In a wide range of social and economic situations, people are rewarded on the basis of an assessment of their past performance, but often an objective account of such performances is not available. As result, the assessors are forced to rely on their memories of past events.
People usually hope that successes will be remembered and failures forgotten, and they assume that the more recently an event has passed, the fresher it will be in the assessors mind. However, experiments have shown that memory is more complicated and more interesting than you may think.
Our memories actually operate on the principles of similarity and repetition. Similarity refers to the phenomenon wherein current events trigger memories of similar past events, while repetition refers to the fact that recalling the memory of an event increases the likelihood that it will be remembered in the future. People also have the tendency to recall events incorrectly or to "remember" events that never even occurred.
In this paper, Yianis Sarafidis (Assistant Professor of Economics and Political Science, INSEAD) takes a bounded rationality stance, assuming that imperfect memory is an indisputable limitation of human thought.
Building on research in cognitive psychology and neuroscience, he formulates a model of memory, then explores its economic consequences by addressing the issue of how one should time a sequence of informative events in order to manipulate the memories of a forgetful assessor.
Consider, for example, the situation of a politician seeking re-election. She or he has had several months of favorable press coverage, and now plans to announce a popular new tax package. Should this information be released immediately, to reinforce existing good memories of past events? Or should the news be timed for later, in order to make it more memorable on Election Day?
Or what about the employee who is evaluated on the basis of his or her past output performance? How does this shape how much effort the employee decides to exert? And how can he or she be sure the boss will remember it?
After outlining the psychological research that supports his assumptions about memory, Professor Sarafidis presents the model. He considers both the preceding examples: a fixed budget case (a senator running for re-election) and a convex cost case (an employee being evaluated on the basis of his past performance).
The general framework developed in this paper can be applied to almost any setting where products or people undergo periodic assessment, from executives managing the release of news prior to an IPO to lobbying groups seeking to shape public opinion to students timing their class contribution prior to the end of term.
Professor Sarafidis concludes the working paper with concrete advice about the most advantageous ways to time a sequence of informative events and with a brief discussion of possible extensions to his framework.