Decoding Human Cognitive Control Using Functional Connectivity of Local Field Potentials.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Cognitive Science

Background:

• Cognitive control impairments in mental illness lack effective treatments.
• Task engagement is crucial for augmenting decision-making via closed-loop brain stimulation.
• Real-time detection of task engagement is needed for proactive intervention.

Purpose of the Study:

• To develop and validate a neural decoder for detecting task engagement.
• To differentiate task engagement from background brain activity.
• To enable timely interventions for cognitive control disorders.

Main Methods:

• Functional connectivity patterns from local field potentials were extracted.
• Principal Component Analysis (PCA) was used on connectivity patterns.
• Subject-specific linear Support Vector Machine (SVM) classifiers were trained.

Main Results:

• Task engagement was differentiated from background brain activity with 89.7% median accuracy.
• The neural decoder detected task engagement in under 2 seconds.
• The system's speed can be further enhanced with application-specific hardware.

Conclusions:

• A neural decoder can accurately detect task engagement from brain activity.
• Real-time detection facilitates potential closed-loop interventions.
• This approach offers a promising avenue for novel mental illness treatments.