Information and Living Systems: Philosophical and Scientific Perspectives

9. Decision Making in the Economy of Nature: Value as Information

9 Decision Making in the Economy of Nature: Value as Information Benoit Hardy-Vallée All organic beings are striving to seize on each place in the economy of nature —Darwin 1859/2003, 90 Cognition, to use Reuven Dukas’s formula, is the set of “neuronal processes concerned with the acquisition, retention, and use of information” (Dukas 2004, 347). Decision making is one of the principal uses of that information. Classically, decision making is not a topic of discussion in biology and philosophy of biology. The analysis and study of decision making is usually left to philosophy of mind, economics and psychology (see Hardy-Vallée 2007, forthcoming). Philosophers are mostly concerned with the normative features of decisions, that is, what makes a decision rational or not. In philosophy of mind, the standard conception of decision making equates deciding and forming an intention before an action (Audi 2001; Davidson 1980; Searle 2001). According to a different analysis, this intention can be equivalent to, inferred from, or accompanied by desires and beliefs. If desires are represented as utilities, and beliefs as probabilities, then decisions can be represented by rationalchoice theory (RCT) models. The two branches of RCT, decision theory and game theory, formalize the logical relationships between reasons. RCT specifies the formal constraints on optimal decision making in individual and interactive contexts. Rational agents select actions that have the higher subjective expected utility (obtained by multiplying probabilities and utilities) and select equilibrium strategies, that is, n-tuples of states where no player has an advantage to deviate from (see Baron 2000 for an introduction). RCT is also a framework for building predictive models of choice behavior: which lottery an agent would select, whether an agent would cooperate or not in a prisoner’s dilemma, and so on. Experimental economics, behavioral economics, cognitive science, and psychology (I will refer to these disciplines broadly as “psychology”) 254 Chapter 9 use this model to study how subjects make decisions and which mechanisms they rely on for choosing. These patterns of inference can then be compared with RCT (Camerer 2000; Kahneman and Tversky 1979, 1991, 2000; Thaler 1980). Standard philosophical, economic, and psychological analyses of decision making implicitly or explicitly adopt what could be called a “cogitative ” conception. On this account, decision making is a high-level, explicit, and deliberative process analogous to reasoning. Philosophers explain decisions as inferential transitions between propositional attitudes . These transitions can—at least in theory—be made explicit. As Davidson explains, “If someone acts with an intention then he must have attitudes and beliefs from which, had he been aware of them and had he the time, he could have reasoned that his act was desirable” (Davidson 1980, 85). Economists and rational-choice theorists represent agents as Homo economicus: all their preferences are transitive and they select actions by computing probabilities and utilities. In a game-theoretic context, thanks to common knowledge of rationality and backward induction, they infer the equilibrium strategy and act upon it. Psychological research also clearly assumes that deciding is an explicit, “high-level process” (Johnson -Laird and Shafir 1993, 1); decisions are “reached by focusing on reasons that justify the selection of one option over another” (Shafir, Simonson, and Tversky 1993, 34). The fact that it is studied mostly by multiple-choice tests using paper and pen illustrates well how decisions are conceived in psychology: subjects’ decision-making competence is supposedly revealed by questionnaires on probabilistic reasoning. Consequently , it is not surprising that decision making does not stimulate many debates or research in biology and philosophy of biology: it is not construed as a biologically significant capacity. It compares to chess playing: an occasional activity made possible by uniquely human, highlevel cognitive capacities. I would like to suggest here that contrary to common wisdom, decision making is not specifically human, but rather a behavioral control scheme typically found in animals endowed with sensory, motor, and control apparatuses, and, more specifically, in brainy animals (craniates, arthropods, and cephalopods). There are of course some decisions that will fall outside the scope of this analysis: some because they involve multiagent coordination (e.g., jury decision making, institutional processes ), and others (e.g., in ethical or scientific contexts) because they appeal to our theoretical rationality. Humans have, for instance, to Decision Making in the Economy of Nature 255 ponder the justification of a moral claim or the soundness of...