The emergence of communication structure in decision-making groups: An investigation of task complexity and variability of perceived expertise among members

An experiment was conducted to investigate the influence of task complexity and variability of perceived expertise on the pattern of communication in decision-making groups. It was hypothesized that groups would tend to use a relatively centralized communication structure when performing simple tasks and a relatively decentralized structure while performing more complex tasks. It was also hypothesized that groups with high variability in perceived expertise of group members would develop a centralized structure, whereas groups with low variability in perceived expertise of group members would tend to develop a decentralized structure of communication. In addition, the effect of task complexity on the emergence of communication structure was expected to depend on the level of perceived expertise variability. Specifically, differences in the degree of centralization between simple and complex tasks under conditions of high perceived expertise variability were expected to be much less pronounced than differences in the degree of centralization between simple and complex tasks under conditions of low perceived expertise variability. These predictions were tested using a 2 (low variability of perceived expertise versus high variability of perceived expertise) x 2 (low-task complexity versus high-task complexity) factorial design. Task complexity was varied between groups by administering a task that allowed the manipulation of goals, pathways to these goals, and the information processing requirements inherent in the task. Perceived expertise variability was manipulated by providing false feedback to group members to create either the perception of highly discrepant task expertise among members, in the high-variability condition, or the perception of near-equal expertise among members, in the low-variability condition. As predicted, results communication in groups in low-task complexity conditions tended to be more centralized than communication in high-task complexity conditions. Also, communication in groups in high perceived expertise variability conditions tended to be more centralized than communication in low perceived expertise variability conditions. However, the anticipated Task Complexity x Variability of Perceived Expertise interaction, with the effect of task complexity on the emergence of communication structure expected to depend on the level of perceived expertise variability, was not supported.