1

　Original Contribution Kitasato Med J 2012; 42: 1-5　

The magnitude of stereopsis in peripheral visual fields

Hiroshi Mochizuki,1 Nobuyuki Shoji,1,2 Eriko Ando,2 Maiko Otsuka,2

Kenichiro Takahashi,2 Tomoya Handa1,2

1 Department of Ophthalmology, Doctoral Program of Medical Science, Graduate School of MedicalSciences, Kitasato University

2 Department of Rehabilitation, Orthoptics and Visual Science Course, School of Allied Health Sciences,Kitasato University

Objective: We investigated the magnitude of the stereopsis of peripheral visual fields using"CyberDome" (Panasonic Electric Works Co., Ltd.) developed to test binocular visual function.Methods: Sixteen volunteers (mean age, 21.1 ± 1.5 years old), with normal visual fields and stereopsis,were enrolled in this study. The subjects were asked to fix their attention on a central fixation targeton the screen, and the stereopsis target was shown with a binocular disparity so that it stood out fromthe surrounding images. In this system, stereopsis targets were indicated randomly at 10°, 20°, and 30°on 8 meridians radiating from the central target.Results: The perception of stereopsis declined significantly the further away a target was from thecentral target (mean stereoacuity of 8 meridians radiating on 10°, 474 ± 202 arcsecs; 20°, 725 ± 704arcsecs; and 30°, 1,223 ± 1,101 arcsecs). Moreover, stereoacuity from the central target to 30°directly below it tended to be better than that in other directions.Conclusions: When the stereo images appeared in the peripheral visual fields, normal subjects,between the ages of 20 to 22 years old, recognized stereo images of 3,600 arcsecs and over within thecentral 30° of the visual field. Furthermore, the stereopsis in the lower visual field was clearer than inthe other visual fields.

Key words: stereopsis, binocular visual function, peripheral visual field

Introduction

t is well known that stereopsis is the most advancedbinocular visual function. Stereopsis is disturbed by

poor visual acuity in one or both eyes, aniseikonia,strabismus, abnormal retinal correspondence, or defectivedevelopment of the visual cortex.1 Numerous studies2,3

have been conducted on stereoacuity at the fovea, butexamination of past reports revealed only a few studieson detailed peripheral stereoacuity. Previous studies onperipheral stereopsis4,5 have suggested that stereopsisdeclined with distance from the center of the visual field.Sometimes, patients with visual field defects complainof a lack of a sense of distance and, for example, haveslipped on a stairway, or have failed, by misjudgingdistance, to quickly pick up a pen from a desk. Stereopsishas a close relationship with quality of life6-8 andunderstanding stereopsis on peripheral visual fields is

important to consider for the daily life of visual fieldimpaired patients.

Recently, three-dimensional (3D) imaging technologyhas been developing rapidly and has spread gradually inhomes. The CyberDome (Panasonic Electric Works Co.,Ltd.) is the virtual reality display system and can project3D images. Handa et al. improved CyberDome to beable to measure a binocular function and reported that itwas useful in the evaluation of binocular function, forexample, simultaneous perception, fusion amplitude, andstereopsis.9 This device uses special image-correctingtechnology to provide a distortion-free image on the domescreen. This technology contributes to provide accurate3D images on peripheral visual fields. In this study, weinvestigated the magnitude of stereopsis of 8 meridiansradiating from the center within 30° of peripheral visualfield by using the new CyberDome device.

I

Received 25 May 2011, accepted 27 July 2011Correspondence to: Hiroshi Mochizuki, Department of Ophthalmology, Doctoral Program of Medical Science, Graduate School of MedicalScience, Kitasato University1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0373, JapanE-mail: dm07034m@st.kitasato-u.ac.jp

2

Mochizuki, et al.

