Brain-computer interface-based neurofeedback training enables transferable control of cortical state switching in...

Area of Science:

• Neuroscience
• Cognitive Science
• Brain-Computer Interfaces

Background:

• Behavioral flexibility depends on dynamic neural activity and cortical state transitions.
• The capacity for humans to deliberately learn and generalize control over these neural dynamics is not well understood.

Purpose of the Study:

• To investigate whether humans can volitionally learn to control neural dynamics related to cortical state transitions using a brain-computer interface (BCI).
• To determine if this learned control can generalize beyond the training context and improve behavioral flexibility.

Main Methods:

• Participants trained with a real-time BCI providing feedback on sensorimotor activity.
• Comparison with a double-blind sham control group.
• Latent-state analysis to examine neural dynamics during transitions.
• Measurement of reaction times for muscle contraction and relaxation.

Main Results:

• Trained participants demonstrated volitional modulation of sensorimotor oscillations without BCI.
• Enhanced interregional phase coupling and steeper broadband spectral slope in the medial frontal cortex were observed during transitions.
• Training-induced neural reorganization correlated with improved performance, indicated by reduced reaction times.

Conclusions:

• Learned control over sensorimotor cortical state transitions is achievable through BCI training.
• This learned control enhances behavioral flexibility and generalizes beyond the training context.
• The findings highlight the potential of BCI for modulating neural dynamics and improving cognitive functions.