2017

Social Network Processes in Collaborative Decision-Making

Presenter: Christina Sirabella

Research Category: Computer and Information Sciences
Student Type: null
Additional Authors: Tina Lee
PI: Christina Sirabella, Christoph Riedl

In many situations, effective decision-making depends critically on individuals’ ability and willingness to share information. Previous research documented failure of groups colocated in the same room to share relevant information leading to a “shared information bias.’’ Increasingly, groups are collaborating virtually through technology mediated communication. Taking this into account, this research project examines collective decision-making using a hidden profile style experiment conducted online.

Individuals are assigned a random color name and set of private (unshared) and public (shared) clues for the online mystery game. They complete a preliminary answer sheet based only on clues they were given, then interact with other individuals via a group chat. After the group chat phase, individuals complete another answer sheet based on clues received through the chat in addition to ones they were given. The accuracy of individuals’ answers and the information shared in the chat is then analyzed.

In a preliminary study, we have collected data from 56 individuals. Of the 21 groups observed, 14% had all 5 participants complete the experiment. Those groups were able to accurately solve at least one of the three components of the mystery. A preliminary analysis suggests that user-dropout is an issue, hindering us from performing complete and accurate analyses of every trial. Going forward, we have changed the format of the game to enhance user experience and increase the rate of retention. We are confident that our future research combined with the data already collected will allow us to provide additional insights for collaborative decision making.