Complex dynamic behaviors, including oscillations, are ubiquitous in the nervous system. Our research examines the role of neural control mechanisms in generating these dynamics. However, the nervous system is also subjected to the influence of random inputs, i.e., noise. A basic question is how to determine which part of the dynamics is generated by the control mechanism and which part reflects the influence of noise. We use an interdisciplinary approach involving experimentation in concert with computer and mathematical modeling. 

Our research focuses on the dynamics of neural control mechanisms involving 2-3 neurons. These systems are small enough to permit experimental manipulation while at the same time are well enough characterized so that simple, yet plausible mathematical models can be posed. Thus, direct comparisons between theory and observation become possible. 

Most of our research involves the pupil light reflex. The study of this reflex is facilitated by the use of an experimental clamping method which allows us to electronically insert feedback ("area comparator" in figure) into the reflex arc. Currently we are working on the influence of noise on stability of the pupil reflex. Since this reflex can be regarded as a paradigm for neural feedback control, we anticipate that by studying its dynamics we will be able to find general precepts applicable to other neural systems. 

 Milton JG (1998) Epilepsy and the multistable nervous system. In: Self-organized Biological Dynamics and Control by External Stimuli (J Walleczuk, ed). Cambridge University Press: Cambridge (in press).

 Hunter JD, Milton JG, Thomas PJ and Cowan JD (1998). A resonance effect for neural spike time reliability. J. Neurophys. (accepted).