (EPSRC funded, 3.5 yrs,
Sept 2019 start)*

General information & apply (funding is primarily targeted at UK students):

Project description:

*Closing date 7th Jan 2019*

*Dynamical System Modelling of the Early Auditory Pathway*

This interdisciplinary project will develop computational and mathematical
models of the auditory system to understand how complex stimuli like speech
are encoded by spiking neurons in the midbrain.

The auditory midbrain is a key hub in the auditory processing pathway,
functioning as an important junction that relays and shapes neural signals
as they ascend towards auditory cortex.  Knowledge of the way in which
complex sounds, e.g. speech, are encoded in the midbrain is crucial for
understanding how dysfunction in the earlier auditory processing pathway
(cochlea, auditory nerve, cochlear nucleus) leads to different types of
hearing loss (a problem affecting 1 in 6 people in the UK). Working with
neural recordings from the auditory midbrain in gerbils, a commonly-used
animal for the study of low-frequency hearing, this project will develop
mathematical and computational models of the auditory processing pathway.
The aim is to understand the different roles of the patterns of inputs to
midbrain neurons and their intrinsic response properties (e.g. their
spiking rate) in shaping their responses to complex sounds.

The project will use a dynamical systems approach to model the intrinsic
properties of individual neurons in the midbrain in a biologically
plausible way (working with, e.g. adaptive exponential integrate-and-fire
neurons or the Hodgkin-Huxley equations). Inputs to these neurons will be
based on established cochlear models and the biological details of the
auditory nerve and cochlear nucleus. The resulting model will produce
firing patterns directly comparable with neural recordings provided by the
experimental supervisor. This data will be used to train and parameter fit
the model using e.g. Bayesian optimisation or genetic algorithms. The
resultant model will have explanatory power for the extent to which
midbrain responses are shaped by its inputs from cochlear nucleus. Further,
it will make predictions, testable in new experiments, of how midbrain
responses will be affected by different dysfunctions of the early auditory
system relating to hearing loss.

The successful candidate will receive training dynamical systems theory and
in the development and analysis of individual neuron and neural network
models. An interdisciplinary approach, incorporating known biological
details of the auditory processing pathway, will require the candidate to
learn the relevant biology and neuroscience along with mathematical and
computational techniques. The project will involve working closely with
experimental neuroscientists and experimental data. This project provides a
unique opportunity to receive training in mathematical modelling in close
collaboration with experimentalists using cutting-edge methods recording
spikes simultaneously from hundreds of neurons. Experience working on such
interdisciplinary projects is highly sought after.

Candidates with quantitative backgrounds (mathematics, physics,
engineering) and from neuroscience programmes are encouraged to apply.
Programming experience, knowledge of dynamical systems theory and
experience in biological modelling are a plus.

For further information, please contact me at the email address above.
*For further information, please contact Dr James Rankin, email
j.a.rankin at exeter.ac.uk <j.a.rankin at exeter.ac.uk>*

See also another project (co-supervised with Joel Tabak):
*Determination of parameter dependencies across diverse populations of
neuro-endocrine cells*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.tnb.ua.ac.be/pipermail/comp-neuro/attachments/20190103/44321fa1/attachment-0001.html>

More information about the Comp-neuro mailing list