Phase Sensitivity of Bipolar EEG Training Protocols

Abstract

Introduction. When performing bipolar EEG training the reward waveform is dependent on the phase relationship between the two sites at the center frequency of the reward band. It is also driven by the instantaneous amplitude at the two sites. Effectively, the training rewards an increase in differential mode activity while discouraging common mode activity. This follows from the fact that the in-phase component of the signal is suppressed in the differential amplifier, and only the out-of-phase component of the waveforms is propagated and rewarded. These relationships can be represented visually with a vector calculation. Method. Two vectors representing the signals at the two sites at a given frequency were combined with varying phase angle and for several representative magnitudes to yield the net output amplitude. Surrogate data were used to obtain distributions illustrating the above functional relationships. Results. Varying phase angle between the two vectors has the greatest effect on the resultant magnitude when there is a high degree of amplitude parity. The influence of phase is more modest when comodulation is low. In cases of high amplitude disparity bipolar training reduces effectively to single-site or referential training. Discussion. The theoretical rationale behind bipolar training as a primary approach to self-regulation is modest at this time. The present modeling of signal interaction illustrates that phase relationships play a governing role in bipolar training in the general case. That dependence is modulated by the degree of comodulation between the sites. Hence, bipolar training may activate different mechanisms of control depending upon the degree of amplitude parity between the two sites. The incorporation of such thinking will likely lead to a further refinement of treatment protocols.