Measuring Cognitive Abilities in the Wild: Validating a Population-Scale Game-Based Cognitive Assessment.

Area of Science:

• Cognitive Science
• Psychometrics
• Human-Computer Interaction

Background:

• Traditional cognitive assessments are time-consuming and lack ecological validity.
• There is a need for rapid, accessible, and engaging methods for cognitive phenotyping.
• Citizen science platforms offer opportunities for large-scale data collection in naturalistic settings.

Purpose of the Study:

• To develop and validate Skill Lab, a novel game-based tool for simultaneous cognitive ability assessment.
• To evaluate the feasibility of rapid, in-the-wild cognitive phenotyping.
• To establish a foundation for population-scale cognitive benchmarking and personalized diagnostics.

Main Methods:

• Development of Skill Lab, a game-based suite with mini-games and established cognitive tasks.
• Large-scale validation (N = 10,725) on a citizen science platform.
• Construction and validation of predictive models for eight cognitive abilities based on in-game behavior.

Main Results:

• Skill Lab models reliably predict individual cognitive abilities from in-game behavior.
• The game-based measures are five times faster than traditional tasks.
• Findings replicate age-related cognitive decline in a large sample (N = 6369).

Conclusions:

• Game-based cognitive assessment is feasible for rapid, in-the-wild evaluation.
• Skill Lab demonstrates potential for broad applications in personalized learning, employment, and psychiatry.
• This approach advances population-scale cognitive benchmarking and individualized mental health diagnostics.