Generally, an increase of binocular disparity within a certain range increases the level of stereoscopic depth. However, when the amount of disparity passes over the range, the change of stereoscopic depth is perceived in the form of a double image .
In this way, the level of stereoscopic depth determines whether the stereoscopic image appears single or double. Furthermore, the amount of visual discomfort increases with an increase of binocular disparity since it is caused by the inconsistency of accommodation and vergence. In this experiment, the threshold value of MOS was set to 3, the middle point of the five-point scale, to suggest the allowable range of visual discomfort. The result showed that the MOS of positive disparity based on the threshold dropped from the fourth level image (3.5%). For 3D content with negative disparity, the index showing visual discomfort dropped below the threshold level from the fourth level image (3.0%) just like for positive disparity. For 3D content with depth budget, the index dropped from the third level (3.0%). As a result, it is estimated that the allowable range of positive disparity is 2.5~3.5%, negative disparity is2.0~3.0%, and depth budget is 2.0~3.0%.
Visual fatigue caused by disparity
2) Motion depth
Most 3D content consists of videos containing certain types of motion rather than static images. Hence, the change rate of binocular disparity as well as the binocular disparity itself may affect visual discomfort, and this can be more important in certain situations. From the viewpoint of inconsistency of accommodation and vergence, which is the origin of visual discomfort, visual discomfort caused by stereoscopic images containing an object moving in the direction of depth differs from the discomfort caused by images without any motion.
With static stereoscopic images, ocular movements are required to move just once to a specific vergence angle, while such vergence movements happen repeatedly when an object in a stereoscopic video moves in the depth direction. Furthermore, the vergence angle of the object and actual vergence angle of eyes do not exactly match , and this error can increase with the movement speed in the depth direction.
In this experiment, the MOS threshold of visual discomfort was set to 3 to suggest the allowable range of visual discomfort by the change of binocular disparity. Based on this threshold value, the motion depth for visual discomfort was set to 2%, 4% and 6%, and the amount of visual discomfort increased with an increase of motion depth and with an increase of the depth change rate of binocular disparity.
Moreover, the visual discomfort level from images with persisting stimuli of the change of binocular disparity was higher for images with one-time change. When the motion depth was 6%, the difference of visual discomfort depending on the change rate of binocular disparity depth changed significantly for each stimulus level. Figure 37 shows the allowable range of binocular disparity from the viewpoint of visual fatigue. The images with a change of disparity of 0~2% and 0~4% among the repeated stimuli caused visual discomfort if the change rate of binocular disparity was higher than one second.
The 0~6% image caused visual discomfort when the change rate of binocular disparity was higher than two seconds. The image with one-time stimulus caused visual discomfort for all cases when the change rate of binocular disparity was higher than 0.5 seconds.
Visual fatigue caused by motion depth
Motion loop (0% ~ 2%)
Motion once (0% ~ 2%)
Motion loop (0% ~ 4%)
Motion once (0% ~ 4%)
Motion loop (0% ~ 6%)
Motion once (0% ~ 6%)
This research was performed to provide a production guideline for the safety of 3D watching. However, 3D content involves more variables than just the content itself, for example, viewer and/or environmental factors, and those variables all affect each other. This implies that it is difficult to measure the allowable range as an independent variable, and further study is required. Even though the research result is limited to true values in an actual production environment, it still can be used as baseline data to set the default values of stereoscopic variables in 3D content production.
 Lipton, L. (1982). Foundations of the stereoscopic cinema-A study in depth. New York, NY: Van Nostrand Reinhold.
 Recommendation ITU-R BT.500-11 – Methodology for the subjective assessment of the quality of television pictures, 2002.
 Recommendation ITU-R BT.1438 – Subjective assessment for stereoscopic television pictures, 2000.
 Richards, W. (1970). Stereopsis and stereo blindness. Experimental Brain Research, 10, 380-388.
 Ogle, K.N. (1950), Researches in binocular vision. NewYork:Hafner.
 Ogle, K.N. (1952), On the limits of stereoscopic vision. Journal of Experimental Psychology, 44, 253-259.
 Collewijn, H., Erkelens, C.J., & Regan, D. (1986), Absolute and relative disparity: a re‑evaluation of their significance in perception and oculomotor control. In E.L., & Zee, D.S. (Eds.), Adaptive Processes in Visual and Oculomotor Systems (pp. 1777‑1184).
