Designing for Awareness

This project explores solutions to the issue of lack of situational awareness in mobile messaging. We aim to understand how we can leverage virtual assistants to inform availability of their users. At the same time, another focus of this project is to make functions of virtual assistants open and understandable to their users especially when these technologies involve AI and Machine Learniing. This project encompasses multiple research aspects including user modeling, technology design, user-centric privacy, human-agent interactions and explainable AI.
More information about this project can be found here

Publications related to this project:

Jain, Pranut, Rosta Farzan, and Adam J. Lee. "Context-based Automated Responses of Unavailability in Mobile Messaging." Computer Supported Cooperative Work (CSCW) 30.3 (2021): 307-349.

Jain, Pranut, Rosta Farzan, and Adam J. Lee. "Adaptive modelling of attentiveness to messaging: A hybrid approach." Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization. 2019.

Jain, Pranut, Rosta Farzan, and Adam J. Lee. "Are You There? Identifying Unavailability in Mobile Messaging." Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems. 2019.

Team Communication in Games

This project aims to undertand human player's communication behaviors in collaborative competitive eSports games. We are particularly interested in player's implicit communications during the game, the intention of these communication, and the means of such communication (e.g. using Pings, or a sequence of actions).

Community Science for Air Quality Monitoring in Pittsburgh

Given its industrial past and present as a 'Steel City', Pittsburgh's air quality has a significant public health impact on its residents. We are working with community-based organizations and using citizen-science principles to deploy a network of low-cost air quality monitors to make environmental data more legible and accessible to communities. In collaboration with a local environmental non-profit organization, UpstreamPgh, we designed and deployed a ‘Social Sensor Network’ of low-cost environmental sensors in the Nine Mile Run Watershed in Pittsburgh, PA for a 3-month pilot project. A network of sensors automatically reporting air quality and other metrics was virtually connected to a closed social media platform (Slack), involving a group of tech-savvy participants with a strong interest in environmental issues. The project was designed to include automatic, active, interactive, and passive forms of engagement to give participants various avenues to contribute to data collection, exploration, and interpretation. We are currently working on expanding the network to other communities, and designing workshops to use the data collected to tell data stories connecting local issues to global issues.

Participatory design

"Co-creation" is any act of creativity undertaken by two or more people (Sanders & Stappers, 2008). In the world of human-computer interaction, a very common application of co-creation is that of participatory design, a process in which researchers and users work together over time to create systems. The process enables allows the final product to incorporate the expertise of researchers, who are often experts in a relevant subject, with that of users, who are experts in the perspective of a system user. Participatory design has been shown to lead to more equitable designs which enjoy more acceptance and are more effective in their purpose.

In this work, we examine a unique configuration of stakeholders: researchers working alongside parent/teenager dyads to design a mobile app for the creation and management of rules for online safety. We apply a unique two-condition design in which teenagers and parents design both individually and together with their dyadic partner, and we manipulate the order in which these phases occur. In doing so, we're able to observe how ideas flow between design phases, better understanding how each participant attends to ideas to their partner while contributing their own. We also apply inductive coding to the themes of the generated designs in order to better understand parent and teenager perspectives on the design of such interventions. Last we examine how the act of co-creation contributes to shared understanding within the dyad and present recommendations for future researchers who seek to conduct parent/teen dyadic design, whether it's in a single co-creative session or a longer-term participatory design context.