Spectral analysis of electroencephalogram during perceptual-motor learning

Spectral analysis of the electroencephalogram was utilized to determine attention demands during learning of a perceptual motor task. Fifteen subjects participated in the task which consisted of 15 trials of the mirror star learning task. EEG was monitored bilaterally from the occipital and parietal lobes during each trial and analyzed for intensity and mean frequency components within five different bandwidths. All subjects rapidly improved performance in terms of both time and error reduction although the former measure appeared to be a better fit of the supradiagonal form of matrix typically found for this type of learning. Mean alpha frequency increased slightly during the task while delta frequency decreased.