The effects of expectations of loss of control and experiences of success and failure on the development of learned helplessness

Learned helplessness (Seligman, 1975; Abramson, Seligman, & Teasdale, 1978) has recently gained prominence as a model of depression The theory states that an expectation of loss of control over reinforcements results in cognitive and motivational deficits, and, in certain cases, reduced self-esteem and increased depressed affect. While the theory has received some experimental support, at least three major areas of uncertainty remain in the theory: 1) Perceptions of loss of control have never actually been shown to cause helplessness; 2) most studies supporting the theory have confounded an experience of failure with an expectation of loss of control; 3) some studies have found facilitated cognitive and motivational functioning following an experience of loss of control. The present study was designed to investigate these three areas of uncertainty.

A total of 165 students were randomly assigned to 11 groups, ten of which were given experimental treatments and one of which was a No Treatment group. Students in the experimental groups received an experimental treatment based on a combination of two independent variables: a) Success/Failure; b) Level of Expectation of control over future reinforcements. The Success/Failure manipulation consisted of students being given five concept-formation problems. Students in the Failure condition were induced to fail each problem, while students in the Success condition were allowed to solve each problem. The Level of Expectation manipulation consisted of stating one of five probability numbers to each student. This probability number was presented as the students' best chance of scoring well on an anagram task which was given after the concept-formation task. All students were told that they had a 20% chance of scoring well on the task if they did not try hard to score well, and either an 80%, 75%, 50%, 25%, or 20% chance of scoring well on the task if they did try hard to score well. Only the group being told that they had a 20% chance of scoring well if they did try hard to score well and a 20% chance if they did not try hard to score well fit Seligman's (1975) definition of loss of control. The Success/Failure and Level of Expectation variables were crossed, resulting in a 2(Success, Failure) X 5(80%, 75%, 50%, 25%, 20% Level of Expectation) factorial design. Students in the No Treatment group were given neither of the experimental manipulations.

Following the experimental manipulations, students were given an anagram task to measure cognitive and motivational functioning along with checklists to measure depressed affect and self- esteem. Experimental predictions hypothesized that both failure and expectation of loss of control would produce impaired cognitive and motivational functioning, decreased self-esteem, and increased depressed affect. A combination of success and a high level of expectation of control over reinforcements was predicted to produce facilitated cognitive and motivational functioning, increased self-esteem, and decreased depressed affect.

Results showed that the Level of Expectation variable had no effect on cognitive or motivational functioning, self-esteem, or depressed affect. Decreased self-esteem was found when students experiencing failure were compared with students who experienced success. Increased depressed affect was found when students experiencing failure were compared with students who were not given the experimental manipulations. In addition, no facilitation was produced from any combination of the independent variables.

These results conflict with the learned helplessness theory and suggest that failure, rather than an expectation of loss of control, produces decreased self-esteem and increased depressed affect. Thus, it is possible that because studies producing learned helplessness commonly confound failure with loss of control, Abramson et al. (1978) chose the wrong variable as the variable responsible for producing learned helplessness. Failure is different from an expectation of loss of control as it includes an experience of reduction in the amount of positive reinforcement.

The present study produced no cognitive or motivational deficits. One explanation for this result is that anagrams are inadequate measures of cognitive and motivational functioning. Additional research, utilizing improved dependent measures and refined methodology, with the intention of continuing to investigate the effects of failure and loss of control as independent variables is suggested.