Rational choice theory (RCT) is standardly interpreted not as a causal theory, but as a normative one. Experimental economics, by replacing underlying assumptions of the standard interpretation of RCT by more realistic ones – by including pro-social preferences to the utility function, for instance – has improved the predictability of the RCT. That was not sufficient, however, to overcome another theoretical difficulty that RCT faces, which is the fact that its postulated entities are not rooted in any causal mechanism. In this paper, we show how neuroeconomics, because it is not limited to behavioral manifestations, may pave the way to the development of a natural science of decision-making.
In some contexts, the actual behaviors of individuals can appear as aberrant with rational agent-based theories. Behavioral theory tries to explain these anomalies, especially with respect to social preferences. The purpose of this paper is to show how experimental economics has provided an accurate measure and further characterization of these preferences. Based on a very selective review of experiments, we show how a fruitful dialogue has been established between behavior analysis in the laboratory and in theory.
The paper discusses the basis of the experimental method in general, its adaptation to the study of economic phenomena, and the recent philosophical thoughts on the nature of laboratory experiments. It also inventories the main experimental procedures that characterize best practices of economists—some of them quite ingenious—by drawing a parallel with what distinguishes them from former practices established by psychologists. In particular, the issue of monetary incentives is carefully examined. The paper emphasizes the immense progress that has been accomplished in a few decades while taking into account the disagreements that still persist between some experimentalists and the ongoing existence of some important “open” questions. In this regard, the paper tries to clarify a major methodological questioning which only recently has been approached more seriously: the issue of “external validity” of experimental results (i.e., their “generalizability” to the outside world).