Decision theory: what's in it for me?

Abstract

Now that a great deal is known about visual sensory processing, the following question has gained interest in the neurophysiological community: how is visual and other sensory information used to make a decision? More generally, how are decisions made? The framework of decision theory suggests that the options are evaluated in terms of expected payoff and outcome probability: choices with higher payoffs and choices yielding more likely payoffs are preferred. This is true not only in humans, but also in animals. In a recent study 1 Platt M.L. Glimcher P.W. Neural correlates of decision variables in parietal cortex. Nature. 1999; 400: 233-238 Crossref PubMed Scopus (1204) Google Scholar , Platt and Glimcher investigated the neural basis for decision-making in awake rhesus monkeys by recording from single neurons in posterior parietal cortex (area LIP). The animals performed a decision task that involved making a saccade to one of two targets, which was cued during the trial. The authors asked whether neurons represent not only the location of a visual stimulus and the outcome of a decision (in this case an eye movement), but also the expected payoff or outcome probability of a decision. To answer this question, they varied either the expected gain or the outcome probability across blocks of trials, in separate experiments. In both cases, they found that the activity of parietal neurons was modulated according to the probability of a gain (a juice reward) as a result of the monkey’s response; in other words, neurons do in fact represent the decision variable (expected gain or outcome probability). In additional experiments, the authors estimated, using a psychophysical method, the value that the monkey attached to the two choices in a trial, and found that the estimate was also represented in the neural activity. The gap between sensory processing and complex decision-making is being narrowed by insightful experiments such as these.