Human-AI Interaction

My current research examines human interaction with artificial intelligence. Recently, this involved a large-scale study of human interaction with AI agents with social awareness. More details about this work can be found at

Context-Aware Computing

Computational awareness of the broad context of interaction is essential for providing support that goes beyond simple recommendations. Using a variety of unobtrusive methods, we create models of context that enable systems to reason about what tasks a user is trying to perform, what information a user may need, and what type of support will be most useful.

Visualization of Activity over Time

My graduate school research examined interactive visualization of activity over time.

The ability to record human activity has been extended it into situations that have typically not been accessible to digital technology, enabling examination of fine detail of action captured in real-world settings. This recording can be for personal purposes, to understand one’s own activity for reflection or task resumption, such as perusing one’s web browser history of. This recording can also be for research purposes, to better understand human behavior when performing real-world, situated tasks.

A promising way to improve our ability to understand all of this information is interactive information visualization, taking advantage of the the interactive possibilities made possible by modern computers to show trends in data and invite people to explore data in a variety of ways.

My research seeks to answer two complementary questions: How should multiple streams of activity over time be visualized? and How does interaction with this visualization affect one’s understanding of the data?