 Golding, J.F. (1998), Motion sickness susceptibility questionnaire revised and its relationship to other forms of sickness. Brain Research Bulletin, 47 (5), 507-516.
 Golding, J.F., & Markey, H.M. (1996), Effect of frequency of horizontal linear oscillation on motion sickness and somatogravic illusion. Aviation, Space, and Environmental Medicine, 67, 121‑126.
Liaison statement to the WHO
Source: Document 6/316
Working Party 6C
DRAFT COMMUNICATION FROM STUDY GROUP 6 TO
THE WORLD HEALTH ORGANIZATION
Visual fatigue and other possible health hazards
due to prolonged viewing of stereoscopic (3D)
ITU-R Study Group 6 (“Broadcasting Service”) has initiated studies on the specifications to be recommended for stereoscopic (3D) television for broadcast use, which is currently attracting some interest on the part of some television broadcasters and of their audiences.
Some ITU members have pointed out that there are indications in the medical literature to the effect that extended viewing of stereoscopic programme material, as displayed on currently available
3D presentation devices, can cause, for example, viewers’ eye fatigue, nausea, dizziness, discomfort, headache and other possible health hazards.
We ask the World Health Organization to kindly advise us on any evidence that they may have on whether viewing stereoscopic 3D television presentations that would be typical of normal television home viewing using currently available displays7 may cause any possible medical issues and to which extent.
We look forward to the WHO kind reply and we thank them for their help on this matter.
[Please address your reply to (Counsellor)…]
The Influence of Alzheimer’s Dementia on dynamic 3D perception
and fatigue while watching 3D Televison
Subjects and methods
Subjects included Alzheimer’s dementia (AD) patients and an age-matched normal control group. All subjects had stereo acuity levels of under 800 arcsec. Informed consent was obtained from all of the patients.
Sex (man: woman)
73.73 ± 6.79
68.14 ± 7.47
14.86 ± 4.37
24.48 ± 2.91
Corrected visual acuity (Left)
0.57 ± 0.27
0.73 ± 0.32
Corrected vVisual acuity (Right)
0.55 ± 0.26
0.75 ± 0.30
Log seconds of arc of Titmus test
168.38 ± 169.09
175.86 ± 258.85
The amount of visual acuity and number of refractive errors were measured. The angle of ocular deviation was obtained with the alternate prism-cover test. The level of stereo acuity was examined with the Stereo Fly test (Stereo Optical Co., Chicago, IL, USA) for near stereopsis.
The 3D video, which was produced by the national broadcasting system of South Korea, was shown for 18 minutes on a 55-inch 3D high-definition television. The viewing distance was 2.7 metres. After watching 3DTV, a survey was performed to evaluate the degree of subjective 3D perception and the symptoms. The questionnaire was comprised of 6 items, which included the degree of 3DTV perception and frequently reported symptoms after watching 3D TV (as shown in Table 18). Each item was answered according to a five-category scale (1-5). A value of 1 corresponded to no impact and a value of 5 corresponded to an impact too severe to watch 3DTV. The degree of subjective 3D perception and level of discomfort were also compared.
For statistical analysis, the Independent-Samples Kruskal-Wallis Test and Kendall’s Taucorrelation were used with SPSS 20.0K for Windows.
Feel stereoscopic vision
* 1: Never experienced, 2: mild impact, 3: moderate impact, 4, severe impact, 5: very severe impact.
Experimental results and discussion
Sixty-six subjects were enrolled in this study, 37 with AD and 29 age-matched subjects without AD. The mean age between the two groups was 73.73 ± 6.79 and 68.14 ± 7.47 years, respectively. The mean cognitive function test (MoCA) between the two groups was 14.86 ± 4.37 and 24.48 ± 2.91, respectively. A cognitive function test showed significant differences between the AD and the control subjects (p < .001).
The mean stereo acuity levels examined with the Stereo Fly test did not differ. However, the mean subjective dynamic 3DTV perception scores of the AD group were significantly different from those of the control group (3.05 ± 1.49 vs. 3.79 ± 1.37, respectively) (Fig. 39).
The scores show that the AD group could not sufficiently perceive dynamic 3D content as well as the control patients. Furthermore, the degree of 3DTV perception was correlated with the scores on a cognitive test (Korean version of Montreal Cognitive Assessment, K-MoCA) (Fig. 40).
There was no difference in the amount of 3D discomfort between the two groups. Therefore, in terms of safety, it can be concluded that 3DTV is not harmful for those with AD (Fig. 41).
