Cortical alpha rhythm, biofeedback, and the determinants of subjective state

A critical assumption in the rationale for the clinical application of voluntary control over central nervous system (CNS) activity is that there exists a direct relationship between specific CNS activities and specific subjective mood states. The experiment reported here was based on the idea that a joint manipulation of both cognitive and physiological (i.e., alpha electroencephalographic [EEG]) variables would provide information required for an explication of the processes involved in subjective mood change. Three main variables were considered in the present experiment. First, the instructions were designed to induce a "set" or expectations for either positive or negative changes in subjective mood. Second, the type of feedback (alpha or not alpha) was varied. Those in the not-alpha feedback condition received feedback contingent on the absence of alpha activity, whereas those in the alpha feedback condition received feedback contingent on the presence of alpha activity. Third, an estimate of each individual's baseline alpha density was obtained, and on this basis subjects were classified as high-baseline or low-baseline subjects. Baseline alpha density, alpha density during training, and subjective mood were all considered as continuous variables. The results of the present study indicate that biofeedback of the cortical alpha rhythm is neither a sufficient nor a necessary condition for changes in subjective mood. In addition, the type of instructions designed to set subjects for positive or negative alterations in subjective states is predictive of changes in subjective state reported during alpha biofeedback training as well as of success at controlling and changing EEG during feedback training. The data presented here permit a direct examination of the actual relationship of EEG changes to changes in subjective mood state. Similar shifts in subjective mood state can be obtained with either increases or decreases in alpha activity. Further, both positive and negative subjective experiences can be associated with increases in alpha activity.