Is dyslexia caused by a visual deficit?

A number of studies have found that dyslexic readers suffer from visual deficits A causal relationship has yet to be established. One step toward establishing such a relationship would be to demonstrate that the degree of dyslexia can be altered by manipulating a purely visual variable. One such variable would be the size of the printed text. In a recent study (O'Brien, Mansfield, & Legge, 2000), reading speed for dyslexic readers and controls was assessed for several different sizes of printed characters. This study, however, did not specifically address the effect on reading performance of character size. I have, therefore , re-plotted the data to investigate this issue. The results are shown in Fig. 1. As can be seen, in the case of both whole sentences (Fig. 1A) and random words (Fig. 1B), there is a clear tendency for dyslexic individuals (filled symbols) to read more slowly than controls (open symbols), and, most importantly, the magnitude of the difference (Fig. 1C) increases with decreasing letter size. If we assume that the degree to which dyslexic readers read more slowly than controls can be taken as a measure of the severity of dyslexia, then Fig. 1C shows that the severity of dyslexia depends, in part at least, on the letter size. The primary alternative to a visual deficit theory of dyslexia is the view that dyslexia is of phonological Changing the letter size clearly affects the visual appearance of a text but does not in any way influence its phonological contents. For this reason, it is difficult to attribute any effect of letter size to phonological influences. Although additional causal factors, among them phonological factors, cannot be ruled out, the present results clearly point to some causal role for vision in dyslexia. A theoretical and experimental case for a visual deficit in specific reading disability. Fig. 1. (A, B) Reading speed as a function of letter size for dyslexic readers (filled symbols) and controls (open symbols). Data are re-plotted from Fig. 4 of O'Brien et al. (2000) (and represent reading speeds at contrast levels where reading speed is independent of contrast). Panels (A) and (B) show reading speed for whole sentences and random words, respectively. (C) Difference in reading speed between dyslexic readers and controls (in words per minute) for sentences (filled symbols) and random words (open symbols). Negative values denote slower reading on the part of the dyslexic readers.