Learning to read requires the mapping and integration of visual (letters) and auditory (speech sounds) signals. Tracking how the brain's multisensory processing network reorganises during the earliest stages of formal literacy instruction can reveal experience-dependent neuroplasticity underlying successful reading acquisition. The authors used longitudinal magnetic resonance imaging to follow children beginning to learn to read, focusing on functional and structural changes tied to audiovisual integration.
The study followed twenty-three German-speaking children who were six years old at study entry. Participants completed four fMRI sessions distributed across a 12-month period that corresponded to the first year of formal literacy instruction. The longitudinal design allowed assessment of within-subject changes in neural responses as reading skills emerged.
During each fMRI session participants were presented with three stimulus types: single letters, single speech sounds, and audiovisual combinations of letters and speech sounds. Audiovisual combinations were presented in both congruent and incongruent formats. The primary functional contrast reported examined the difference between congruent and incongruent audiovisual conditions, intended to index neural sensitivity to matching versus mismatching letter–sound pairs.
The investigation employed 3 Tesla functional and structural MRI across the four sessions. Longitudinal changes in BOLD response amplitude were analyzed using linear mixed-effects models, enabling the authors to model session-wise trends and condition effects while accounting for repeated measures. The summary reports functional BOLD results; specific structural MRI measures, acquisition parameters, and detailed preprocessing or modeling steps were not provided in the source summary.
Across the 12-month period, the congruent-minus-incongruent contrast showed significant longitudinal modulation in a left-lateralised network. Regions with reported effects included the central opercular cortex, the anterior supramarginal gyrus, the inferior frontal gyrus (pars triangularis), and the frontal pole. The observed longitudinal effect pattern was primarily driven by a gradual increase in response amplitude across sessions for the congruent-versus-incongruent comparison.
The progressive increase in BOLD amplitude for congruent relative to incongruent audiovisual stimuli is interpreted as evidence of refinement in audiovisual integration mechanisms during early literacy acquisition. The implicated left-lateralised regions overlap with established language-processing and multisensory-integration areas, consistent with the idea that reading instruction recruits and reorganises language-related circuitry as children learn to link letters and sounds.
The study description indicates that both functional and structural MRI data were acquired longitudinally. However, the source summary focuses on functional BOLD changes and does not provide detailed structural MRI findings, measures (for example, cortical thickness or white-matter indices), or structural longitudinal effects. Likewise, specific behavioral measures of reading skill progression, correlations between neural change and reading outcomes, and exact timing of the four sessions relative to instruction milestones were not reported in the source abstract.
The text summarized here is an abstract of a preprint. It therefore omits many methodological specifics that are important for appraisal: precise MRI acquisition parameters, preprocessing pipelines, statistical thresholds, correction for multiple comparisons, and detailed structural MRI results were not described in the provided summary. The sample size is reported (n = 23), but demographic or behavioral detail beyond age and language (German-speaking six-year-olds) was not included in the abstract. As a preprint, the findings have not been peer reviewed.
In this longitudinal 3T MRI study of children beginning formal reading instruction, neural responses to congruent versus incongruent audiovisual letter–sound pairings increased over the first year of learning within a left-lateralised language and multisensory network. The authors interpret the rising BOLD amplitudes as a refinement of multisensory integration mechanisms as reading skills develop. The work illustrates the potential of longitudinal MRI to map experience-dependent functional reorganisation during early literacy, while structural results and other methodological details were not reported in the abstract.