Dave Hammond, 3/4/03 <br> <br> 

State switching in a simple bistable neural network model
<br>
<br>
Bistability is a property inherent in many models of neural networks,
and has been proposed as a possible mechanism for working short-term
memory. To be useful for memory, it must be possible to adjust the
state of the network by external input. In this talk I will examine
this problem in detail for a very simple two population firing rate
model, and describe how the ability to 'reset' the state of the
network depends upon the model parameters.
<br>
<br>