Memory for Reward in Probabilistic Choice: Markovian and Non-Markovian Properties

Pigeons were rewarded with food for pecking keys in various forms of two-armed bandit situation for an extended series of daily sessions in two experiments. The average daily preference (S=R/[R+L]) is very well fit by a markovian linear model in which predicted preference today is an average of predicted preference yesterday and reinforcement conditions today: s(N+1) = as(N) + (1-a)A(N+1), where A(N+1) is set equal to 1 when all rewards are for the Right response, and 0 when all are for the Left, and a is a longterm memory parameter. This linear model explains some apparent paradoxes in earlier reports of memory effects in two-armed bandit experiments. Nevertheless, closer examination of the details of preference changes within each experimental session showed several kinds of non-markovian effects. The most important was a regression at the beginning of each experimental session towards a preference characteristic of earlier sessions (spontaneous recovery). This effect, but not a smaller, less reliable non-markovian reminiscence effect, is consistent with a very simple rule, namely that the effect on preference of each individual reward for a Right or Left response is inversely related to how long ago the reward occurred. Thus, animals learn to prefer the rewarded side each day because these rewards are recent; but they regress to earlier preferences overnight because the most recent rewards become relatively less recent with lapse of time.

Effect of Ability Level upon Retention of a Balance Skill after Two Years

The interactive effects of ability level and a 2-yr. layoff upon forgetting of a balance skill were examined. In an earlier study, Carron and Marteniuk (1970) tested 150 male Ss on the stabilometer, ranked them according to their initial ability level and selected 3 groups ( N = 20 in each) which consisted of the best performers (high-ability group), the worst performers (low-ability group) and the performers scoring closest to the median (average-ability group). No differences were evident following a 1- or a 7-day layoff but a significant interaction (resulting from a reminiscence effect for the high-ability group combined with forgetting for the average- and low-ability groups) was present following a 14-day layoff. Ss from the three ability groups were retested (high ability, N = 9; average ability, N = 8; and low ability, N = 13). While all groups showed forgetting and the high-ability Ss had a considerably smaller percentage of forgetting (23.8% for the high-ability versus 41.5% and 46.0% for the average- and low-ability groups), the interaction between ability groups and forgetting just failed to reach significance.