Computer Science TechTalk Series
When: Wednesday, January 28, 2026
Where: RB 104
Time: 3:00 PM
Title: Can Video Games Measure Cognitive Abilities? Machine Learning Approaches for Connecting Gameplay Data to Cognitive Task Performance
Presented By: Ming Chen – Indiana University Indianapolis
Abstract: Executive function (EF), which encompasses skills such as inhibition, cognitive shifting, and working memory, is central to learning and problem-solving, yet it can be difficult to measure in a way that is both reliable and engaging. In this talk, I will share my dissertation work on game-based measurement: instrumented video games that collect gameplay telemetry while players complete EF-like challenges. I studied three custom EF mini-games (Dragon Tower, Magic Forest, and Lava Cave) alongside established EF tasks (Flanker, DCCS, and N-back) in a repeated-measures dataset of 137 undergraduate participants.
The reasons why simple correlations often understate the relationship between gameplay behavior and EF performance will be explained first, and then I will present a machine-learning pipeline that connects the two. The pipeline (1) applies unsupervised learning to EF task outcomes to identify interpretable performance profiles (e.g., different speed–accuracy tradeoff styles), and (2) evaluates supervised models (KNN and Random Forest) that predict EF outcomes and profile membership from game metrics. The results highlight both promise (e.g., meaningful performance profiles and informative game–task signals) and practical limitations such as sample size, class imbalance, and feature quality.
Actionable takeaways for future game analytics will be provided at the end, including how to design telemetry that supports inference, how to enhance feature robustness, and how to create gameplay tasks that are engaging while still yielding clean measurement data.
Bio: Dr. Ming Chen is a Lecturer in the Media Arts and Science Program and the Human-Centered
Computing Department at Indiana University Indianapolis. She earned her Ph.D. in Educational Psychology from the CUNY Graduate Center and an M.S. in Computer Science from the Georgia Institute of Technology. Her work lies at the intersection of game development, player modeling, and machine learning, with a focus on how gameplay telemetry can be used to evaluate cognitive abilities (e.g., executive function) and support scalable, engaging digital assessments. She teaches game design and development and also develops indie games in her spare time.
Light refreshments will be served