Guiding functional reorganization of motor redundancy using a body-machine interface

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Neuroscience

Background:

• Body-machine interfaces (BMIs) translate body movements into commands for external devices.
• Users of BMIs face challenges in understanding interface operation and efficiently controlling devices due to redundant motion signals.
• This study investigates the connection between understanding interface mechanics and improving device control efficiency.

Purpose of the Study:

• To determine if aligning body movement spaces with interface-encoded spaces improves BMI control.
• To investigate how this alignment facilitates redundancy resolution and enhances device control efficiency.
• To explore the implications for controlling assistive devices and rehabilitation.

Main Methods:

• Twenty participants controlled a computer cursor using arm movements with inertial sensors.
• An adaptive mapping strategy was compared to a fixed mapping, with mappings updated based on principal components of user movements.
• Tasks included reaching and tracking, with movement variance distribution analyzed.

Main Results:

• Participants rapidly improved reaching times and learned to distribute movement variance.
• Adaptive mapping reduced the mismatch between user movement subspaces and the interface map's potent subspace.
• This optimization enhanced movement efficiency without compromising task generalization.

Conclusions:

• Aligning the interface's potent subspace with the user's movement subspace enhances BMI control efficiency.
• This principle is vital for optimizing the control of assistive technologies.
• For rehabilitation, interfaces should be adapted to encourage new movement patterns.