Literacy skills are crucial for children's development. Individual differences in reading development, observable even in preschool, highlight the importance of the home literacy environment (HLE). HLE, encompassing literacy-related interactions, resources, and attitudes, is often assessed through parental questionnaires. Studies consistently show a positive association between supportive HLE and enhanced vocabulary and reading skills in children, suggesting vocabulary may mediate the HLE-reading relationship. Neuroimaging studies have linked socioeconomic status (SES), a proxy for HLE quality, to structural and functional differences in brain regions associated with literacy, particularly the left perisylvian areas, including the left inferior frontal gyrus (IFG). Direct investigations into the relationship between HLE and brain development using parental questionnaires or home recordings show associations between home reading exposure, maternal shared reading quality, and activity in left perisylvian regions during language tasks. However, there's limited evidence directly linking HLE to brain activity during word recognition tasks in children. This study aims to investigate this relationship using fMRI and a word adaptation task in French-speaking 8-year-olds, testing if HLE is associated with fMRI activity during word recognition and if this relationship is mediated by vocabulary skills. The word adaptation task, previously used with adult English speakers, involves passively presenting blocks of identical (adaptation) or different (no-adaptation) words, measuring neural adaptation as the difference in fMRI activity between these blocks. The study hypothesizes that more frequent HLE is associated with greater neural adaptation to words, mediated by vocabulary skills, and this effect is specific to word recognition (not general symbolic processing).

Existing research strongly supports the link between the home literacy environment (HLE) and children's literacy skills. Studies using parental questionnaires have consistently demonstrated that children exposed to richer HLEs exhibit improved vocabulary and reading skills. The mediating role of vocabulary in this relationship is a recurring theme, suggesting that frequent home literacy activities may enhance reading efficiency by fostering stronger vocabulary foundations. Neuroimaging research adds another layer to this understanding, showing correlations between socioeconomic status (SES), often a proxy for HLE quality, and brain structure and function in regions vital for literacy. Lower SES is often associated with reduced cortical thickness and surface area in left perisylvian regions. Previous neuroimaging studies have examined HLE’s impact on brain development through various measures, such as parental questionnaires and home recordings. These studies have linked HLE to white matter integrity and left perisylvian activity during story-listening tasks in preschoolers. However, many of these studies focus on pre-reading children or use auditory tasks, not directly assessing the relationship between HLE and word recognition in school-aged children. This gap in the literature necessitates the current study.

This study used a two-session design. In the first session, parents (n=73) completed questionnaires assessing the frequency of various home activities, including literacy practices (using a 6-point Likert scale). Children's (ages 7.52-9.22 years) reading fluency and vocabulary were assessed using standardized tests (Alouette-R and NEMI-2). Seven children were excluded due to speech therapy, low IQ, speech delays, or ADHD, resulting in a behavioral sample of 66 children. In the second session, a subset of the children (n=58) underwent fMRI. Fourteen children were excluded from fMRI analysis due to incomplete data or excessive motion, leaving a final fMRI sample of 44 children. fMRI data was acquired using a Siemens Prisma 3T MRI scanner during a word adaptation task (adapted from Perrachione et al., 2016), which involved passively viewing blocks of identical (adaptation) or different (no-adaptation) monosyllabic nouns. A control digit adaptation task used Arabic numerals (1-8) instead of words. A target detection task was embedded to assess attention. fMRI data was preprocessed using SPM12, including outlier removal and normalization to the MNI space. Statistical analysis used a GLM to model brain activity during adaptation and no-adaptation blocks. Word adaptation effects were calculated by subtracting adaptation block activity from no-adaptation block activity. Whole-brain analyses identified regions showing significant adaptation effects, and ROI analyses focused on regions identified by Perrachione et al. (2016) in typical adult readers. Correlations assessed the relationship between home literacy practices, vocabulary, and neural adaptation in the ROIs. Mediation analysis examined whether vocabulary mediated the relationship between home literacy practices and word adaptation. Additional control analyses investigated the specificity of findings to word recognition (vs. digit processing), the influence of target detection rates, parental perception of child's reading skills, and parental reading skills.

Behavioral analyses revealed a small but significant positive correlation between overall frequency of home literacy practices and children's vocabulary scores (partial r(64) = 0.250, p = 0.023), but no significant relationship with reading fluency (accuracy or speed). fMRI analyses replicated Perrachione et al.'s (2016) findings, showing significant word adaptation effects in a left-lateralized network including the left pIFS, pdSTS, and pTF. Crucially, more frequent home literacy practices were positively associated with greater word adaptation in the left pIFS (r(42) = 0.322, Pcorr = 0.048), even after controlling for SES. This relationship remained significant after controlling for parental income and education (partial r(42) = 0.302, p=0.026). Vocabulary was positively correlated with word adaptation in the left pIFS (partial r(42) = 0.317, p=0.020). Mediation analysis confirmed that vocabulary mediated the relationship between home literacy practices and word adaptation in the left pIFS. This was statistically significant (path ab, coefficient = 0.015, STE = 0.014, 95% CI Lower = 0.001, p = 0.046). Control analyses using a digit adaptation task showed no significant relationships between digit adaptation and home literacy practices or vocabulary, highlighting the specificity of the findings to word recognition. Additional analyses controlled for potential confounders such as task engagement (target detection rates) and parental perception of the child's reading skills, but the key findings remained robust.

This study provides the first evidence directly linking home literacy practices to fMRI activity during a word recognition task in children. The positive association between home literacy practices and neural adaptation in the left pIFS supports the idea that early HLE significantly influences brain regions involved in reading. The mediating role of vocabulary clarifies the mechanism through which HLE impacts reading-related brain activity; richer HLE leads to better vocabulary, which in turn enhances sensitivity to repeated printed words in the brain. The lack of similar effects in the digit adaptation task emphasizes the specificity of the effect to word processing. The findings align with models proposing that HLE's effects on literacy are mediated by foundational skills like vocabulary. Although the observed effect sizes were relatively small, the consistency across analyses strengthens the results. These findings have important implications for interventions aimed at improving literacy outcomes by targeting HLE.

This study demonstrates a significant association between home literacy practices, vocabulary development, and neural activity related to word recognition in 8-year-old children. The findings highlight the mediating role of vocabulary in this relationship and emphasize the importance of early HLE for fostering reading skills. Future longitudinal studies should track these relationships from younger ages and examine their stability across different age groups, tasks, and reading abilities. This research could benefit from larger samples and incorporating more comprehensive HLE measures and reading-related outcomes to further refine our understanding of the multifaceted relationship between HLE and reading development.

This study's cross-sectional design prevents causal inferences. While the results support the proposed model, other interpretations remain possible. The reliance on parental questionnaires may introduce recall and social desirability biases. The relatively small sample size, focused age range (8-year-olds), and limited reading assessment (word recognition only) restrict the generalizability of the findings. Future studies should adopt longitudinal designs, more objective HLE assessments, and broader reading assessments. Addressing these limitations will enhance the understanding and application of these findings.