Single-cell neuronal model for associative learning.

Abstract

Recently, a novel cellular mechanism, activity-dependent neuromodulation, was identified in sensory neurons mediating the gill and tail withdrawal reflexes in Aplysia. This mechanism may explain associative learning on a behavioral level. The present study was designed to mathematically model subcellular events that may underlie this mechanism and to examine the ability of the model to fit available empirical data. In this associative model, the reinforcing or unconditioned stimulus (US) leads to non-specific enhancement of transmitter release from sensory neurons by activating a cAMP cascade. Spike activity in sensory neurons, the conditioned stimulus (CS), transiently elevates intracellular Ca2+. The CS-triggered increases of intracellular Ca2+ "primes" the cyclase and amplifies the US-mediated cAMP synthesis. As a result of pairing specific amplification of cAMP levels, transmitter release is enhanced beyond that produced by unpaired stimuli or by application of the US alone. The model is capable of fitting empirical data on activity-dependent neuromodulation and predicts a characteristic interstimulus interval (ISI) curve. At the subcellular level, the model's ISI function is related to the time course of the buffering of intracellular Ca2+. The magnitude and duration of the pairing specific enhancement of transmitter release is related to the levels and time course of intracellular cAMP stimulation.