Vision screening and vocational aptitude: A factor analysis approach

Abstract

For a good vision screening battery to quickly and accurately reflect the status of the human visual system it should be relevant, reliable, and streamlined. Because the early visual system has limited functional architecture, many simple measurements of the visual system may in fact be measuring the shared computations and parallel processes of other visual functions, making much of the measurement process redundant. This can make a screening battery repetitious and therefore inefficient. The purpose of this research is to investigate these redundancies in a large occupational screening dataset using factor analysis. 192 subjects participated in the Operational Based Vision Assessment (OBVA) Laboratory Automated Vision Testing (AVT) procedure. The AVT includes digital tests for visual acuity, luminance and cone contrast sensitivity, motion coherence, stereopsis, and binocular motor function. Psychometric thresholds and fusional ranges were collected from each subject and a factor analysis was utilized to investigate independent latent variables in the dataset. A promax rotation revealed 5 factors that explained 74% of the total variance: (1) medium and high spatial frequency vision, (2) stereoacuity and horizontal fusional range, (3) cone contrast sensitivity, (4) motion perception, and (5) low spatial frequency vision. Practically, these results suggest that the screening battery can be reduced to 5 independent measurements that capture much of the variance in the dataset. Furthermore, the factors predicted operational and vocational aptitude better than any single variable. More interestingly, these relationships also reiterate known computational processes within the human visual system, such as the parallel processing of low and high spatial frequency content.