The purpose of this paper is to revisit the influence of delayed feedback on learning, performance and strategy search. Two main working hypotheses have been used as a guideline of the study. First hypothesis claimed that when learning environment is "lenient" (errors ar not heavily punished), postponed feedback aids learning process. Second hypothesis asserted that when learning environment is "exacting" (penalities imposed for errors are severe), postponed feedback adversely affects learning process and performance.

An experiment with real people has been designed and conducted to test these hypotheses.

While the experimental data disconfirmed both theses, it showed an interesting phenomenon. Subjects in the lenient environment, when presented with delayed feedback, tended to change their strategy more often than the subjects receiving immediate feedback. On reverse, delaying the feedback in the exacting environment, made the subjects to be more consistent in the application of their strategy.

Full text