Culture and Computer-Mediated Communication

Although the Internet provides many new opportunities for interaction with people from across the world, bridging nations via technology does not guarantee that the cultures of the nations involved are similarly bridged. Mismatches in social conventions, work styles, power relationships and conversational norms can lead to misunderstandings that negatively affect the interaction. My students and I have been examining these cultural processes in computer-mediated communication. Our goal is to understand how cultural dimensions influence CMC and to develop tools to reduce the problems that arise in intercultural communication.

Multilingual Communication

This project, with Dan Cosley at Cornell and Hao-Chuan Wang at National Tsing Hua University in Taiwan, focuses on understanding the problems that arise when some members of a group must communicate using a non-native language and on building and evaluating new tools to improve multilingual communication. As part of this project, graduate student Tina Yuan is conducting interview and survey studies of non-native English speakers in the U.S. and graduate students Ge Gao and Bin Xu are investigating ways to enhance machine translation so that it is more useful for interpersonal communication. This project is supported in part by NSF awards #1025425 and #1318899, and by a Google Faculty Gift

Social-computational support of civic engagement in public policy making

This new project (with Claire Cardie, Cynthia Farina, and Gilly Leshed) combines behavioral/social science research with natural language processing and human-computer interaction design to build and evaluate tools to support public policy discussions. It draws on the CeRI (Cornell eRulemaking Initiative) platform developed by Farina and colleagues. Our goals are to draw more people into online policy discussions, to foster useful and satisfying discussion, and to move people toward consensus-building. This project is supported in part by NSF award #1314778.

Collaboration in Intelligence Analysis

This project, with Sara Kiesler at CMU, and Gilly Leshed and Claire Cardie at Cornell, focuses on helping analysts make sense of large amounts of intelligence information. As part of this project, we are conducting laboratory studies of information sharing, developing NLP tools to help analysts make sense of large amounts of data, and building and evaluating new interfaces for collaborative analysis. This project was supported in part by NSF awards #0968450 and #1025184.

Leveraging the Internet to motivate energy-saving behavior

This project combines psychological research on persuasion with communication technologies to motivate people to conserve energy. We have been exploring techniques such as personalized feedback, public commitment, and individually tailored recommendations for action in the context of a website accessed via computer or mobile devices. Our initial work focused on developing a website testbed,, to explore methods of motivating people to conserve energy. We later modified the site for use in the CALSGreen project, in which students and staff in Cornell campus buildings could explore energy saving options and commit to taking action. Most recently, graduate student Xiying Wang has been working on the development and evaluation of a mobile phone application that combines some of the original StepGreen features, such as suggestions for actions, with new collaboration features for families and housemates who need to collectively manage their energy use. The work was initially funded by NSF grants #0745885 and #0803733.

Recently completed projects

Large Scale Collaboration

This project, with Sara Kiesler at CMU and Suzanne Weisband at University of Arizona, examined how people distribute their time, effort, and communication across multiple partners, tasks, and work teams. One line of research, conducted by former CMU doctoral student Peter Scupelli and Yan Xiao and colleagues at the University of Baltimore Medical Center, focused on understanding how features of the physical environment of OR suites influences how people communicate and coordinate using large displays. A second line of research, conducted by CMU doctoral student Aruna Balakrishnan, examined how visual information displays such as social network diagrams can improve collaborative intelligence analysis.

Gestural Communication in Collaborative Physical Tasks

This collaborative project with Jie Yang and Jane Siegel examined the types and functions of gestures in collaborative physical (3D) tasks and develops technologies to allow remote gesturing in video conference systems. We developed the DOVE system that allows remote communicators to draw directly on live video feeds from a local worksite, making it easy for them to point out objects and locations in the workspace. The value of DOVE for collaboration was demonstrated in our 2004 HCI journal article. For more information and initial publications, please see our project website.

Communication and Social Interaction in Online Support Forums

This project, funded in part by the Robert Wood Johnson Foundation Health e-Technologies Initiative, focused on developing measures to evaluate the health benefits of participation in online support forums such as chatrooms, electronic bulletin boards, and email distribution lists. Discourse analysis and other techniques are being used to understand the impact of online support on the development of social relationships, participants’ knowledge of their illness, and attitudes toward treatment.

The Importance of Shared Visual Spaces for Collaborative Tasks

This large project was a collaboration with Robert Kraut, Jane Siegel, and Jie Yang at Carnegie Mellon University, and Susan Brennan at SUNY Stony Brook. The project combined theoretical analysis of the communications requirements of collaborative tasks, stylized and applied laboratory studies of the effects of visual information on communication and performance in collaborative physical tasks, and the development of new video technologies.

The Development of Shared Mental Models and Group Performance

This collaborative project with Javier Lerch, Bob Kraut, and Alberto Espinosa, used lab and applied field studies to understand how groups achieve mental models, the trade-offs between the development of shared mental models and cognitive overload, and the effects of these models on team coordination and overall effectiveness.