Academic journal article Cognitive, Affective and Behavioral Neuroscience

A Neural Network Model of Foraging Decisions Made under Predation Risk

Academic journal article Cognitive, Affective and Behavioral Neuroscience

A Neural Network Model of Foraging Decisions Made under Predation Risk

Article excerpt

This article develops the cognitive-emotional forager (CEF) model, a novel application of a neural network to dynamical processes in foraging behavior. The CEF is based on a neural network known as the gated dipole, introduced by Grossberg, which is capable of representing short-term affective reactions in a manner similar to Solomon and Corbit's (1974) opponent process theory. The model incorporates a trade-off between approach toward food and avoidance of predation under varying levels of motivation induced by hunger. The results of simulations in a simple patch selection paradigm, using a lifetime fitness criterion for comparison, indicate that the CEF model is capable of nearly optimal foraging and outperforms a run-of-luck rule-of-thumb model. Models such as the one presented here can illuminate the underlying cognitive and motivational components of animal decision making.

Foraging behavior has been characterized as an inherently dynamic process (Mangel & Clark, 1988; McFarland, 1977; McNamara & Houston, 1986; Stephens & Charnov, 1982). In order to survive, actively feeding animals must make a number of important foraging decisions on any given day. For example, an animal must decide where to search for food, what foods to consume, and how long to forage before moving to a new location. Assessing the costs and benefits of foragers' decisions has led foraging modelers to borrow theoretical concepts from economics and decision theory, such as utility, risk sensitivity, and average rate maximization (see Stephens & Krebs, 1986, for a review). In addition, experimental psychologists have shown that decision making under risk is heavily influenced by short-term emotional reactions (Townsend & Busemeyer, 1989; Tversky & Kahneman, 1974, 1981).

Animal foraging theory, like human decision-making theory, has moved back and forth between normative and descriptive models. The normative models are usually some variant of optimal foraging theory (OFT), which assumes that the principles of evolutionary biology apply to animal decision making in a foraging context (Krebs, 1978; McFarland, 1977). Animals that forage efficiently are favored by natural selection and thus should have evolved as optimal (or nearly optimal) decision makers. As a normative model, OFT prescribes how an animal should behave in order to maximize a particular currency (e.g., net energy intake or probability of survival), which is then interpreted in terms of a forager's lifetime fitness. Although OFT does not explain the proximate mechanisms that underlie actual behavior, it gives insights into some decision-making processes that might be used by animals.

Yet an increasing number of scientists recognize that both animal and human decision making is heavily influenced by short-term contexts and may therefore deviate from maximization of lifetime fitness. For example, Reboreda and Kacelnik (1991) and Marsh and Kacelnik (2002) have noted that foraging starlings are sensitive to risk and not just to expected value in choices among both amounts and delays of food. Like Tversky and Kahneman's (1981) human subjects deciding between monetary gambles, these starlings tend to be risk seeking for choices among losses (and, sometimes, risk averse for gains). This suggests that a foraging theory is required that incorporates short-term emotional influences and yet predicts adequate, if not optimal, fitness.

How can such emotional influences on decision making be understood quantitatively? There have been a few models of foraging decisions using neural networks (Mangel, 1990; Niv, Joel, Meilijson, & Ruppin, 2002a, 2002b; Real, 1992). However, none of those models incorporates biologically realistic principles for dynamic affective evaluation in real time. One such principle is opponent processing, based on comparison of current stimuli with ongoing expectations (Solomon & Corbit, 1974). Opponent processing has played a major role in neural models of cognitive-emotional interactions in human decision making (Grossberg & Gutowski, 1987; Leven & Levine, 1996). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.