Picture perception in Chinese dyslexic children: an eye-movement study.

Background: Currently, whether or not there is visuospatial impairments in Chinese dyslexic children is still a matter of discussion. The relatively recent application of an eye-tracking paradigm may offer an opportunity to address this issue. In China, in comparison with reading studies, there have not been nearly as many eye movement studies dealing with nonreading tasks such as picture identification and whether Chinese children with dyslexia have a picture processing deficit is not clear. The purposes of the present study were to determine whether or not there is visuospatial impairments in Chinese dyslexic children. Moreover, we attempted to discuss whether or not the abnormal eye movement pattern that dyslexic subjects show during reading of text appropriate for their age is a consequence of their linguistic difficulties.

Methods: An eye-link II High-Speed Eye Tracker was used to track the series of eye-movement of 19 Chinese dyslexic children and 19 Chinese normal children. All of the subjects were presented with three pictures for this eye-tracking task and 6 relative eye-movement parameters, first fixation duration, average fixation duration, average saccade amplitude, mean saccade distance, fixation frequency and saccade frequency were recorded for analysis.

Results: Analyzing the relative parameter among three pictures, except for the fixation frequency and the saccade frequency, other eye-movement parameters were significantly different among the three pictures (P<0.05). Among the three pictures, the first fixation duration was longer, and the average fixation duration, the average saccade amplitude and the mean saccade distance were shorter from picture 2 to picture 3. Comparing all eye-movement parameter between the two groups, the scores of average saccade amplitude (P=0.017) and the mean saccade distance (P=0.02) were less in the dyslexia group than in the normal group (P<0.05), other parameters were the same in the two different groups (P>0.05).

Conclusions: The characteristics of the pictures can significantly influence the visuospatial cognitive processing capability of the Chinese children. There is a detectable disability for the Chinese dyslexic children in the visuospatial cognitive processing: their saccade amplitude and mean saccade distance are shorter, which may be interpreted as specific for their reading disability.