Postdoctoral Position Available in Computational Neuroscience



From: Mark Goldman (mark_g@MIT.EDU)
Date: Thu Oct 31 2002 - 12:46:04 CET


Postdoctoral Position Available in Computational Neuroscience
Laboratory of Dr. Mark S. Goldman, Wellesley College, Wellesley, MA
    
I am looking to hire a postdoc to work in my computational neuroscience
laboratory at Wellesley College. The general goal of the laboratory is
to connect the biophysical properties of single neurons to the robust
behavior of neural networks. Work will primarily be theoretical but is
expected to be performed in close collaboration with experimental
studies. Typical projects will range from developing methods for
characterizing experimental data to building analytical and
computational models of particular neurobiological systems.
    Current laboratory interests are described below. References to
papers on some of these topics can be found on my web page:
http://hebb.mit.edu/people/markprojects.html. The selected candidate
will also have the flexibility to develop projects that meet his or her
own interests. The expected start date would be summer or fall of 2003.
    I would be happy to meet with any interested candidates at the SFN
meeting in Orlando. I can be contacted at mark_g@mit.edu.

Sincerely,
Mark Goldman

------------------------------------
CURRENT LABORATORY INTERESTS:
1)Robustness of persistent neural activity in the oculomotor neural
integrator.
    Integrator neurons receive transient eye movement command signals
yet maintain sustained firing at a rate proportional to eye position.
Because integrator neurons continue to fire in the absence of external
inputs, the integrator network is a model system for the study of
short-term memory. Current we are studying how single neuron properties
can add robustness to integrator network models.

2)Robust production of digestive rhythms in the crab stomatogastric system
The stomatogastric network is the best studied system for determining
the effects of neuromodulators on network function. The networks of
different crabs produce similar outputs even though the properties of
individual identified neurons in different crabs can be highly
dissimilar. By contrast, small changes in the properties of these
neurons induced by neuromodulators result in large and consistent
changes in network output. We are investigating the organizational
principles of the network that lead to this interesting and seemingly
paradoxical behavior.

3)Processing of natural images
Broad classes of images from the natural world, although appearing
highly different, are statistically quite similar. Properties of
cortical neurons and networks that are specialized to deal with the
statistical similarities of natural images will be characterized,
identified, and modeled.

4)Connecting local synaptic plasticity rules to global network functions
    Much recent experimental work has been spent identifying spike-based
plasticity rules that operate at single synapses. Simultaneously, much
theoretical work has tried to derive global network learning rules that
allow networks to optimally perform a variety of functions. This
project seeks to develop concrete connections between high level global
learning algorithms and single neuron experimental findings in systems
with well characterized neurobiological functions.



 
Home   Login   Meetings   Courses   Belgium   Maillist   Credits
  Servers   Links   Archive

Please send comments and suggestions to Mike Wijnants.

 
Page last updated on Thursday, 31-Oct-2002 12:56:59 CET © BBF 1998 all rights reserved