Operant (biofeedback) control of left-right frontal alpha power differences: potential neurotherapy for affective...

Two experiments were done with subjects from a paid pool of undergraduates. In each study, there were five 1-hour sessions on each of 5 days: (1) Baseline: Rewards given for randomly selected 20% of the 700-ms sequential epochs; mean and SD of baseline power differences determined. 2) Exploration: Subjects were rewarded when right minus left alpha differences in an epoch were greater than the baseline mean plus about .85 SD (p = .20); subjects told to discover how to generate rewards. (3)-(5). Training: Subjects were paid (over and above the $8/h flat rate) in proportion to their hit rates. In the first study (in which active filters passed 8-12 Hz activity, and the rectified, integrated amplitude was utilized), 6 of 8 subjects met learning criteria (a significant difference between baseline and training scores). In the second study (in which on-line FFTs were used to extract alpha power), 3 of 5 subjects met learning criteria.