Twenty subjects, aged 17 to 25, were given from 5 to 10 sessions of training in controlling alpha. They were divided into three groups, respectively reinforced for increasing alpha from the central area, reinforced for decreasing alpha from the central area, and given noncontingent reinforcement. Compared with the initial baseline, the alpha of the noncontingent subjects did not change, while those reinforced for increases were reliably higher and those reinforced for decreases reliably lower than the noncontingent group. A slight trend toward improvement during successive sessions was not reliable. Since the experiment was conducted in the Soviet Union, the subjects had no expectations of an "alpha experience." Although tests showed a slight elevation in mood at the end of the sessions, there were no differences among the groups. There was an increase in reports of fatigue after the training sessions. There were no reports of using visual or somatomotor maneuvers as a means of controlling alpha. Furthermore, alpha rhythm control was not found to be consistently correlated with changes in heart rate, respiration, or mood, as determined by cross-correlation analysis.