Risk Sensitivity: Economics meets Behavioural Ecology meets Psychology meets Neurobiology Alex Kacelnik, Oxford University, UK Since the consequences of most actions are stochastic, those studying how organisms act often consider how stochasticity (or risk) influences decision making. Economists have faced this problem directly since Bernoulli (1738) explained how risk aversion is generated by a non-linear utility function, psychologists since Kahnemann & Tversky (1979) discovered switches between risk aversion and risk proneness as a function of how the problem is presented (or "framed"), and biologists since Caraco (1980) and others predicted switches in risk attitude according to energetic state. All these theories are grounded in non-linear relations between payoffs and consequences (utility or fitness). As an alternative, I will describe Scalar Utility Theory, an approach that accounts for the main features of risk-sensitive behaviour in human and non-human animals using knowledge of information-processing in the absence of stochasticity of the inputs, and present evidence in its support. SUT binds Economics, Psychology and Behavioural Ecology because it predicts risk aversion for positive outcomes such as the size of food, water rewards or monetary gains and risk-seeking for undesirable outcomes such as delays to reward or monetary losses. It differs from alternative models in not requiring utility to be non-linear on payoff, in not predicting a switch of risk attitude with state and in its strong attachment to empirical descriptions of cognitive processes. At least for one dimension of reward (delay to reward) SUT is offering some hope of a forthcoming link to neurobiology. Key words: risk sensitivity, scalar utility theory, optimality, decision making.

Back to/ Retour à: Alex Kacelnik

Back to invited speakers	Retour aux conférenciers invités
Back to summary	Retour au sommaire