The goal of this work is to identify and understand the needs of users who collaborate with an AI or data-driven tool, in order to identify opportunities and applications to better design new tools. There is a great deal of work that seeks to develop AI assistants and collaborators that can assist human users in various tasks. However, much of this work focuses on simple tasks (such as email classification) and, further, very little of it seeks the input of the users in designing the tools. As a result, there is little understanding of what the application contexts and design needs are for AI assistants that collaborate with human users in complex, higher-risk contexts. The purpose of this research is to gain a better understanding of these criteria in the context of dynamic-decision-making domains, specifically, esports games and engineering design.

Team: Erica Kleinman, Murtuza Shergadwala, Magy Seif El-Nasr

Publications:

Kleinman, E., Shergadwala, M. N., & Seif El-Nasr, M. (2022, April). Kills, Deaths, and (Computational) Assists: Identifying Opportunities for Computational Support in Esport Learning. In CHI Conference on Human Factors in Computing Systems (pp. 1-13).

Shergadwala, M. N., & Seif El-Nasr, M. HUMAN-CENTRIC DESIGN REQUIREMENTS AND CHALLENGES FOR ENABLING HUMAN-AI INTERACTION IN ENGINEERING DESIGN: AN INTERVIEW STUDY. In International Design Engineering Technical Conferences and Computers and Information in Engineering Conference 2021, online, (pp. accepted). American Society of Mechanical Engineers

To Collaborate or Get Involved: Contact the project lead (Erica) at emkleinm@ucsc.edu

The goal of this work is to generate a formal understanding of how players interact with, interpret, and make meaning from visualized gameplay data. The current state-of-the-art of data-driven tools for games has focused predominantly on evaluating design and usability. As such, there is currently no formal understanding of how players interact with and extract meaning from data, which would inform the design of future systems. An initial paper explores this question in the context of Stratmapper, an interactive, spatiotemporal visualization system for gameplay data, and seeks to generate a unique taxonomy specific to the domain. A follow-up paper expands our understanding of meaning-making to process visualizations.

Team: Erica Kleinman, Nikhita Preetham, Zhaoqing Teng, Andy Bryant, Murtuza Shergadwala, Magy Seif El-Nasr

Publications:

Kleinman, E., Preetham, N., Teng, Z., Bryant, A. and Seif El-Nasr, M. (2021). "What Happened Here!?" Towards a Taxonomy for User Interaction with Spatio-Temporal Game Data Visualization. Proc. ACM Hum. Comput. Interact. 5, CHIPLAY.

Kleinman, E., Villareale, J., Shergadwala, M., Teng, Z., Bryant, A., Zhu, J., and Seif El-Nasr, M. (2022). Towards an Understanding of how Players Make Meaning from Post-Play Process Visualizations. IFIP-ICEC.

To Collaborate or Get Involved: Contact the project lead (Erica) at emkleinm@ucsc.edu

This body of work explores how Self-Regulated Learning occurs in the context of esports. An initial study highlighted similarities and differences across skill levels. A new study highlights how third-party assistants support SRL processes across three phases of the Cyclical Phase Model of SRL and identifies opportunities to improve that support in future development. We are currently exploring opportunities to expand this work into other theories of SRL, analyses of other tools, and other games. Part of this work was conducted in collaboration with SenpAI.GG.

Team: Erica Kleinman, Magy Seif El-Nasr

Collaborators: SenpAI.GG

Publications:

Kleinman, E., Gayle, C., & Seif El-Nasr, M. (2021). "Because I'm Bad at the Game!" A Microanalytical Study of Self-Regulated Learning in League of Legends. Frontiers in Psychology, Educational Psychology.

Kleinman, E., Habibi, R., Yao, Y., Gayle, C., and Seif El-Nasr, M. (2022). A Time and Phase for Everything" - Towards A Self Regulated Learning Perspective on Computational Support for Esports. CHIPlay.

To Collaborate or Get Involved: Contact the project lead (Erica) at emkleinm@ucsc.edu

This work examines whether instilling a virtual agent with fairness and social intelligence would positively influence users' perceptions, attitudes, and behaviors towards the virtual agent. We focus on two facets of fairness: procedural fairness(explaining the procedures used to plan, implement, and reach a decision), and informational fairness (providing adequate justification for a decision). Using social intelligence theory, we will develop tactics that explain agents’ reasoning and enhance users’ sense of autonomy.

Team: Erica Kleinman, Atieh Kashani, Reza Habibi, Sai Siddartha Maram, Magy Seif El-Nasr

Collaborators: Nagwan R. Zahry, Assistant Professor, Strategic Communications, University of Tennessee at Chattanooga

Publications: To Come

This work builds on previous studies and seeks to create a stronger understanding of players' experiences and needs with and from AI assistants for esports. The results of this work will help derive more concrete design requirements and best practices for computational assistants for esports.

Team: Erica Kleinman, Magy Seif El-Nasr

Collaborators: Mobalytics

Publications: To Come

To Collaborate or Get Involved: Contact the project lead (Erica) at emkleinm@ucsc.edu