Procedural Rationality Through Exploration?Discovering How to Effectively Learn from Outcome Feedback
Sprache des Vortragstitels:
Sharing Behavioral Economics in Person (Jahreskonferenz der Society for the Advancement of Behavioral Economics)
Sprache des Tagungstitel:
Domain-general learning mechanisms often enable decision makers to learn from outcome feedback which actions tend to achieve a desired end. However, in a new and complex environment decision makers must explore how to learn effectively, i.e. how to elicit and evaluate outcome feedback that enables them to obtain the requisite knowledge despite the vastness of the search space. Our experiments investigated in a dynamic business simulation whether, and if so how frequently, participants discover an effective procedure to learn the optimal turnpike. We also probed whether the payoff scheme affects learning procedures and performance. In different studies, we found that a (relatively small) number of participants discovered an effective learning procedure and succeeded in approximating the optimal policy. In line with the method of heuristics, the effective learning procedure involved (1) a simplification of the search space and (2) the application of domain-general learning rules to the simplified space.