Age? It's in the Game: An Exploratory Study on Detection of Cognitive Aging through Card Games
summary
Built a computer-vision auto-labeler that extracts play traces from commercial FreeCell and engineers features; trained a Logistic Regression classifier on data from 52 participants 130 rounds to test digital biomarkers of cognitive aging. Key signals: average think time, game result, move variability. Model separates young and older groups (AUCs ~0.81–0.88, combined AUC 0.734, accuracy 0.63) but fails on the middle-aged bin. Toolkit works for FreeCell but needs more data and clinical batteries to generalize.
AI-written summary · about
chapters (9)
- 0:00Start
- 0:36COGNITIVE AGING
- 1:56GAMES FOR DIGITAL BIOMARKERS
- 2:26PROBLEM
- 2:43RESEARCH QUESTION
- 4:45CORRELATION WITH AGE GROUP
- 5:34ROCCURVE
- 7:13TAKE AWAY
- 11:43SECRET SLIDES
related talks
21mExamining the relationship between brain activity and video game performance
06mMachine Learning for Phone-Based Relationship Estimation: The Need to Consider Population ...
18mLongevity & Lifestyle
21mHow to find the right audience for your game
02h 02mUsing Open Game Data to Understand Game-Based Learning
23mKeynote: Digital Psychometrics and it's Future Effects on Technology
07mBig Data and Cardiac Rehabilitation
49mData Science in Gaming