Classification characteristics of fine motor experts based on electroencephalographic and force tracking data

Area of Science:

• Neuroscience
• Cognitive Science
• Machine Learning

Background:

• Expertise development is often studied by comparing expert and novice performance.
• Machine learning offers a data-driven method to analyze complex datasets like electroencephalographic (EEG) activity.

Purpose of the Study:

• To investigate the differences in electroencephalographic (EEG) patterns and force output between experts and novices during motor tasks.
• To determine if machine learning can classify tasks and expertise levels based on these physiological and performance metrics.

Main Methods:

• Applied classification algorithms to electroencephalographic (EEG) data and force output variables from participants performing four force modulation tasks (sinusoidal and steady force tracking) with both hands.
• Distinguished between different tasks performed by individuals and between expert and novice groups.

Main Results:

• Tasks were accurately classified based on EEG patterns and force output in both novices and experts.
• Classification of group membership (expert vs. novice) was at chance level.
• Follow-up analysis revealed highly individual EEG patterns in experts, suggesting specialized central processing developed over time.

Conclusions:

• Continuous practice in a work context fosters the development of highly individual and task-specific central control patterns in fine motor experts.
• While task execution is distinguishable, the general expertise level is not readily identifiable from EEG and force output alone, emphasizing individual neural adaptations.