Fig 1.
Experiments were performed in a dark room using six 32-inch liquid crystal displays (LCDs; MultiSync V321; NEC, Japan; 60 Hz), arranged in a hexagon. Eye movements were recorded by an eye tracker (EMR-9; NAC, Japan; 60Hz). Head and body movements were recorded by two small, light-weight sensors from an electromagnetic motion tracking system (FASTRAK; Polhemus RS-232, USA; 60 Hz).
Fig 2.
Example of visual search stimuli.
(A) White target and distractors were presented randomly on a gray background in one display; (B) A typical frame presentation sequence for one trial.
Fig 3.
Example horizontal eye and head movement trajectories.
Four examples of eye (blue), head (black), body (green), and gaze (red) rotational trajectories during the first two seconds of each trial. The plus and star indicate the onset and offset, respectively, of each gaze and head movement. A saccade is the distance between the plus and star. Upward deflection of the traces indicates rightward movement. The two vertical dashed lines across each trace represent head movement onset and offset. One saccade (A), two saccades (B), three saccades (C), and four saccades (D) in one head movement are shown.
Fig 4.
Vertical eye and head movement trajectories.
Four examples from the same trials as Fig. 3. Notations are the same as in Fig. 3, but upward deflection of the traces indicates upward movement.
Fig 5.
Percentage of head movements with different numbers of saccades.
The bars indicate the average and standard error for seven subjects. The same data are also plotted in a pie chart to show that a larger percentage of head movements had multiple versus single saccades.
Fig 6.
Contribution of head movement rate to gaze shifts.
(A) Head movement amplitude against the total gaze shift amplitude for different numbers of saccades in horizontal and vertical components for seven subjects. Each gray dot shows the individual head movement with saccades. The red dots indicate the average data, which are binned in 10° for horizontal and 3° for vertical. (B) Comparison between average horizontal and vertical movements. Lines are linear functions fitted to the average data across subjects. (C) Distribution of horizontal and vertical gaze amplitudes without head movements for all seven subjects. (D) Average slopes for head movements with different numbers of saccades for each subject. Error bars represent standard error.
Fig 7.
Average duration and speed of head movements as a function of gaze shifts.
(A) Head duration, and (B) maximum speed plotted against total gaze amplitude change. Different symbols represent head movements with different numbers of saccades. Lines are linear functions fitted to the average data across subjects.
Fig 8.
Influence of head orientation on eye position distribution.
(A) Distribution of eye positions with different head orientations is shown in different panels. The horizontal head orientation is binned into 15 groups and eye position data are analyzed for each group. The range of head orientation for each group is shown at the top of each panel. The ordinate shows the percentage of fixations from all head directions and the abscissa shows horizontal eye position relative to the head. The red line shows a Gaussian function fitted by a least square method. (B) The average of the peaks of the Gaussian functions fitted to each data set from seven subjects as a function of head orientation. Error bars represent standard error.