When prior knowledge overrules new evidence: Adaptive use of decision strategies and the role of behavioral routines 1The paper is based on a presentation held at the 16th SPUDM conference in Leeds, 1997. The research was supported by a grant from the Deutsche Forschungsgemeinschaft to Klaus Fiedler and Tilmann Betsch, within the Sonderforschungsbereich 504 at the Universities of Mannheim and Heidelberg. We are grateful to Frank Siebler for programming the Mouselab. Thanks are due to Susanne Haberstroh, Arndt Bröder, Bronwyn Jones and Esther Goldbloom for many helpful comments on earlier versions of the paper.

This paper focuses on behavioral routines in adaptive decision making. In an experiment consisting of two phases, participants worked on recurrent, multiattribute choice problems. In the first phase, routines were induced by relying upon the human ability to adapt to situational changes by changing decision strategies. To induce strategy change, time pressure was varied as a within factor. Payoffs were manipulated so that an adaptive change in strategy led participants to maximize choice frequency for one out of three options (routine acquisition). After a one week time lapse, participants worked on similar problems, containing the previously preferred routine option. In this second phase, payoffs favored deviation from the routine option. Results showed that choices were almost perfectly calibrated to payoffs under low time pressure. However, if time pressure increased, participants were more likely to prefer the routine option, even though search strategies were still used adaptively and evidence discouraged routine selection. Results are discussed with reference to the model of adaptive decision making ( Payne, Bettman & Johnson, 1993 ), and the MODE model of attitude-behavior relation ( Fazio, 1990 ).