Central and peripheral letter recognition in specific reading disability

In 1987, Geiger and Lettvin tested letter recognition in the right visual field and found that in late adolescents and adulthood SRDs had a letter recognition advantage in peripheral visual areas compared to normal readers at 7.5°, 10° and 12.5° of visual angle from fixation. They concluded that whereas normal readers attended to stimuli within a narrow range surrounding their point of fixation, so that information in the periphery was lost, dyslexics could identify letters in the periphery better than normal readers. The results also suggest that differences in low level visual processing which has been shown to be deficient in younger SRDs may persist into adulthood. The results may also be taken as further evidence for a visual difference between SRDs and normals for which there is increasing evidence. Specifically, Geiger and Lettvin's results may have some bearing upon Sustained/Transient functioning, hypothesized to be different in young SRDs from normal readers. However, the flaws in the design and the reporting of the experimental procedure demand that their experiments be repeated under more controlled conditions. Thus the present thesis reports two experiments which repeat as closely as possible and extend the findings of Geiger and Lettvin. In the first experiment, letter recognition in both left and right visual fields was investigated using a backward masking paradigm. In preliminary testing, the threshold durations for unmasked target stimuli at 7 retinal positions, at fixation and three each in the right and left visual fields, were measured in order to determine individual target durations for subjects in the masking study. The results showed no significant differences for duration threshold for isolated target letters although there was a trend for the target duration thresholds in the SRI) group to be longer in peripheral visual areas. The masked letter recognition functions in the groups showed that the SRD group experienced significantly more masking than controls, but that this depended upon the retinal position and inter-stimulus interval. Both groups showed fewer errors with increasing ISIs. Controls showed more errors at each ISI with increasing Retinal eccentricity, but the SRD group showed more masking in central retinal positions at the lowest ISIs. Both groups, but especially the SRD group, showed flatter error functions across ISI at the extreme retinal positions, and especially in the Right visual field. The most striking result of Experiment 1 was that SRDs experienced significant levels of masking in central vision for a single backwardly masked target. A second experiment was conducted which investigated simple three letter word recognition under a number of forward and backward masking conditions. The three masking conditions were isolated target word, target word followed by mask, and target word embedded within a series of forward and backward masks. The results showed that simple word recognition required longer threshold durations for the SRD group to correctly identify the words under all the masking conditions. The results were discussed in terms of the role of visible peristence and sustained and transient channel functioning as well as the transient channel deficit hypothesis developed by Lovegrove et al. (1986). The results appear to support the transient deficit hypothesis in these older age groups and, in addition, showed that a transient system deficit is likely to have implications for sustained channel functioning.