Understanding the time variant connectivity of the language network in developmental dyslexia: new insights using Granger causality

Abstract

The reading process takes place in a neuronal network comprising the inferior frontal, posterior dorsal and posterior ventral brain areas. It is suggested that developmental dyslexia is caused by a disruption of the two posterior network areas. What remains debatable is whether these areas are affected in their functionality or whether the neuronal networking (connectivity) of these areas suffer from a disturbed information transfer. Thus, it is of major interest to investigate the time flow of the directed information transfer (time variant connectivity) within the neuronal reading network of dyslexic subjects. We investigated adolescents with dyslexia and normal-reading controls with functional magnetic resonance imaging and electroencephalography (EEG) with a paradigm addressing basic visual, orthographic and phonological processing. EEG data were analyzed with the time variant Granger causality index (tvGCI) to investigate the temporal order of the directed information transfer (time variant causal connectivity: which network node passes when information to which network node) during reading in dyslexic readers. Results show that the reading network of dyslexic readers comprises the same brain areas as identified in normal-reading subjects. The tvGCI analysis of the network profiles of dyslexic readers indicates that dyslexics show a difference in timing and localization of connectivity within this reading network compared to normal readers. Dyslexic readers use right hemisphere language areas to counterbalance posterior left hemisphere processing deficits. The compensatory involvement of homologue right hemisphere brain areas for the reading process may be the neurobiological background for the significantly longer reading times by dyslexics.