John T. Lanzetta

EFFECTS OF PROBLEM UNCERTAINTY AND PRIOR AROUSAL ON PRE-DECISIONAL INFORMATION SEARCH

Fifty-four undergraduate Ss at the University of Delaware were presented sequentially 10 decision problems. Prior to making a decision on a problem, S had the option of receiving up to five additional information items. Independent variables were the initial uncertainty of the decision problem and prior “emotional” arousal. Dependent measures were the number of information items taken, the utilization of time resources, and Ss arousal level (GSR). Neither independent variable affected the amount of information acquisition prior to decision, but differences in problem uncertainty within the high uncertainty conditions were negatively related to information search (r = -.706). Decision time increased linearly with problem uncertainty and decreased across sequential problems. In general, decision-making became more facile with increasing experience.

Choice among Four Alternative Gambles as a Function of Monetary Level of Play

In a series of earlier studies by Lanzetta and Kanareff (1961, 1962), Ss were presented with a decision situation which permitted the acquisition of additional items of information prior to decision. Although the contents of these information items appeared relevant to the making of a better decision, rhey in actuality had nothing to do with the probabiliry of obtaining a fixedsize payoff. Instead, probability of payoff was pre-programmed to be a positive linear function of the number of information-seeking responses (queries) made. Thus, amount of information-seeking pe7 se, and not effective decision-making, was made instrumental in obtaining the payoff. Under a variety of cost-payoff conditions, it was observed that Ss failed, in general, to seek the maximum amount of information available, which would have yielded them the greatest probabi l i ty of payoff a n d t h e m a x i m u m expec ted profi t . T h e results thus appeared to be discrepant with any information-seeking formulation based on the principle of expected-value maximization. Further, a highly consistent trend was observed for the mean number of queries per trial to decrease as the monetary level of play, i.e., the absolute amounts of cost per query and size of payoff, increased. The higher the absolute cost of additional information, the lower the frequency with which it was acquired, although the information was equally instrumental under all conditions in obtaining a payoff, and the payoffs were increased commensurably under conditions of high information cost so as to compensate for the higher cost. In view of previous findings by Edwards (1953) and others with respect