Brain Cortical Area Characterization and Machine Learning-Based Measure of Rasmussen's S-R-K Model

Area of Science:

• Neuroscience
• Cognitive Psychology
• Human Factors Engineering

Background:

• The Skill, Rule, and Knowledge (S-R-K) model categorizes human behavior based on cognitive control needs.
• Current S-R-K model lacks quantifiable metrics, hindering objective performance measurement.
• This study addresses the need for neurophysiological characterization of the S-R-K model.

Purpose of the Study:

• To neurophysiologically characterize the S-R-K model by analyzing operator cerebral cortical activity.
• To develop a machine learning model for estimating cognitive control behaviors.
• To establish objective metrics for differentiating Skill, Rule, and Knowledge levels.

Main Methods:

• Participants performed tasks simulating Skill (tracking), Rule (navigation), and Knowledge (unfamiliar) conditions.
• Electroencephalogram (EEG) was recorded during task execution.
• Global Field Power (GFP) in EEG frequency bands and Brodmann areas (BAs) were analyzed.

Main Results:

• Distinct S-R-K patterns were identified in cerebral cortical activity.
• Machine learning models were built using EEG features and BAs.
• The models successfully estimated participants' cognitive control behaviors.

Conclusions:

• Objective measurement of S, R, and K levels is possible through brain activation analysis.
• Findings align with existing literature on cognitive functions at different control levels.
• This research provides a neurophysiological basis for the S-R-K model.