CNS*1997
The Annual Computational Neuroscience Meeting
July 6 - 10, 1997, Big Sky, Montana
CNS*1997 Workshops
The role of chaotic dynamics in neural systems.
Peter Rowat
It is proposed to hold a discussion on the following topic:
Is the occurrence of low-dimensional deterministic chaos in a neural system simply an epiphenomenon like the rattling sounds in a car when driving on a rough road, or is it an essential functional component of the system, like the action of your car springs on the rough road?
Some questions that could be raised:
Recent work in a small motor-pattern-generating system has shown that an isolated neuron may display low-dimensional chaotic dynamics, and that intercellular connections in small networks may modify - increase or decrease - the Lyapunov exponents associated with neurons having chaotic dynamics. What functional advantages does this confer on a motor system?
Is the unexpectedly high variability of the interspike interval in cortical spiketrains due to deterministic chaos rather than stochastic noise - high-dimensional chaos - in the environment? Van Vreeswick and Sompolinsky have proposed that this is due to an approximate balance between the excitatory and inhibitory synaptic inputs to a neuron in a large cortical-like network.
A closely related question arises from the observations by Mainen and Sejnowski of precisely timed spike firing patterns in response to stimuli with small, synaptic-input like, fluctuations. Could this result from the small fluctuations "controlling" the endogenously chaotic dynamics of individual cells?
Is the chaos observed in olfactory systems (Freeman et al) generated endogenously by the chaotic dynamics of individual neurons, or is it a consequence of the connections between neurons, or by some combination of intrinsic cellular and network dynamics?
A physical system in a chaotic parameter regime can switch between distinct strange attractors in response to a changing input much faster than a corresponding non-chaotic system. Is this property utilized in biological sensory-motor systems?