Artificiality, Reactivity, and Demand Effects in Experimental Economics

A series of recent debates in experimental economics have associated demand effects with the artificiality of the experimental setting and have linked it to the problem of external validity. In this paper, we argue that these associations can be misleading, partly because of the ambiguity with which artificiality has been defined, but also because demand effects and external validity are related in complex ways. We argue that artificiality (understood as unfamiliarity of the experimental environment) may be directly as well as inversely correlated with demand effects. We also distinguish between the demand effects of experimentation and the reactions that they may trigger and that might endanger experimental validity. We conclude that economists should pay more attention to the way in which subjects construe the experimental task and learn to exploit subjects' reactivity to expectations in their experiments