Functional magnetic resonance imaging has helped University of Washington researchers discover that too much connectivity between the language-processing regions of the brains of dyslexic children may contribute to reading disabilities. Follow-up scans produced functional evidence that these abnormal connectivity patterns were mitigated after only three weeks of specialized reading training.
Functional magnetic resonance imaging has helped University of Washington researchers discover that too much connectivity between the language-processing regions of the brains of dyslexic children may contribute to reading disabilities. Follow-up scans produced functional evidence that these abnormal connectivity patterns were mitigated after only three weeks of specialized reading training.
Radiology professor Dr. Todd Richards and neuropsychologist Virginia Berninger, Ph.D., director of the university's Learning Disability Center in Seattle, conducted fMRI on 18 dyslexic children with known reading problems and 21 children who were good readers and spellers. All subjects exhibited normal intelligence and were in the fourth through sixth grades. FMRI was acquired while they performed word tasks, such as deciding whether nonsense words, such as "pleak" and "leeze," could have the same sound.
The imaging protocol involved a standard fMRI EPI-planar pulse sequence (TR/TE 3000/50 msec) with 21 6-mm-thick slices acquired with a 1-mm gap between each. Data were processed using FSL software from the Oxford (University) Center for Functional Magnetic Resonance Imaging of the Brain in the U.K. FSL software and custom software written by Richards were used to calculate brain correlation values from user-defined seed points. Each functional scan lasted five minutes, 42 seconds.
Baseline scans revealed that compared to normal controls, the dyslexic children exhibited significantly more connectivity between left inferior frontal gyrus seed points and the right inferior frontal gyrus while completing word tasks.
"The results were surprising because we were expecting to see deficits in connectivity, not over-connectivity, in dyslexia," Richards said in an interview. "Some brain regions are too strongly connected functionally in children with dyslexia when they are deciding which sounds go with which letters."
Different areas of the brain must communicate with each other properly in order to decode written language, according to Berninger. The brain center for this communication is like the conductor of an orchestra, she said. The left inferior frontal gyrus may serve as the conductor for language, ensuring that information is shared with the correct timing and in the proper sequence. In people with dyslexia, that conductor is ineffective.
But the trial also demonstrated that the abnormal pattern observed in dyslexic children can be modified. Functional MRI was again performed on the dyslexic children after they participated in three weeks of alphabetic, phonetic, and word recognition training.
Scans performed after treatment revealed that the amount of connectivity from the left inferior frontal gyrus seed point decreased in the brains of children with dyslexia to become more like the connectivity patterns of the control subjects, suggesting that focused instruction can normalize this aspect of brain activity among dyslexic children.
"We had hints in previous studies that the ability to decode novel words improves when a specific brain region in the right hemisphere decreases in activation," Richards said. "This study suggests that the deactivation may result in a disconnection in time from the comparable region in the left hemisphere, which in turn leads to improved reading. Reading requires sequential as well as simultaneous processes."
The study was reported in the October issue of the Journal of Neurolinguistics.
For more information from the Diagnostic Imaging archive:
Brain imaging focuses on connectivity and activation
Functional MRI snoops out how, why we decide to buy
Schizophrenic patients who hear voices demonstrate abnormalities in voice-processing brain regions