Background: The Taylor Visagraph II is a relatively new device designed to evaluate eye movements made during reading. It uses goggles with infrared optics to detect the eye movements and computer software to determine the number of fixations, regressions, and other characteristics of the eye movements. The system also calculates several scores including a school grade equivalent for the reader. Potential applications include the diagnosis of eye movement-related reading problems and the assessment of vision therapy results. Goals of the project were to assess the operation of the system and to determine the validity and reliability of data it produces.
Methods: Fifty first year optometry students served as subjects. Each read 5 standard Taylor Level 10 (College) paragraphs during each of two sessions while eye movements were assessed by the Visagraph II.
Results: The Visagraph II operated correctly for 498 of the 500 trials; operator error caused two malfunctions. Significant differences were found between scores from the first paragraph read and subsequent paragraphs. Nonlinear relationships were found between most of the variables especially those involving grade levels and spans of recognition. Session to session comparisons produced relatively high split-half correlations. Correlations between Optometry Admission Test Reading Comprehension scores and all but two Visagraph II scores were not significantly different from zero. Also of interest was the wide range of reading eye movement skills; several of the optometry students appeared to have skills that placed them at below fourth grade -level.
Conclusions: The Visagraph II is a potentially useful device for the assessment of eye movements made during reading. It performed properly over a large number of trials and produced data that seemed to be reliable indicators of reading skills. Paragraph to paragraph variations in mean scores suggest caution in interpreting small changes in performance across paragraphs and the need for at least one practice trial before usable data are obtained. Further, an understanding of the nonlinear relationships between some of the Visagraph II scores can help to prevent misinterpretations or inappropriate use of statistical comparisons that assume linearity.
Files are restricted to Pacific University. Sign in to view.