Date of Award

2017

Document Type

Open Access Master's Thesis

Degree Name

Master of Science in Applied Cognitive Science and Human Factors (MS)

Administrative Home Department

Department of Cognitive and Learning Sciences

Advisor 1

Elizabeth Veinott

Committee Member 1

Shane Mueller

Committee Member 2

Adam Melnitsky

Abstract

This thesis compares different plan evaluation techniques in a series of experiments. The Premortem plan evaluation method can help people reduce overconfidence and generate more reasons a plan might not succeed. This research evaluates the extension of the Premortem to shorter planning time periods, evaluates the effectiveness with team generated and executed plans, and compares the use of this technique among individuals and teams. In Experiment 1, 52 Army Cadets operating in teams completed six time-constrained field exercises that required planning, half using the Premortem and half using a standard Military plan evaluation process. When teams used the Premortem they had fewer fouls and less fixation with no change in planning and execution time. In Experiment 2, 72 individual participants from university organizations used the Premortem Method or Worst-Case Scenario Method to evaluate their group’s plan for an engineering task. Results from Experiment 2 indicated that there was no statistically significant difference in the number of reasons and solutions generated between methods. However, the two methods had significantly different distributions of reasons and solutions across categories, indicating that the methods were prompting participants to approach the plan differently. To further examine the relative effectiveness of these two plan evaluation methods, and the influence of group dynamics, Experiment 3 compared the efficacy of the Premortem and Worst-Case Scenario Method amongst groups and individuals in face-to-face settings with a complex and unfamiliar plan. Eighty-two participants generated more reasons with the Premortem Method than the Worst-Case Scenario Method, and groups generated more solutions than individuals. The distribution of reasons was also significantly different across categories, indicating that the underlying mechanisms are changing how participants view the problem and generate reasons. These studies extend prior work by validating that the Premortem is effective in short planning horizons, demonstrating that it works for individuals and teams, and clarifying potential boundary conditions. This research advocates several directions for future research, and suggests possibility of future implementation as a virtual tool or application.

Download

Share

COinS