Ophthalmology in China ›› 2014, Vol. 23 ›› Issue (1): 37-42.doi: 10.13281/j.cnki.issn.1004-4469.2014.01.010

Previous Articles     Next Articles

Visual comfort and fatigue measured by eye movement analysis when watching three-dimensional displays

ZHANG  Li,   Ren-Jie,   Xu-Liang,   Zhang-Jing-Shang,   Zhao-Jing   

  1.  Beijing Institute of Ophthalmology; Beijing Ophthalmology & Visual Sciences Key Lab.;Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing 100005, China
  • Received:2013-12-15 Online:2014-01-25 Published:2014-01-22
  • Contact: XU Liang, Email: xlbio1@163.com

Abstract: Objective To assess and compare visual fatigue and visual discomfort induced by viewing stereoscopic displays using eye movement analysis. Design Observational cross-over study. Participants Twenty eight healthy subjects (age: 20~30 years), -3.0D  < refractive error <+1.0 D. Methods Three dimensional (3D) displays used the pattern retarder-spatial domain technology with linearly polarized three dimensional technologies. During this study performed at two subsequent days, 28 subjects watched a video shown on a 3D display or a two-dimensional (2D) display. Main Outcome Measures Blink frequency and duration, fixation duration, pupil size, and saccade frequency, duration, speed and amplitude of the saccades.  Results Blinking frequency was significantly lower when watching 3D display than watching the 2D display (left eye: P=0.02; right eye P=0.01). With increasing duration of watching the movie, blinking frequency increased during the 3D watching and remained unchanged during the 2D watching. Blinking duration did not differ significantly between the 3D watching and 2D watching.  Pupil diameter did not differ significantly between 2D viewing and 3D viewing (left eye, P=0.86; right eye P=0.64).  Fixation duration was significantly shorter during 3D viewing than during 2-dimensional viewing (left eye, P=0.02, and right eye P=0.02). Fixation duration did not differ significantly between the two eyes (P=0.94 for 3D viewing and P=0.98 for 2D viewing). The saccade parameters (frequency, duration, amplitude, velocity and acceleration) differed significantly between 3D viewing and 2D viewing.  Except for the difference in the saccade duration (P=0.046 for the left eye; P=0.035 for the right eye), all other parameters of the saccade measurements (frequency, saccade amplitude, saccade velocity and acceleration) showed highly significant differences between 3D viewing and 2D viewing (All P<0.001). Compared to 2D viewing, 3D viewing  was associated with a higher saccade amplitude, higher saccade frequency, shorter saccade duration and faster saccade speed.  The mean saccade frequency decreased from 980.8 / minute (left eye) in the first 5 min interval to 937.2 / minute (left eye) in the last 5 minute interval during the 3D viewing.  Saccade amplitude between the left eye and the right eyes did not differ significantly (P=0.94 for 2D viewing, and P=0.65 for 3D viewing). Saccade speed between two eyes did not vary significantly (P=0.90 for 2D viewing, and P=0.82 for 3D viewing). Conclusion 3D viewing, as compared with 2D viewing, was associated with lower blinking frequency, faster saccadic speed, higher saccadic frequency, and larger saccadic amplitude, while pupil size did not vary between them.  At the end stage of the 3D movie session, the increased values decreased to supra-normal values. This pattern of eye movements suggests that 3D viewing of movies, as compared to 2D viewing, is associated with an increased level of alertness which decreases to supra-normal values after 40 minutes of watching.  (Ophthalmol CHN, 2014, 23: 37-42)

Key words: stereoscopic viewing, blink, fixation, saccade, pupil size