A Neural Model of Biased Oscillations in Aplysia Head-Waving Behavior
Wood, Richard J.
MetadataShow full item record
A long-term bias in the exploratory head-waving behavior of Aplysia can be induced using bright lights as an aversive stimulus: coupling onset of the lights with head movements to one side results in a bias away from that side (Cook & Carew, 1986). This bias has been interpreted as a form of operant conditioning, and has previously been simulated with a neural network model based on associative synaptic facilitation (Raymond, Baxter, Buonomano, & Byrne, 1992). In this article we simulate the head-waving behavior using a recurrent gated dipole, a nonlinear dynamical neural model that has previously been used to explain various data including oscillatory behavior in biological pacemakers. Within the recurrent gated dipole, two channels operate antagonistically to generate oscillations, which drive the side-to-side head waving. The frequency of oscillations depends on transmitter mobilization dynamics, which exhibit both short- and long-term adaptation. We assume that light onset results in a nonspecific increase in arousal to both channels of the dipole. Repeated pairing of arousal increments with activation of one channel (the "reinforced" channel) of the dipole leads to a bias in transmitter dynamics, which causes the oscillation to last a shorter time on the reinforced channel than on the non-reinforced channel. Our model provides a parsimonious explanation of the observed behavior, and it avoids some of the unexpected results obtained with the Raymond et al. model. In addition, our model makes predictions concerning the rate of onset and extinction of the biases, and it suggests new lines of experimentation to test the nature of the head-waving behavior.