The efficiency of eye-movements while tracking sequential lights was compared for 15 dyslexic children and 15 controls, matched on age and non-verbal intelligence. Contrary to Pavlidis (I98 I), no significant differences were found between the two groups on two measures of efficiency. A hierarchical cluster analysis did not demonstrate unique patterns of errors for dyslexics, with the exception of a single female dyslexic.

One of the difficulties facing any attempt to investigate a research finding is to know what effect certain differences in procedure or equipment will have on the result. Our previous attempt at looking at non-reading eye-movements in developmental dyslexics (Stanley, 1978) produced essentially the same results as another research group (Adler-Grinberg & Stark, 1978), despite considerable differences in procedure and equipment. With respect to Pavlidis' (1981 a ) work, we were somewhat at a disadvantage, in so far as his original paper does not provide the rationale now given (Pavlidis, 1983) for a number of procedural moves which might otherwise be considered arbitrary. While appreciating the value of having his position clarified for us in a way not provided by his previous publication, we would like to comment on a number of points he has made about our paper (Stanley et al., 1983).