The maintenance and adjustment of visual gaze are functions both of eye and of head movement. Compared to the factors that affect eye movement during gaze displacement, the factors that affect head movement have received relatively little attention. Most experimenters have restricted head movements to determine how the eye acquires information. Information is commonly acquired without head movements (e.g., it is likely you are moving your eyes without moving your head as you read the words on this page). However, when visual information beyond the written page is acquired head movements in addition to eye movements are likely to occur. The purpose of this experiment was to study gaze displacement under more natural conditions in which the eyes and head were both free to move. Specifically, the purpose of this experiment was to identify the pattern of gaze displacement as a function of cognitive task difficulty. Visual information was presented at an eccentricity of 40 degrees of visual angle either to the left or to the right of a center fixation point (0 degrees eccentricity) to 24 undergraduate subjects (12 male). On each trial, five integers were presented in rapid succession; subjects were asked to count the number of odd integers or to arithmetically manipulate the integers. When subjects expected to perform the more difficult task of arithmetic manipulation, they were more likely to make head movements in the direction of the stimuli, F-1,F-16 = 9.34, P < 0.05. Those subjects who made head movements to acquire the peripherally presented visual information made significantly more errors than the 'non-movers', F-1,F-24 = 6.03, P < 0.05. These results suggest subjects who moved their head to acquire peripheral information may have found the task to be more difficult than subjects who did not make head movements. The results confirm that head movements play a critical role in the intake of information even in structured laboratory tasks and suggest the common method of restricting head movements to study eye movements may not be a natural method of information intake. (C) 1997 Elsevier Science B.V.