Comparative Study Of Different Types Of Stereoacuity Tests In 2-to 3-Year-Old Children

Objective:There are various tests for assessing stereopsis.Based on the state of the eyes during the tests,they can be divided into two kinds: the tests conducted in real depth and the tests conducted while separating the images seen by the left and right eyes with the help of dissociation techniques.As for tests in the form of pictures,they can also be divided into two patterns: contour-based stereograms and random-dot-based stereograms.At present,there are controversies about whether there are differences between results of the tests conducted in real depth or with binocular dissociation for different populations.Similar controversies exist between contour-based and randomdot-based stereograms.Furthermore,few studies recruited young children to compare stereopsis measured by different types of stereotests.This study aims to compare the stereoacuity measured in real depth and binocular dissociation and to compare the stereoacuity measured by contour-based and random-dot-based stereograms.Methods:(1)A total of 130 children aged 2–3 years were recruited.The Frisby near stereotest(FNS)was modified to measure stereopsis in real depth.The test distance was calculated according to interpupillary distance of the participant,which was generally about 0.40 m.A naked-eye 3D smartphone was used to establish the test system for binocular dissociation.The test distance was 0.43 m.For both systems,the size of test patterns that imitated the Frisby test was the same.Besides,the disparities were set as480″,240″,120″ and 60″ in both systems.The stereoacuity of 72 children was measured using these two tests.(2)The contour-based stereotest and random-dot-based stereotest were conducted using a naked-eye 3D smartphone.The contour-based symbol imitated the quantitative measurements of the Stereo Fly Test.The random-dot symbol imitated the quantitative measurement of the Random Dot 3 Stereo Acuity Test.For both tests,the test distance was 0.43 m and the disparities were set as 480″,240″,120″ and 60″.The stereoacuity of 58 children was measured using these two tests.The Wilcoxon signed-rank test was used to explore the differences between stereoacuity measured by different groups(Real depth vs.Binocular dissociation and Contour-based stereograms vs.Random-dot-based stereograms).The kappa statistic was used to determine the agreement between stereoacuity measured by the different groups.Results:The stereoacuity was significantly different between real depth and binocular dissociation(Wilcoxon signed-rank test: Z=-4.082,P<0.001).Stereoacuity was higher when measured by binocular dissociation than when measured in real depth.There was substantial agreement between stereoacuity measured using these two tests(κ=0.658;95% CI: 0.497–0.819).(2)The stereoacuity was significantly different between the results for contour-based and random-dot-based stereograms(Wilcoxon signed-rank test:Z=-3.275,P= 0.001).Stereoacuity was higher when measured using random-dotbased stereograms than when measured using contour-based stereograms.There was moderate agreement between stereoacuity measured using contour-based and randomdot-based stereograms(κ=0.522;95% CI: 0.339–0.704).Conclusions:For 2-to 3-year-old children,stereopsis tested in real depth was better than that tested by binocular dissociation.In addition,stereopsis was better when measured using contour-based stereograms than when measured using random-dot-based stereograms.