Subjects and Methods

The subjects were 16 volunteers (21.1 ± 1.5 years old,spherical equivalent of right eye: -2.97 ± 2.49D, sphericalequivalent of left eye: -2.87 ± 2.50D) with a correctedvisual acuity of 1.0 (decimal visual acuity) or better inboth eyes, and stereoacuity better than 100 seconds ofarc (arcsecs) in a Titmus stereo test (Stereo Optical Co.,Inc., IL, USA). There was no significant abnormality inthe visual field test using a Humphrey Field Analyzer(Carl Zeiss Meditec Inc., CA, USA), like that determinedfrom the criteria proposed by Anderson and Patella.10

Informed consent was obtained from all subjects afterthe purpose and experimental procedures to be used inthis study were carefully explained to them. We certifythat all applicable institutional and governmentalregulations concerning the ethical use of humanvolunteers were followed during this research.

Subjects were able to experience visual perceptiongiving the impression of being surrounded by visualimages, a feature of the CyberDome (Panasonic ElectricWorks Co., Ltd.) hemispherical visual display system.Visual images to the right and left eyes were projectedand superimposed on the dome screen, allowing testimages to be seen independently by each eye usingpolarizing glasses. The hemispherical visual display was1.4 meters in diameter. Corrected images are reflected

by a flat mirror in front of the projectors, and a compactimage is projected onto the dome screen. Synchronousprocessing is performed by two slave computers and onemaster computer to project a three-dimensional image.9

At examination time, the distance between the displayand the subject's face (which was fixed by the chin rest)was 1 meter (Figure 1). This device uses a special imagecorrecting technology to provide a distortion-free imageon the dome screen. This technology contributes toprovide accurate 3D images on peripheral visual fields.For this reason, we think that the CyberDome is suitablefor this study.

There is one central fixation target (without parallax)on the center of the display and 24 sites where targets(size, 4°; shape, diamond; color, blue) are indicated at10°, 20°, and 30° in the visual field on all 8 meridiansradiating from the center (Figure 2). When the fixationwas on the central fixation target, sometimes only one ofthe 24 targets showed parallax. The subject would answer,"Yes," when he or she recognized a parallax target andthen stated where it was. The parallax angle was changedfrom 420 arcsecs to 840, 1,800, 3,600, and 7,200 arcsecs

Figure 1. The CyberDome

The CyberDome consists of 3 PCs, 2 projectors, 1 hemispherical screen,and 1 chin rest. The chin rest is placed 1 meter in front of the screen.

Figure 2. The central fixation target and peripheral targets

One central fixation target (blue circle) and 24 peripheraltargets (blue diamonds) were placed in the 10°, 20° and30° peripheral visual fields on 8 meridians radiating fromthe center and were projected on a hemispherical screen.One of the 24 peripheral targets has parallax, and thesubjects answered where this target was. In this figure, the20° straight-down target has parallax.Up, upper meridian radiating; UpRt, upper right; Rt, right;LoRt, lower right; Lo, lower; LoLt, lower left; Lt, left;UpLt, upper left.

3

Stereopsis of peripheral visual fields

Figure 3. Stereoacuity of peripheral visual fields

There were significant differences between the 30° peripheral visual field and the 10° and 20° fields.**P < 0.01 (Bonferroni test)

Figure 4. Stereoacuity of peripheral 10°, 20°, and 30° from the central visual field of each of the 8 directions

The magnitude of stereopsis in the lower visual field tended to be superior to that in any other part of the 30°peripheral visual field. *P < 0.05 (Bonferroni test)Abbreviations on the horizontal axis are the same as in Figure 2.

4

each for 10 seconds. Examination at 7,200 arcsecsresulted in a target whose parallax was too large andcaused a double image. For analysis, therefore, we usedparallax targets that were 3,600 arcsecs or smaller. Theminimum parallax at which the subject recognized a 3Dtarget correctly was defined as the magnitude of stereopsisat that site. Also, observation of the subject's fixation onthe central fixation target was confirmed visually. If thecentral fixation was unstable, the result was not recorded,and a retest was performed.

For statistical analyses the Kruskal-Wallis test andthe Bonferroni test were used. P values of less than 0.05were considered statistically significant. For stereoacuity,the smaller values indicate better stereopsis ability.

