Please note: This PhD seminar will take place online.
Valerie Platsko, PhD candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Kate Larson
Social networks, from institutional structures governing communication to social media, shape communication choices. While misinformation has been a common focus of research on networks, rational agents may have incentives to withhold information even if that information, while potentially imprecise or not fully-informative, is guaranteed to be truthful or can be easily verified. Strategic choices to withhold information may lead to undesirable outcomes if too little information is shared, even in the absence of active misinformation. Our work considers how aspects of social network design affect the ability of rational agents to share verifiable information in the presence of competing incentives.
We show that network structure, as well as agents’ knowledge of the network structure (such as the visibility of friends-of-friends in an online social network) affects the ability of rational agents to communicate in equilibrium, as well as the likelihood that a collective decision made based on shared information accurately reflects agents’ informed preferences.