Longitudinal analysis of letter and speech sound association

1028 

Abstract Submission 

Joanna Beck^1,2, Katarzyna Jednoróg¹

¹Laboratory of Neurobiology of Language, Nencki Institute of Experimental Biology PAS, Warsaw, Mazovia, ²Bioimaging Research Center, Institute of Physiology and Pathology of Hearing, Warsaw, Poland

Joanna Beck  
Laboratory of Neurobiology of Language, Nencki Institute of Experimental Biology PAS|Bioimaging Research Center, Institute of Physiology and Pathology of Hearing
Warsaw, Mazovia|Warsaw, Poland

Katarzyna Jednoróg  
Laboratory of Neurobiology of Language, Nencki Institute of Experimental Biology PAS
Warsaw, Mazovia

The initial step in reading development involves acquiring letter-speech sound (LS) associations, with literature consensus on the left superior temporal cortex as the integration site in adults and children (e.g., Blau et al., 2009). This research focuses on understanding the brain development of LS integration in children. Polish children from kindergarten to 8th grade participated in a computer-based letter-to-phoneme game, which measured their reaction time and accuracy in judging whether the letters and speech sounds were identical. Data showed that children master accurate LS association within a year of reading instruction. However, they needed more time (around three years) to automatize this ability, as reflected by decreasing reaction times in the game.
Longitudinal fMRI analysis (N = 67, age at TP1: mean = 6.93, TP3: mean = 8.92), in the prior study, revealed significant changes in brain activation during the first two years of education, indicating a shift in sensory areas activation. Brain activity decreased in response to unimodal speech sounds and letters but increased for multimodal LS pairs (Beck et al., under review).

Twenty-eight children from Beck et al. (under review) underwent an additional fMRI scan in the 8th grade (TP4: mean = 14.26). Using fMRI, brain activity was measured during the presentation of visual letters, speech sounds, congruent and incongruent LS pairs. Participants pressed a button for the word "cat," the picture of a cat, and the spoken word "/cat/."

LS association was measured on two levels:
A. Lower (basic sensory aspects) - multimodal enhancement - stronger response to multimodal congruent condition compared to unimodal conditions [ i.e super-additive effect (congruent LS > letters + speech sounds), max criterion (congruent LS > letters) ∩ (congruent LS > speech sounds) or mean criterion (2 * congruent LS > letters + speech sounds)] or opposite effect, multimodal suppression (i.e sub-additive effect).
B. Higher (orthographic and phonological): (in)congruency effect, i.e congruent vs. incongruent LS pairs

Pre-processing and analyses were conducted using BrainVoyager 22.2.1 software, with results reported at p < .005 corrected for multiple comparisons at the cluster level.

Extending prior analyses (Beck et al., under review), sensory areas exhibited decreased activity in response to unimodal speech sounds and letters, while multimodal LS pairs showed increased activity. Lower-level comparisons revealed engagement in the left inferior frontal gyrus and superior temporal gyrus for multisensory enhancement (super-additive effect). No significant differences were observed at the higher level of integration.

Contrary to expectations, the developmental trajectory of Polish children did not exhibit the anticipated congruency effect. While initial hypotheses suggested multisensory enhancement and (in)congruency effects as children automated this skill, our study only partially confirmed these expectations. Possible factors include insufficient number of participants, limited stimuli presentations, and potential fMRI paradigm influences. These results may also be linked to the unique timing of children's brain development, prompting further exploration of these factors and their implications.

Reading and Writing ¹

Normal Brain Development: Fetus to Adolescence ²

Activation (eg. BOLD task-fMRI)

Development

FUNCTIONAL MRI

Language

Learning

Other - Reading; Multimodal integration