In the preliminary experiments, it was observed:
• 3D perception decreased more in the dementia group than in the control group even when there was no structural difference between their eyes.
• Watching 3DTV did not cause any different level of discomfort for the control group nor for those in the AD group.
• In terms of safety, watching 3DTV did not harm the AD patients as long as they complied with the “3D video safety guidelines”.
All of the above observations should be considered when developing viewing safety guidelines of stereoscopic 3D systems.
Differences when using the Titmus fly test and the 3D perception questionnaire
in the AD group and the control group
NS; not significant, *p < 0.05
Correlation 3D perception questionnaire with K-MoCA scores (p < 0.01)
No significant difference in the components of the 3D TV discomfort scale
Mendez MF, Cherrier MM, Meadows RS., Depth perception in Alzheimer’s disease. Percept Mot Skills. 1996 Dec. 83 (3 Pt 1) 987-95.
Mittenberg W, Choi EJ, Apple CC, Stereoscopic visual impairment in vascular dementia. Arch Clin Neuropsychol. 2000 Oct. 15 (7) 561-9.
Mittenberg W, Malloy M, Petrick J, Knee K., Impaired depth perception discriminates Alzheimer’s dementia from aging and major depression. Arch Clin Neuropsychol. 1994 Jan. 9 (1) 71-9.
S.H. Kim, J.H. Park, Y.H. Kim, and S.B. Koh, Stereopsis in drug naive Parkinson’s disease patients, Can J Neurol Sci, 201138: 299-302.
T. Müller, D. Woitalla, S. Peters, K. Kohla, and H. Przuntek, “Progress of visual dysfunction in Parkinson’s disease,” Acta Neurologica Scandinavica, 2002 105: 256‑260, 2002.
The role of 3D Television in terms of refractive errors in children
Subjects and methods
Healthy volunteers aged 6 to 12 years were recruited. Informed consent was obtained from the parents of all the subjects. Before enrolment, manifest refraction, slit lamp examination, fundus evaluation with fundus camera, alternate prism-cover test and near stereopsis test with the Stereo Fly (Stereo Optical Co., Chicago, IL, USA) were performed. Volunteers with strabismus, best corrected Snellen visual acuity less than 20/20, near stereoacuity worse than 60 seconds of arc, anisometropia more than 2.00 D or with any structural abnormalities in the cornea, lens, retina or optic nerve were excluded.
A 3D video was shown for 50 minutes on a 3D high-definition television (UN55C7000WF, Samsung Electronics, Suwon, South Korea) with a diagonal screen size of 139 cm. Subjects wore liquid-crystal shutter glasses. The illumination of the room was 5 lux and the viewing distance was 2.8 metres. The viewing distance and duration were set following the recommendations of the TV manufacturer and the International Telecommunication Union (2.2 metres or more for a 55 inch HDTV, respectively).
The 3D content used in this study was produced with the true 3D shooting technique by the national broadcasting system of South Korea for 3D test-run broadcasts. The image disparity ranged from 1 to 1 degree and the reference depth was zero screen disparity.
Refractive errors were obtained objectively with an autorefractor (RK-F1, Canon, Tokyo, Japan) before and immediately after watching 3DTV. It was rechecked after a 10-minute rest. At each point, the measurement of the refractive errors was repeated until we obtained the same value for three consecutive measurements of each subject. The repeatability of the autorefractor was assessed using a coefficient of variation (0.64%), and it was highly reliable as we reported in the previous study. Spherical equivalent (sphere + 1/2 cylinder) of the right eye was used in this study. The refractive errors before and after watching 3DTV were compared. The subjects whose spherical equivalent of baseline refractive error was worse than 0.75 diopters (D) were included in the myopia group, and the other subjects were in the non-myopia group. The number of refractive changes after watching 3DTV was compared between the two groups. The refractive changes of the subjects who showed myopic shifts were compared between each point to assure that the myopic shift persisted after 10 minutes of rest.
The mean age of the subjects was 9.23 ±1.75 years.
Sixty normal subjects were enrolled. Their mean age was 9.23 + 1.75 years (6~12).
The refractive errors before and after watching 3DTV are shown in Table 19. There were no statistically significant differences between each time point.
