Sriram, N. and Howard, R., National University of Singapore, Singapore
Herrnstein and Prelec (1991) detailed Melioration, a manifestation of bounded rationality in inter-temporal choice. In a coin-toss experiment their subjects pressed one of two keys and were presented with a delay before monetary reward. One key (X) always resulted in a shorter delay than the alternative (Y). But the delays for both keys increased in proportion to the number of X's in the window of the last k choices. In trying to maximize reward in a fixed time, subjects systematically deviated from the optimal solution (press only Y) and did not improve with experience on the task. We ran a variation of this paradigm on a sample of prisoners who were classified as Psychopaths (Ps) or NonPsychopaths (NPs) but were equivalent on general intelligence. The delay was constant on each trial of a 100-trial experiment; what varied was the amount of reward that varied as a proportion of X keypresses (X always resulted in larger rewards but also reduced reward for both X and Y) during the current window of 7 trials. NPs showed clear evidence of melioration in that their performance was suboptimal and declined during the second half of the experiment. In contrast, Ps showed no decrement in performance in the second half of the experiment and their absolute performance was superior to the NPs. We interpret these results in the light of findings and theories that link Psychopathy to defective affective processing and propose that individuals with muted sensitivity to feelings are less likely to be susceptible to melioration biases.