Abstract
Introduction: Hypotheses concerning the neurological basis of dyslexia, a common reading disorder, include dysfunction of the magnocellular system in the brain, abnormal temporal processing, and/or deficient phonological skills. Each of these theories has merit, but the link among them has not been identified. Methods: While being functionally imaged with a Siemens Prism 3T MRI scanner, nine subjects with dyslexia and eleven normal readers viewed high contrast checkerboard stimuli contrast reversing at various frequencies. We compared the responses to high and low frequency flicker. The subjects also listened to blocks of either transient (T) sounds with abrupt onsets or onsets, or sustained sounds (S) with ramped onsets and offsets. The auditory stimuli had no linguistic content and the blocks were independent of the visual stimuli. We computed a transient index (TI), defined as the difference in response amplitudes to the two blocks divided by their sum: TI = (T+S)/(T+S). We confined our functional analyses to the lateral (LGN) and medial (MGN) geniculate nuclei in the thalamus. Results: In the visual system, the normal readers exhibited a preference for higher frequency visual flicker in the magnocellular portion of the LGN, whereas the subjects with dyslexia did not. However, larger differences were found in the magnocellular portion of the MGN. Here in the subjects with dyslexia, we observed profoundly attenuated responses to non-linguistic transient but not sustained sounds compared to normal readers. This finding was highly consistent among the subjects in each group: the mean TI measurement in magnocellular slices of the MGN was a 95% accurate diagnostic for dyslexia among the individual subjects. Conclusion: Here we show that people with dyslexia have a selective impairment in the magnocellular portion of the medial geniculate nucleus, the auditory relay in the thalamus. Our finding identifies a core deficit potentially explaining three theories of dyslexia.