Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
The predictive brain in typical and impaired reading: an fMRI study of context effects and statistical learning in reading.
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster B120 in Poster Session B, Tuesday, October 24, 3:30 - 5:15 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Elisa Gavard1, Valérie Chanoine2, Franziska Geringswald2, Yufei Tan1, Eddy Cavalli3, Jean-Luc Anton4, Johannes Ziegler1; 1Aix-Marseille Université, CNRS, LPC, 2Aix-Marseille Université, CNRS, LPL, 3Université Lumière Lyon 2, EMC, 4Aix-Marseille Université, Centre IRM-INT@CERIMED
The predictive brain has become a key concept in language research and prediction is a dominant theoretical framework for understanding how the brain works. Psycholinguistic and neuroimaging research highlights the importance of anticipatory mechanisms in language comprehension, language production, and reading. In addition, a growing number of studies have used statistical learning (SL) paradigms to investigate whether SL abilities can explain inter-individual differences in language processing and reading abilities. The aim of the present study was to focus specifically on the neural network underlying predictive processes and to understand whether making semantic and syntactic predictions in reading relies on domain-general SL abilities and whether people with reading impairments (i.e., dyslexia) show deficits in these domains. To investigate a possible relationship between linguistic prediction and SL abilities in normal and impaired reading, we conducted an fMRI study with a serial reaction time (SRT) task and a predictive reading task in which 50 participants (25 typical readers and 25 dyslexic students) had to read aloud the same target words (e.g., mouse) either in a context of semantically related or unrelated words (cat – dog – rabbit – mouse vs. table – green – flower – mouse) or in the context of syntactically correct or incorrect sentences (she – likes – this – mouse vs. this – likes – she – mouse). We predicted that good readers should be good predictors in both linguistic (reading aloud) and non-linguistic (SRT) domains. It was an open question whether dyslexic students should be better predictors in order to compensate for their lower-level orthographic deficits, or whether they should be worse predictors, as suggested by some previous studies showing SL deficits in dyslexia. As very few neuroimaging studies have focused on semantic versus syntactic prediction, we wanted to dissociate the neural network underlying these predictive processes. It was also of interest to compare the neural underpinnings of linguistic and non-linguistic predictions in both typical and dyslexic readers. Preliminary results of the univariate analysis showed no differences between control and dyslexic participants in our predictive tasks. We also found an overlap between areas involved in the SRT task and the neural networks of semantic and syntactic prediction (specifically in the precentral gyrus and the postcentral gyrus). These results will be discussed in the context of current compensatory theories of dyslexia, the neural basis of reading and statistical learning, and the role of prediction in language processing.
Topic Areas: Reading, Disorders: Developmental