Joint QBI/Maths seminar: When the noise is the signal: Stochastic Synchrony in Olfaction
Event Details
- Date:
- Friday, 05 February 2010
- Time:
- 11:00 am - 12:00 pm
- Room:
- Building 67, Room 442
- UQ Location:
- Priestley Building (St Lucia)
- URL:
- http://www.pitt.edu/~phase/
- Event category(s):
Event Contact
- Name:
- Mr Reeza Palamoodu Nazer
- Phone:
- 66416
- Email:
- r.nazer@uq.edu.au
- Org. Unit:
- Queensland Brain Institute
Event Description
- Full Description:
- Speaker: Prof G. Bard Ermentrout
Department of Mathematics University of Pittsburgh
Title: When the noise is the signal: Stochastic Synchrony in Olfaction
Abstract: Rhythmic oscillations in the 20-40 Hz range are often found in the
olfactory bulb (OB, sense of smell). These oscillations indicate that
there must be synchronization between the principle neurons in the OB,
namely the mitral cells. There is very little direct coupling between
these neurons so that the synchrony must arise from other
mechanisms. A population of inhibitory cells (granule cells, GCs) is
believed to mediate the synchrony. However, the output of GCs is
asynchronous and stochastic, but does provide strongly correlated
inputs to the mitral cells. In this talk, I will describe a mechanism
through which correlated noise is able to synchronize uncoupled
oscillators; stochastic synchronization. I will start with some
simulations of conductance-based models. Then, I will show that under
some reasonable circumstances, these large dimensional systems can be
reduced to a class of simple scalar models called phase oscillators. I
then quantify the rate of synchrony, the input to output transfer of
correlation and what statistical features of the noise optimize the
the degree of synchrony. The theoretical results are verified through
a series of experiments involving the application of correlated
broadband currents to patch-clamped mitral cells. I close with some
recent results on the interaction between correlated inputs and
coupling. Mathematical methods from random dynamical systems,
averaging, and variational principles are used to solve the resulting
reduced systems. This work is joint with Nathan Urban, Roberto
Fernandez-Galan, Sashi Marella, Aushra Abouzeid, and Cheng Ly.
Directions to UQ
- Google Map:
-
- Directions:
- St Lucia Campus | Gatton campus.
Event Tools
Share This Event
Print
Email
Share
