Sensorimotor EEG operant conditioning: Experimental and clinical effects

Neurophysiological studies in cats have established a functional relationship between waking 12–15 Hz sensorimotor cortex rhythmic EEG activity (the sensorimotor rhythm or SMR) and a similar pattern during sleep, the sleep spindle. Both result from oscillatory thalamocortical discharge involving ventrobasal thalamus and sensorimotor cortex, and both are associated with a state of suppressed motor excitability. Enhancement of the SMR with operant conditioning methods in the cat clearly led to reduced seizure susceptibility. The experimental application of this approach to seizure control in epileptics has resulted in (A) evidence that EEG patterns can be manipulated significantly in man with operant conditioning, (B) suggestive observations concerning a potential component of pathology in epilepsy, and (C) strong preliminary evidence that SMR operant conditioning in epileptics is specifically therapeutic. Current research has focused upon the EEG during sleep in epileptics with primary motor symptomatology. This measure often reveals several hard signs of pathology. These include the presence of abnormal activity in the 4–7 Hz frequency band and the absence or disturbance of activity in the 11–15 Hz frequency band. Power spectral analysis is being utilized to quantify these sleep EEG components in five groups of epileptic patients, studied with different frequency patterns rewarded in an A-B-A design which provides for counterbalancing of order effects. Initial laboratory training is followed by 9–12 months of training at home with portable feedback equipment. Reward contingencies are reversed within each group at approximately three month intervals. Clinical EEG data, blood anticonvulsant measures and patient seizure logs supplement sleep EEG data obtained before training and after each phase of the design. Early results have again indicated specific therapeutic benefits following training of high frequency rhythmic central cortical activity.