A neurocognitive investigation of phonological processing in dyslexia

This thesis firstly reviews the nature of the phonological deficit in developmental dyslexia. An established hypothesis is that the phonological representations are inadequately specified in dyslexia. Another perspective, considered in this thesis, is that inefficiency in resolving phonological competition compromises the functional availability of phonological information. The neural basis of the phonological deficit is also examined, particularly in relation to the left posterior inferior temporal lobe (BA37) and the left angular gyrus. Six experiments - four behavioural and two neuroimaging - were carried out, with adult developmental dyslexics, and controls matched for age and performance IQ. In Experiment 1, word-reading and picture-naming speed and accuracy were investigated in a single trial format. In Experiment 2, phonological neighbourhood density was manipulated in a continuous naming format. Dyslexic participants were differentially impaired by those factors that increased the level of phonological competition - the presence of orthography, the rate of presentation, and phonological neighbourhood density. These results supported the phonological competition hypothesis. Experiment 3 investigated the neural basis of these impairments during reading and naming. Dyslexic participants showed reduced activation in left BA37 during both tasks. This indicated a general impairment in phonological retrieval, independent of orthographic decoding. Experiment 4 aimed to establish whether similar regions would show abnormal activation during auditory word repetition. In fact, dyslexic participants showed reduced activation in the right hemisphere, particularly the anterior superior temporal lobe (BA22). This was attributed to an attentional bias in the dyslexic group towards phonological aspects of the auditory input. Experiment 5 demonstrated an effect of length in a phonological awareness task. Experiment 6 established that the phonological deficit can have a general impact, influencing performance in tasks that are ostensibly non-verbal. In the light of these findings, the nature of the phonological deficit in dyslexia is reevaluated and suggestions for further research are discussed