The changes of refractive errors after watching 3DTV
1.70 ± 1.79
Immediately after watching
1.75 ± 1.85
After 10 minutes of rest
1.69 ± 1.80
The myopic group consisted of 34 subjects (56.7%). Table 20 shows a comparison of the change of refractive error between the myopia and non-myopia group. The refractive errors did not change significantly for either of the groups. Also, the mean change of refractive errors immediately after watching 3DTV did not differ between the two groups (P = 0.541).
The refractive errors of the non-myopia and myopia group before & after watching 3DTV
(n = 26, < 0.75D)
(n = 34, > 0.75D)
0.10 ± 0.56
2.93 ± 1.39
Immediately after watching
0.10 ± 0.69
3.01 ± 1.41
After 10 minutes of rest
0.10 ± 0.62
2.92 ± 1.42
The distribution of refractive changes before and immediately after watching 3DTV is shown in Fig. 42.
The distribution of the changes of refractive errors after watching 3DTV
Myopic shift (D)
The myopic shift was observed in 31 subjects. The refractive errors before watching 3DTV in these subjects (1.83 ± 1.92 D) did not differ from those in the others who did not show any myopic shift (1.53 ± 1.68 D; P = 0.636). The ages of the subjects did not differ either (P = 0.994). In subjects with myopic shift, the refractive errors significantly changed immediately after watching 3DTV (P < 0.001) and the mean amount of myopic shift was 0.29 ± 0.23 (0.13-1.00) D. However, it was resolved after 10 minutes of rest and the refractive errors before watching 3DTV and after 10 minutes of rest did not differ significantly (P = 0.122).
2 Subjects with exodeviation
There was no significant change in terms of exodeviation and refractive errors after watching 3DTV. Table 21 shows the changes in refractive errors and the angle of deviation after watching 3DTV.
Changes of refractive errors (D) and angle of exodeviation (PD) after watching 3DTV
Angle of exodeviation (PD)
13.04 ± 5.25
12.96 ± 5.11
0.09 ± 2.45
Refractive errors in the right eye (D)
2.15 ± 1.55
2.14 ± 1.57
0.01 ± 0.22
Yano S, Ide S, Mitsuhashi T, Thwaites H., A study of visual fatigue and visual comfort for 3D HDTV/HDTV images. Displays 2002; 23: 191-201.
Schor CM, Tsuetaki TK., Fatigue of accommodation and vergence modifies their mutual interactions. Invest Ophthalmol Vis Sci 1987; 28: 1250-1259.
Hoffman DM, Girshick AR, Akeley K, Banks MS., Vergence-accommodation conflicts hinder visual performance and cause visual fatigue. J Vis 2008; 8:33. 1-30.
Lambooij M, Ijsselsteijn W, Fortuin M, Heynderickx I., Visual Discomfort and Visual Fatigue of Stereoscopic Displays: A Review. Journal of Imaging Science and Technology 2009; 53(3).
Kim SH, Suh YW, Song JS, Park JH, Kim YY, Huh K, et al., Clinical Research on the Ophthalmic Factors Affecting 3D Asthenopia. J Pediatr Ophthalmol Strabismus 2012; 49: 248-253.
Kim SH, Suh YW, Yun CM, Yoo EJ, Yeom JH, Cho YA., 3D Asthenopia in Horizontal Deviation. Curr Eye Res 2013; 28: 614-619.
Lin LL, Shih YF, Tsai CB, Chen CJ, Lee LA, Hung PT, et al., Epidemiologic study of ocular refraction among schoolchildren in Taiwan in 1995. Optom Vis Sci 1999; 76: 275-281.
Wu HM, Seet B, Yap EP, Saw SM, Lim TH, Chia KS., Does education explain ethnic differences in myopia prevalence? A population-based study of young adult males in Singapore. Optom Vis Sci 2001; 78: 234-239.
Jung SK, Lee JH, Kakizaki H, Jee D., Prevalence of myopia and its association with body stature and educational level in 19-year-old male conscripts in seoul, South Korea. Invest Ophthalmol Vis Sci 2012; 53: 5579-5583.
Yoon KC, Mun GH, Kim SD, Kim SH, Kim CY, Park KH, et al., Prevalence of eye diseases in South Korea: data from the Korea National Health and Nutrition Examination Survey 2008-2009. Korean J Ophthalmol 2011; 25: 421-433.
Chung K, Mohidin N, O’Leary DJ., Undercorrection of myopia enhances rather than inhibits myopia progression. Vision Res 2002; 42: 2555-2559.
