Paradigm Shift in Sensorimotor Control Research and Brain Machine Interface Control: The Influence of Context on...

Area of Science:

• Neuroscience
• Sensorimotor Systems
• Computational Neuroscience

Background:

• Neural activity in the primary motor cortex (M1) correlates with movement kinematics and dynamics.
• Recent work indicates M1 and primary somatosensory cortex (S1) activity are modulated by context, like reward value, during movement and observation.

Purpose of the Study:

• To investigate reward modulation in M1 during brain-machine interface (BMI) control.
• To determine if reward level alters neural tuning to kinematic and dynamic variables in M1.
• To explore the potential for improved BMI performance by accounting for reward context.

Main Methods:

• Recording neural activity from M1 during tasks involving movement and BMI control.
• Analyzing neural tuning functions in relation to kinematic, dynamic, and reward variables.
• Classifying reward expectation from M1 activity on a per-movement basis.
• Gating multiple linear BMI decoders using classified reward expectation for offline performance evaluation.

Main Results:

• Reward level significantly changes the neural tuning functions of M1 units for kinematic and dynamic variables.
• Reward-modulated neural activity is present and detectable during BMI control.
• Classifying reward expectation from M1 activity allowed for gating of BMI decoders.
• Gating BMI decoders based on reward expectation improved offline performance.

Conclusions:

• M1 neural activity is modulated by reward context, influencing tuning to movement variables.
• This reward modulation persists during BMI control, offering opportunities for decoder enhancement.
• Accounting for reward context in BMI design can lead to more robust and accurate decoders.
• Future work will focus on integrating this reward-based gating system into online BMI performance.