Abstract
Studies suggest that poor reading can be related to poor binocular visual skills, such as vergence. It has been difficult to relate reading directly to such skills, however, because most datasets lack sufficient quantitative measures of relevant skills. In a prior study, MKP had gathered data from 1,062 students, ages 8-11; each subject had quantitative symptom and optometric measures. On re-examining this dataset with WPF, we constructed a variable called Functional Binocular Vision (FBV), by using a probabilistic approach that prioritized the estimation and comparison of linear measures. Besides demonstrating that FBV is robust and reliable, we also showed that it is not related to visual acuity. In the present study, we examined the relationship among FBV, acuity, and reading. In the original MKP study, a subset of subjects participated in an in-school intervention designed to improve visual skills using a computer program. The program kept track of 10 indicators (latency and accuracy for each of 5 training modules, each relating to a defined visual skill). These students also had sustained oral reading fluency measures taken during the intervention. Regression analysis of FBV variables against oral reading fluency scores had a significant adjusted R squared (R2) of 0.61; when acuity entered the regression equation, the R2 dropped to 0.44. When we examined the variables that changed with visual skill training, we found that binocular (FBV) and symptom (CISS) scores changed with reading, but acuity did not. We conclude: (a) Functional Binocular Vision (FBV) is a robust and useful construct that can predict oral fluency reading performance independent of Snellen acuity, and (b) training visual skills can improve reading outcomes.