Abstract
Several lines of evidence support the hypothesis that the phonological route, rather than the lexical route, requires efficient visuo-attentional mechanism. In particular, impaired spatial attention has been repeatedly described in dyslexic children showing poor phonological decoding skills. A deficit in the magnocellular-dorsal stream has been hypothesized as the possible neurobiological substrate of the attentional deficit in developmental dyslexia. Here, we investigated the impact of different attentional mechanisms on reading, in proficient readers, using different techniques. In a word naming task, we jointly manipulated two orthographic properties of the stimuli: i.e. lexicality (word vs. pseudo-word) and length (short vs. long) that are diagnostic for the distinction between the lexical and phonological routes. In the first study, we investigated the neurobiological basis of reading aloud delivering single pulse TMS to parietal regions. Both the left angular gyrus, part of the reading network, and the right angular gyrus, part of the attentional fronto-parietal network, slowed down the performance. Crucially, in accordance with the magnocellular-dorsal theory, the stimulation of the area controlling spatial and non-spatial attention was more detrimental especially for processing long pseudo-words. In the second study we evaluated, by means of regression modeling, how different attentional abilities can predict performance in a naming task. Spatial and non-spatial attentional mechanisms accounted for a large portion of variance and again, they were especially involved in explaining the length effect in pseudo-word processing. Our findings confirm that visuo-attentional factors play an important role in determining reading ability and that they can be involved in reading difficulties. Moreover, they support the assumption of computational models, such as the CDP+ model of reading aloud, that phonological decoding requires an efficient attentional mechanism that shifts serially through the sub-lexical units.