Learning in the absence of feedback – an experimental study

Sequences of decisions where the consequences obtained at one stage may determine the choice alternative available at the next one are known as dynamic decisions. Since a long series of decisions need to be made, one important feature of this type of decision is the lack of immediate outcome feedback. How can decision makers learn in the absence of outcome feedback? In this paper, I report an experimental study of learning in the absence of immediate feedback. The setting is a computerized treasure hunt game, where the treasure can only be found at a fixed location. Results indicate that subjects seem to develop a mental map of the relative values of different locations, starting from the locations that are closest to the solution.