Neural initialization of audiovisual integration in prereaders at varying risk for developmental dyslexia

Abstract

Learning letter-speech sound correspondences is a major step in reading acquisition and is severely impaired in children with dyslexia. Up to now, it remains largely unknown how quickly neural networks adopt specific functions during audiovisual integration of linguistic information when prereading children learn letter-speech sound correspondences. Here, we simulated the process of learning letter-speech sound correspondences in 20 prereading children (6.13-7.17 years) at varying risk for dyslexia by training artificial letter-speech sound correspondences within a single experimental session. Subsequently, we acquired simultaneously event-related potentials (ERP) and functional magnetic resonance imaging (fMRI) scans during implicit audiovisual presentation of trained and untrained pairs. Audiovisual integration of trained pairs correlated with individual learning rates in right superior temporal, left inferior temporal, and bilateral parietal areas and with phonological awareness in left temporal areas. In correspondence, a differential left-lateralized parietooccipitotemporal ERP at 400 ms for trained pairs correlated with learning achievement and familial risk. Finally, a late (650 ms) posterior negativity indicating audiovisual congruency of trained pairs was associated with increased fMRI activation in the left occipital cortex. Taken together, a short (<30 min) letter-speech sound training initializes audiovisual integration in neural systems that are responsible for processing linguistic information in proficient readers. To conclude, the ability to learn grapheme-phoneme correspondences, the familial history of reading disability, and phonological awareness of prereading children account for the degree of audiovisual integration in a distributed brain network. Such findings on emerging linguistic audiovisual integration could allow for distinguishing between children with typical and atypical reading development. Hum Brain Mapp 38:1038-1055, 2017. © 2016 Wiley Periodicals, Inc.

Emergence of the neural network underlying phonological processing from the prereading to the emergent reading stage: A longitudinal study

“…The VWFS is located in the left occipitotemporal cortex, and a posterior‐to‐anterior hierarchical gradient for the representation of increasingly larger word fragments from simple letters in the posterior fusiform/inferior occipital gyri to whole words in the middle fusiform gyrus has been reported (Vinckier et al, ). Recent developmental studies have further suggested that the specialization process of the VWFS starts in the posterior area, as the print‐sensitive activation has been shown to emerge in this region (Brem et al, ; James, ) and correlate with electrophysiological responses associated with letter‐sound congruency effects (Karipidis et al, ) in prereaders after preliteracy training. In line with these findings, the functional network between the LIPC and the posterior VWFS (i.e., LpOTC) observed in the current study may reflect the establishment of the neural pathway for letter‐sound correspondence in our very young participant sample, following their decoding/reading experience over the developmental time course of reading acquisition.…”

Reading-induced shifts of perceptual speech representations in auditory cortex

“…Furthermore, the amplitude of unimodal (speech) and crossmodal (text-speech) responses in this region has been found to scale with individual differences in phonological and/or reading skills in typical readers 7, 15, 16 and pre-readers 17 , and to show an overall reduction in dyslexic readers 14, 18, 19 . However, it remains debated whether and how learning to read changes the representation of speech at the level of the auditory cortex 1, 20 .…”

Rapid changes in brain activity during learning of grapheme-phoneme associations in adults

“…This was carried out for the training and testing blocks in 3 different learning stages on Day 1 and Day 2. Based on earlier literature (I Karipidis et al 2017;Xu et al 2019;Raij et al 2000) brain dSPM source waveforms (500ms to 800ms after stimulus onset) were extracted from left and right superior temporal cortices (labels: "superiortemporal" + "bankssts") Blomert 2011;van Atteveldt et al 2009;Xu et al 2019; as defined by the Desikan-Killiany Atlas (Desikan et al 2006).…”

“…Interestingly the fast learning-related changes in brain activity seem to be linked to cognitive performance (Karipidis et al 2018;I Karipidis et al 2017). Multisensory integration effects were found in a distributed brain network after a short letter-speech sound training (<30 min) in preschool children (I Karipidis et al 2017) with promising implications for identifying poor-reading children and predicting reading outcomes in pre-readers (Karipidis et al 2018).…”

“…Artificial grapheme-phoneme training paradigms that simulate the initial stage of learning to read in alphabetic scripts have provided interesting insights into the brain mechanisms of learning graphemephoneme associations. Learning-related brain changes have been reported at the time scale of minutes (Hä mä lä inen et al 2019; I Karipidis et al 2017), hours (Taylor et al 2014;Brem et al 2018) and days Hashimoto and Sakai 2004;Madec et al 2016;Quinn et al 2017;Karipidis et al 2018) after initial training. Combining information from visual and auditory modalities has been suggested to involve at least two possible mechanisms: (Hashimoto and Sakai 2004) reported involvement of left posterior inferior temporal gyrus (PITG) and left parieto-occipital cortex (PO) which showed plasticity for forming new links between orthography and phonology when learning novel letters.…”