From mundane details of daily life to tragic warnings of planned suicide, people exchange a massive amount of personal information in social media that defies traditional models of self-disclosure. This project will advance understanding of personal information disclosure in social media contexts and will use this information to develop well-being interventions designed to enhance self-reflection and capacity to productively seek and offer support. Such interventions benefit individuals and society overall by helping people know how and from where to seek support in informal and formal social networks. Knowledge about key drivers of self-disclosure will also be useful in improving the design of other systems that use personal data and will inform public discussions about the use and risks of personal information online. The PIs will work to raise awareness of disclosure benefits and risks through users' participation in experiments, use of systems, and publicizing research results beyond the research community, e.g., by working to move research results into practical applications through the Bronfenbrenner Center for Translational Research at Cornell.

More specifically, this research will yield a multi-theoretical and multi-level model of how individual attributes such as personality and mental health status, technological affordances such as reviewability and audience visibility, and social network properties such as size, density, diversity, and tie strength, individually and in combination, shape anticipated rewards and risks of disclosure and disclosure strategies. These models will incorporate responses by network members into the production cycle of disclosure, and examine how these combined characteristics determine both anticipated and actual outcomes. The researchers will collect examples of actual disclosure behaviors in both text and photos, annotated with information about people's disclosure goals and perceptions, individual characteristics, and social networks, and will use these to validate predictive models of the presence of and responses to disclosure in social media data. These models will help identify when people create meaningful content, which in turn can be used in systems and interventions that support the well-being of social media users, both in the general population and those at risk. These systems and interventions will operate by using disclosed information to facilitate reflection, enrich existing positive psychology interventions, and promote awareness of and effective responses to disclosure and mental health needs.

Agency
National Science Foundation (NSF)
Institute
Division of Information and Intelligent Systems (IIS)
Application #
1405634
Program Officer
William Bainbridge
Project Start
Project End
Budget Start
2014-06-15
Budget End
2020-05-31
Support Year
Fiscal Year
2014
Total Cost
$1,203,821
Indirect Cost
Name
Cornell University
Department
Type
DUNS #
City
Ithaca
State
NY
Country
United States
Zip Code
14850