The geometry of stimulus control

Abstract

Many studies, both in ethology and comparative psychology, have shown that animals react to modifications of familiar stimuli. This phenomenon is often referred to as generalisation. Most modifications lead to a decrease in responding, but to certain new stimuli an increase in responding is observed. This holds for both innate and learned behaviour. Here we propose a heuristic approach to stimulus control, or stimulus selection, with the aim of explaining these phenomena. The model has two key elements.First, we choose the receptor level as the fundamental stimulus space. Each stimulus is represented as the pattern of activation it induces in sense organs. Second, in this space we introduce a simple measure of 'similarity' between stimuli by calculating how activation patterns overlap. The main advantage we recognise in this approach is that the generalisation of acquired responses emerges from a few simple principles which are grounded in the recognition of how animals actually perceive stimuli. Many traditional problems that face theories of stimulus control (e.g. the Spence-Hull theory of gradient interaction or ethological theories of stimulus summation) do not arise in the present framework. These problems include the amount of generalisation along different dimensions, peak-shift phenomena (with respect to both positive and negative shifts), intensity generalisation, and generalisation after conditioning on two positive stimuli

Reward quality influences the development of learned olfactory biases in honeybees

“…(b) Peak shift depends upon the relative difference in the reinforcers Peak shift is observed in generalization gradients under a limited set of conditions that depend in part on the characteristics of learned stimuli (for reviews see Ghirlanda & Enquist 1999. The probability that an animal will generalize a learned response to a new stimulus depends on the perceptual similarity of the new and learned stimuli, and generalization along a stimulus gradient typically follows a Gaussian or exponential function (Shepard 1987).…”

“…The probability that an animal will generalize a learned response to a new stimulus depends on the perceptual similarity of the new and learned stimuli, and generalization along a stimulus gradient typically follows a Gaussian or exponential function (Shepard 1987). Differential learning with two stimuli, one of which is excitatory and the other inhibitory, does not produce a Gaussian generalization function, however (Hull 1943;Mackintosh 1974;Blough 1975;Ghirlanda & Enquist 1999). Instead, a peak-shifted generalization function is often formed, which may result from 'two opposed response tendencies'-an excitatory function and an inhibitory one (Spence 1937;Mackintosh 1974;Blough 1975)-produced by learning to associate two similar conditioned stimuli with two different unconditioned stimuli (US).…”

Intensity Generalization: Physiology and Modelling of a Neglected Topic

“…The main aim of this paper is to evaluate a number of existing theories with respect to their ability to account for intensity generalization. I consider the following models: gradient-interaction theory (Spence, 1936(Spence, , 1937Hull, 1943Hull, , 1949, several theories based on the concept of similarity between stimuli (Shepard, 1987;Nosofsky, 1986;Pearce, 1987), overlap theory (Ghirlanda & Enquist, 1999), the model by Blough (1975) and feed-forward neural networks (see Haykin, 1994). These theories FIG.…”

Vestigial preference functions in neural networks and túngara frogs.

“…Although the shapes of generalization gradients are presumed to influence signal evolution, and to be of interest in their own right, few ethologists have addressed the forces that determine these shapes (but see refs. [19][20][21][22][23][24][25][26].…”

“…A number of groups have begun to use artificial neural network models to investigate the evolution of perceptual mechanisms (24)(25)(26)(27)(28)(29)(30)(31). Because neural network models distribute the representation of a signal across many ''neurons,'' these models often generalize as an automatic result of training (32,33), making them useful tools for the exploration of preference functions.…”