Eysenck suggested that extraverts are chronically cortically under aroused in comparison to introverts. However, many psychophysiological studies do not show the predicted differences. Gale's (1969) explanation is that extraverts engage in compensatory behaviour in the laboratory to increase their level of arousal which would reduce differences between extraverts and introverts. We tested this hypothesis by comparing the amount of movement exhibited by introverts and extraverts in the laboratory. Stelmack has suggested that movement is fundamental to differences between introverts and extraverts. We measured movement using a pressure mat system located in the seat of the chair but unknown to the participant. There were two conditions: a no activity condition; and an activity condition where the participant could self-stimulate by pressing a keyboard to hear various sounds. We found that more extraverted and more neurotic individuals moved more but this difference was confined to the no activity condition. We conclude that in the laboratory, as in ‘real’ life, people may use behaviour to manipulate their levels of arousal even if they are unaware that they are doing so. Furthermore, this finding may partially explain failures to find predicted differences in ‘resting’ levels of cortical arousal associated with extraversion in the laboratory.