Results

All subjects had monocular single vision, and peripheralstereopsis was measured with the CyberDome. The meanstereoacuity was 474 ± 202 arcsecs at a diameter of 10°in the visual field, 725 ± 704 arcsecs at 20°, and 1,223± 1,101 arcsecs at 30°. The farther from the center ofvisual field, the more significantly the magnitude ofstereopsis was declined (Bonferroni test, P < 0.01), andthe more widely it varied (Figure 3).

At diameters of 10° and 20°, the magnitudes ofstereopsis did not differ among the 8 meridians radiatingfrom the center, while at 30°, the magnitudes differedbetween the left and lower parts of the visual field(Bonferroni test, P < 0.05). The stereopsis in the lowervisual field tended to be better than that for the othermeridians at a diameter of 30° in the peripheral visualfield (Figure 4).

Discussion

So far, numerous studies2,3 have been conducted onstereoacuity at the fovea, but a few detailed studies4,5 ofstereopsis in peripheral visual fields have appeared. Inthe present study, within the 30° peripheral visual field,the subjects with normal binocular vision had meanstereoacuity.

It is common knowledge that the stereoacuity at thefovea under the right conditions is several arcsecs.2,3 Themagnitudes of stereopsis in peripheral visual fields were474 arcsecs at 10° in the visual field, 725 arcsecs at 20°,and 1,223 arcsecs at 30° and declined significantly andvaried widely the further the measurement was made (upto an angle of 30°) from the fixation point. We think thatthese results were caused by decreases of amount and ofquality of the visual information in the peripheral visual

fields. In the past, Curcio reported that the greater thedistance from the fovea, the less connection there is withthe retinal ganglion cells and cones, and, in addition, thelower the portion of the cone itself.11,12 The corticalmagnification factor (CMF) indicated how many neuronsin an area of the visual cortex are responsible forprocessing a stimulus of a given size, as a function ofvisual-field location. In the center of the visual field,corresponding to the fovea, a very large number ofneurons process information from a small region of thefield.13 The values of CMF differed by a factor ofapproximately 100 between the representation of thefovea and that of the periphery in the primary visualcortex of primates, but stereoacuity sometimes showeddifferent gradients.14 Accordingly, the magnitude of thedecline of stereopsis increases in the peripheral visualfield, because the amount of visual information decreases.It is known that visual acuity declines markedly outsidethe fovea. While the visual acuity of the fovea was 1.0(decimal visual acuity), in the periphery 20° from thecenter, it decreased markedly to 0.1.15 Moreover, Jenningset al.16 investigated optical characteristics of the humaneye as line-spread function by use of a double-passphotoelectric method and reported optical characteristicdecline farther from the optic axis. Therefore, themagnitude of stereopsis declines in the peripheral visualfield because the quality of visual information decreases.The magnitude of stereopsis in the peripheral visual fieldwas thought to decline according to the amount and thequality of visual information.

The visual functional advantage of one visualhemifield over the other was demonstrated in previousreports for visual acuity, optokinetic nystagmus,multifocal visually evoked potentials, and othercharacteristics.11,17-19 The visual functions of the temporalvisual field are superior to those of the nasal visual field,and the visual functions of the lower visual field aresuperior to those of the upper.20 These facts wereexplained in terms of the embryology of the eye and theanatomy of photoreceptor cells. In the present study,stereopsis of the lower field was superior to that of theupper, in a similar way to the other visual functions.Because the lower visual field plays an important part indaily life, any defect in it contributes to a deterioration ofthe quality of life, for example, in jogging and walking.Therefore, that there is superiority of stereopsis in thelower visual fields makes sense in relation to daily living.

In the present study, it was difficult to set up thevarious amount of the parallax of peripheral targets, sothat it was impossible to change targets in detail at eachmeasurement point. If we could use a different target

Mochizuki, et al.

