The Complexity Construct in Political Psychology: Personological and Cognitive Approaches
Abstract
Measures of the cognitive complexity of leaders have been used to infer the flexibility, open-endedness, and information-orientation of their decision making in international and nonstate confrontations. At present, there are two major methods of "assessment at a distance" used in this context. One uses computer scoring to develop personality profiles of leaders; the other uses a more labor- and time-intensive human scoring system to track changes in the target's thinking to predict the outcome of a particular confrontation. If computer scoring were able to make event-specific predictions, the savings in time and work would be substantial. This study compared the two systems to establish the following: (1) whether the computer-scored system could replace human scoring; and (2) using the example of the South Ossetia War between Georgia and Russia, which method was a better predictor of rising and falling tension. The data confirmed the relevance of integrative complexity measurement in a new context, that of an ongoing confrontation with changing levels of tension, up to and including war, between a major and a minor national power. The correlation between scores from the two methods was low; at high levels of cognitive complexity, it was essentially zero. The human scoring of integrative complexity, which tracks changes in complexity over the duration of a particular event, was closely tied to the course of the confrontation. The computer scoring of cognitive complexity, which profiles complexity as a stable personality characteristic, was not. Thus, although computer scoring has significant advantages in cost and time, it does not accomplish the same goals. The data indicate that the negative trade-off between speed and accuracy is serious enough to opt for the more laborious human tasking if the goal is the prediction of international crisis outcome.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 2010
- Accession Number
- ADA523334
Entities
People
- Peter Suedfeld
Organizations
- University of British Columbia