ESCAPE, AVOIDANCE, AND IMITATION - A NEURAL-NETWORK APPROACH

We present a real-time neural network that integrates classical and op erant processes to describe how animals learn to escape and avoid an a versive stimulus either by trial and error or by imitation. it is assu med that through classical conditioning animals build an internal mode l of the environment and that through operant conditioning animals sel ect from alternative responses. The internal model of the environment provides predictions of the aversive stimulus based on environmental s timuli and the animal's own responses, and these predictions are used to train the operant conditioning block to generate responses that min imize the aversive stimulus. Computer simulations show that the model correctly describes many of the features that characterize escape and avoidance.The network also is able to describe the imitation of a demo nstrator by an observer. During the demonstration, a neural network re presenting the observer stores classical associations between environm ental stimuli and the demonstrator's responses and aversive stimuli, a nd these associations serve to train the operant associations during t he observer's performance. it is assumed that the demonstrator's respo nses evoke a representation of identical responses in the observer and that the demonstrator's unconditioned response to the aversive stimul us serves as an aversive reinforcer for the observer The network contr ibutes to a general theory of adaptive behavior and is relevant to the design of autonomous systems that learn either through trial and erro r or through imitation.