Two experiments on right hemisphere superiority for upright face recognition by normal adults are reported. Experiment 1 showed that right hemisphere superiority for face recognition is affected by the ratio of stimuli to trials used in the experiment. A low ratio of stimuli to trials (Condition A) gave right hemisphere superiorities for both male and female subjects, whereas a high ratio of stimuli to trials (Condition C) led to no visual hemifield difference for male or for female subjects. The use of a stimuli to trials ratio intermediate between those of Condition A and Condition C resulted in a sex difference, with males but not females showing right hemisphere superiority in Condition B. It is argued that both males and females possess asymmetrically organised face processing mechanisms whose operation is dependent upon the level of difficulty of the face memory task used in a particular experiment, but that there is a sex difference in the range across which these mechanisms can operate. Experiment 2 demonstrated that this sex difference does not simply reflect an underlying sex difference in the tendency to use configurational or piecemeal strategies in such experiments.