LEARNING FROM EXPERIENCE, REFERENCE POINTS, AND DECISION COSTS

This paper examines the process by which individuals learn from experience in economic settings. An experiment compared the forecasting accuracy across three groups of subjects. The results support previous findings that individuals utilize reference points during the learning process, Here it is argued that as a result, affine transformations of the payoff function can affect the degree to which the equilibrium decisions will be consistent with the Nash predictions. Similarly, increasing the marginal rewards from making more accurate decisions may have minimal effects. These findings suggest that the appropriateness of normative models as descriptions of individual behavior may depend upon the reference point utilized during the learning process.