The Relationship of Life Stress to Problem Solving: Task Complexity and Individual Differences

Abstract

To examine the hypothesis that a fixed-capacity information-processing model (Kahneman, 1973) can be used to explain the relationship between problem-solv ing errors and life stress as moderated by state anxiety and private body conscious ness (PBC; Miller, Murphy & Buss, 1981), we administered a series of simple (two-term) and more complex (three-term) analogies to undergraduate subjects. The results indicated that performance on the two-term problems was unrelated to the individual-difference variables, to life stress, or to the strategies employed. On three-term problems, however, students high in life stress, state anxiety, and private body consciousness performed more poorly and were more likely to employ nonsystematic scanning of alternative solutions to the analogies. Moreover, non-systematic scanning appeared to mediate the relationship between the individual-difference variables and performance. The data fit a model in which life stress and state anxiety have the potential to consume scarce cognitive resources. People who are highly sensitive to their own autonomic responses (high-PBC) appear more likely to show performance decrements and to use nonoptimal solution strategies as compared with people less aware of their own autonomic responses (low-PBC). There is substantial evidence that psychological stress, measured with life-events checklists, is negatively correlated with physical (Cohen, Tyrell, & Smith, 1991) and psychological (Nezu & Ronan, 1985) health. The relationship of life stress to cognitive operations, such as problem solving and decision making, however, has received relatively little attention. Taylor and Schneider (1989) have suggested that people "re run" or simulate negative events as a means of coping with past stres sors. This rerunning of past events has at least two functions: to develop plans for solving life problems and to regulate emotions. In both cases, We gratefully acknowledge the assistance of John Klein who programmed the computer ized task and of Susan M. Andersen and three anonymous reviewers for their many helpful comments on earlier drafts of this article.