5

which had a smaller parallax, smaller size, or differentshape, we might investigate more detailed peripheralstereopsis. Furthermore, it was unclear whether or not aminimum detectable parallax was different amongdifferent age groups. Because there were few subjects inthe present study, it is possible that we were able toconfirm only rough outcomes of peripheral stereopsis.A larger study is warranted. However, in our settings inthe present study, when the stereo images appeared inthe peripheral visual fields, the subjects recognized stereoimages of 3,600 arcsecs and over within the central 30°of the visual field. Moreover, these results suggest thatthe stereopsis of the central visual field has excellentfunctionality, reliability, and consistency, within thecentral 10°, among young subjects in particular.

Acknowledgement

We thank Mr. Reynolds CWP for his assistance with theEnglish of the manuscript.

References

1. Rowe F. Binocular Single Vision. In: 2nd ed.Clinical Orthoptics. Oxford, UK: BlackwellPublishing; 2004; 16-27.

2. von Noorden GK, Campos EC. Stereopsis. In: 6thed. Binocular vision and ocular motility. St. Louis:Mosby; 2002; 21-5.

3. Harwerth RS, Schor CM. Stereopsis. In: KaufmanPL, Alm A, editors. 10th ed. Adler's physiology ofthe eye. St. Louis: Mosby; 2003; 493-504.

4. Rady AA, Ishak IG. Relative contributions ofdisparity and convergence to stereoscopic acuity. JOpt Soc Am 1955: 45; 530-4.

5. Yasuoka A, Okura M. Binocular depth and sizeperception in the peripheral field. VISION 2011; 23:103-14 (in Japanese).

6. Nelson P, Aspinall P, Papasouliotis O, et al. Qualityof life in glaucoma and its relationship with visualfunction. J Glaucoma 2003; 12: 139-50.

7. Kuang TM, Hsu WM, Chou CK, et al. Impact ofstereopsis on quality of life. Eye (Lond) 2005; 19:540-5.

8. Rahi JS, Cumberland PM, Peckham CS. Visualimpairment and visual-related quality of life inworking-age adults. Ophthalmology 2009; 116: 270-4.

9. Handa T, Ishikawa H, Shimizu K, et al. A novelapparatus for testing binocular function using the'CyberDome' three-dimensional hemispherical visualdisplay system. Eye (Lond) 2009; 23: 2094-8.

10. Anderson DR, Patella VM. Automated staticperimetry. 2nd ed. St. Louis: Mosby; 1999.

11. Curcio CA, Sloan KR, Kalina RE, et al. Humanphotoreceptor topography. J Comp Neurol 1990;292: 497-523.

12. Curcio CA, Allen KA. Topography of ganglion cellsin human retina. J Comp Neurol 1990; 300: 5-25.

13. Daniel PM, Whitteridge D. The representation ofthe visual field on the cerebral cortex in monkeys. JPhysiol 1961; 159: 203-21.

14. Fendick M, Westheimer G. Effects of practice andthe separation of test targets on foveal and peripheralstereoacuity. Vision Res 1983; 23: 145-50.

15. Wertheim T. Peripheral visual acuity: Th. Wertheim.Am J Optom Physiol Opt 1980; 57: 915-24.

16. Jennings JA, Charman WN. Off-axis image qualityin the human eye. Vision Res 1981; 21: 445-55.

17. Fahle M, Schmid M. Naso-temporal asymmetry ofvisual perception and of the visual cortex. VisionRes 1988; 28: 293-300.

18. Teller DY, Succop A, Mar C. Infant eye movementasymmetries: stationary counterphase gratings elicittemporal-to-nasal optokinetic nystagmus in two-month-old infants under monocular test conditions.Vision Res 1993; 33: 1859-64.

19. Mason AJ, Braddick OJ, Wattam-Bell J, et al.Directional motion asymmetry in infant VEPs --whichdirection? Vision Res 2001; 41: 201-11.

20. Levine MW, McAnany JJ. The relative capabilitiesof the upper and lower visual hemifields. Vision Res2005; 45: 2820-30.

Stereopsis of peripheral visual fields