Ciuffreda KJ, Vasudevan B., Nearwork-induced transient myopia (NITM) and permanent myopia – is there a link? Ophthalmic Physiol Opt 2008; 28: 103-114.
McBrien NA, Young TL, Pang CP, Hammond C, Baird P, Saw SM, et al., Myopia: Recent Advances in Molecular Studies; Prevalence, Progression and Risk Factors; Emmetropization; Therapies; Optical Links; Peripheral Refraction; Sclera and Ocular Growth; Signalling Cascades; and Animal Models. Optom Vis Sci 2008 Dec. 19.
Mutti DO, Mitchell GL, Moeschberger ML, Jones LA, Zadnik K., Parental myopia, near work, school achievement, and children's refractive error. Invest Ophthalmol Vis Sci 2002; 43: 3633-3640.
Saw SM, Nieto FJ, Katz J, Schein OD, Levy B, Chew SJ., Factors related to the progression of myopia in Singaporean children. Optom Vis Sci 2000; 77: 549-554.
Vera-Diaz FA, Strang NC, Winn B., Nearwork induced transient myopia during myopia progression. Curr Eye Res 2002; 24: 289-295.
Walline JJ, Jones LA, Mutti DO, Zadnik K., A randomized trial of the effects of rigid contact lenses on myopia progression. Arch Ophthalmol 2004; 122: 1760-1766.
Walline JJ, Mutti DO, Jones LA, Rah MJ, Nichols KK, Watson R, et al., The contact lens and myopia progression (CLAMP) study: design and baseline data. Optom Vis Sci 2001; 78: 223-233.
Yang Z, Lan W, Ge J, Liu W, Chen X, Chen L, et al., The effectiveness of progressive addition lenses on the progression of myopia in Chinese children. Ophthalmic Physiol Opt 2009; 29: 41-48.
Ciuffreda KJ, Vasudevan B., Effect of nearwork-induced transient myopia on distance retinal defocus patterns. Optometry 2010; 81: 153-156.
Ong E, Ciuffreda KJ., Nearwork-induced transient myopia: a critical review. Doc Ophthalmol 1995; 91: 57‑85.
Wee SW, Moon NJ, Lee WK, Jeon S., Ophthalmological factors influencing visual asthenopia as a result of viewing 3D displays. Br J Ophthalmol 2012 Nov. 96 (11): 1391‑4.
Suh YW, Oh J, Kim HM, Cho YA, Song JS., Three-dimensional display-induced transient myopia and the difference in myopic shift between crossed and uncrossed disparities. Invest Ophthalmol Vis Sci 2012; 53: 5029-5032.
Takeda T, Hashimoto K, Hiruma N, Fukui Y., Characteristics of accommodation toward apparent depth. Vision Res 1999; 39: 2087-2097.
Arunthavaraja M, Vasudevan B, Ciuffreda KJ., Nearwork-induced transient myopia (NITM) following marked and sustained, but interrupted, accommodation at near. Ophthalmic Physiol Opt 2010; 30: 766-775.
Vasudevan B, Ciuffreda KJ., Additivity of near work-induced transient myopia and its decay characteristics in different refractive groups. Invest Ophthalmol Vis Sci 2008; 49: 836-841.
1 Further general information on the subject of 3DTV can be found in Report ITU-R BT.2160-4.
2Disparity describes the actual shift between the same two points in the stereo pair. Thus the disparity would always remain the same irrespective of screen size.
3See for instance: K. Ukai and P.A. Howarth “Visual fatigue caused by viewing stereoscopic motion images: background, theories and observations” – Elsevier B:V., 2007, which states, inter alia, “Viewers should be careful to avoid viewing stereoscopic images for extended durations because visual fatigue might be accumulated. They should be ready to stop immediately if fusion difficulties are experienced”.
4Masaki Emoto, Takahiro Niida, and Fumio Okano – “Repeated Vergence Adaptation Causes the Decline of Visual Functions in Watching Stereoscopic Television”, Journal of Display Technology, Vol. 1, No. 2, December 2005.
5 All test subjects were volunteers able to understand and sign a consent pro forma and all tests were conducted under the guidance of the Institutional Review Board (IRB).
6DPA (The Association for the Promotion of Digital Broadcasting) is an organization of Japanese broadcasters and manufacturers associated with digital TV.
7 It is also known that all currently available stereoscopic displays for television viewing in the home require the use of special 3D glasses, and some 3D viewing uses shuttered “active glasses”.