A survey of thirty-one experimental studies which report data on the c
omparative effects of monetary rewards and opportunity cost shows: (1)
several studies in which increased rewards shift the central tendency
of the data toward the predictions Of rational models; (2) in virtual
ly all cases rewards reduce the variance of the data around the predic
ted outcome. This is consistent with a model in which rewards are bala
nced against decision cost in agent behavior and explicates the argume
nt that when rational models fail it can be attributed to low opportun
ity cost of deviations from the rational prediction.