Citation

Wilde GJS, Gerszke D, Paulozza L. Transp. Res. F Traffic Psychol. Behav. 1998; 1(1): 77-93.

Copyright

(Copyright © 1998, Elsevier Publishing)

DOI

10.1016/S1369-8478(98)00005-9

PMID

unavailable

Abstract

This article describes various procedures for assessing whether the amount of risk a person takes in the execution of a task is too great, too small or just right for the purpose of net benefit maximization. These procedures attempt to mimic human decision making in traffic where the actual accident risk taken by road users may or may not be optimal when overall costs and benefits are being considered. In traffic, however, real costs and benefits are difficult to assess quantitatively. For that reason, laboratory experiments were used in which subjects were asked to perform several tasks with known costs and benefits. These tasks involved two types of item content, (a) general knowledge and (b) magnitude estimation of surface areas, and three different response formats, namely 'Brinkmanship', 'Guessmanship' and 'Narrow Escape'. All three tasks allow the calculation of quantitative measures of actual risk taking and the extent and direction of deviations from optimal risk taking. Subjects engaged in two separate experiments competed for monetary prizes. After a number of trials on the task that was administered first in each experiment, subjects in the experimental groups received detailed information on their degree and direction of deviation from optimal risk taking and it was explained to them how they would have to alter their risk-taking strategy in order to optimize it and thus obtain a larger number of competitive points. The subjects were then asked to respond to more items in the same task. Subsequently, additional tasks were performed in the same test session. These were either similar or different in item content and always different in terms of response format. Control subjects completed the same tasks in the same order without, however, receiving risk-competency training. The combined data appear to indicate that (a) risk-optimization training helps individuals to become more competent risk takers in a given task, and that (b) this enhanced competency generalizes to other tasks, with performance feedback being important for the development of a consistent risk-taking strategy.