Virtual teams are increasingly common in today's organizations, yet they often make poor decisions. Teams that interact using text-based collaboration technology typically exchange more information than when they perform the same task face-to-face, but past results suggest that team members are more likely to ignore information they receive from others. Collaboration technology makes unique demands on individual cognitive resources that may change how individual team members process information in virtual settings compared to face-to-face settings. This experiment uses electroencephalography, electrodermal activity, and facial electromyography to investigate how team members process information received from text-based collaboration during a team decision-making process. Our findings show that information that challenges an individual's prediscussion decision preference is processed similarly to irrelevant information, while information that supports an individual's prediscussion decision preference is processed more thoroughly. Our results present neurological evidence for the underlying processes of confirmation bias in information processing during online team discussions.
Much prior research on virtual teams has examined the impact of the features and capabilities of different communication tools (the nature of communication) on team performance. In this paper, we examine how the social structures (i.e., genre rules) that emerge around different communication tools (the nurture of communication) can be as important in influencing performance. During habitual use situations, team members enact genre rules associated with communication tools without conscious thought via automaticity. These genre rules influence how teams interact and ultimately how well they perform. We conducted an experimental study to examine the impact of different genre rules that have developed for two communication tools: instant messenger and discussion forum. Our results show that in habitual use situations, these tools triggered different genre rules with different behaviors, which in turn resulted in significantly different decision quality. We used heightened time pressure as a discrepant event to interrupt the automatic enactment of habitual genre rules and found that users adopted similar behaviors for both tools, which resulted in no significant differences in decision quality. These findings suggest that the automatic enactment of genre rules for a communication tool may have as powerful an effect on behavior and performance as the actual features of the tool itself. We believe that our results, taken together with past research showing the effects of social structures on communication, call for the expansion of task–technology fit theories to include the role of social structures in explaining the use of and performance from communication tools.