Brain Machine Interface Control Through Neurofeedback Guided Beta Rhythm Modulation

Spencer Torene¹, Frank Guenther², Jason Ritt³
¹Boston University, ²Boston University, ³Boston University

Brain-machine interfaces (BMIs) are an emerging medical treatment for many forms of motor impairments, including paralyzed and “locked-in” patients. Most BMI studies concentrate on decoding intended movement parameters within experimenter defined control periods, when it is assumed the subject wants to move. However, the ability of the user to volitionally start and stop BMI control periods is also a critical requirement for BMI usability. Volitional control of the beta sensorimotor rhythm (SMR) could be used as a “brain switch” by the subject to indicate when the BMI decoder should start or stop decoding. We present a proof of concept that mice can use graded neurofeedback to increase beta SMR power in the motor cortex. Freely moving mice in neurofeedback task received water reward by controlling an auditory pitch cursor using a modified beta power measurement from vibrissal motor cortex. Preliminary results show that: (1) behavioral performance improves over sessions; (2) post-hoc spectral analysis of the local field potential confirms that online beta power estimation is consistent with a true beta SMR; (3) beta activity is correlated to EMG activity, confirming the beta SMR’s association with motor function; (4) mice are able to volitionally modulate beta power, shown by contrasting pre- and post-brain sessions; and, (5) the SMR brain switch decoder model has an extremely low false positive rate of ~0.1% predicating elevated EMG activity.

Keywords: BMI, Beta SMR, Rodent Whisker System


Conference Sponsors