Abstract
Abstract: :
Purpose: The aim of our study was to investigate possible differences in attention to change between groups of normal reading (NR), developmental dyslexic (DD) reading-age control (RC) and learning disabled (LD) children. Methods: 103 school children aged 7-14 years participated. On the basis of Neale reading accuracy and Raven's progressive matrices, 19 were classed as DD, 23 as LD 19 as RC and 42 as NR. The change detection task consisted of four letters in placeholders presented on the monitor around a central fixation point. Trials consisted of an initial presentation time, a gap, and a second presentation in which one of the letters may have changed. The initial presentation time was adjusted by a PEST routine such that all subjects performed in the vicinity of 71% correct detection. A second task compared the effects of cued and uncued trials on detection performance. Results: ANOVA showed a difference between groups in the time required to incorporate four letter items in order to detect change at a performance level of 71% correct (F(3,101)= 5.5, P = .001), and DDs required longer times than NRs (Fisher's PSLD, P = .008). However, when provided with cues in a second experimental phase, children of all groups became highly dependent on the cue to help make a judgement of change, with NRs least affected. Change detection was analysed for spatial location. DDs did not demonstrate a mini-neglect for left visual field detections. However, clear differences were found according to whether the target was in the upper (UVF) or lower visual (LVF) hemifield. Paired t-tests demonstrated a clear upper field change detection advantage for the NR, RC and LD groups (P < 0.05), but an insignificant difference for the DD group (P =0.22). Conclusion: We propose that change detection requires an element of pre-identification related to the coding of letter shape. Children learning to read have to decode, without much context. As new and unknown letter images appear suddenly after each saccade, there emerges a possible rationale for the relationships found between reading ability and change detection. Upper/lower visual field differences in performance have previously been attributed to visual field differences between the magnocellular and parvocellular pathways. The lack of UVF advantage in DDs suggests abnormal dorsal pathway processing.
Keywords: 326 attention • 510 perceptual organization • 539 reading