The Links Between Language Exposure and Brain Structure in Early Childhood

• In Event: Poster Session 01
  In Poster Session: PS 01 Topic - Biological Processes: Neuroscience and Genetics

Abstract

Language exposure predicts children’s linguistic and cognitive skills, but the amount of linguistic input varies across socio-economic status (SES). Conversational experience has been associated (regardless of SES) with greater white matter across language related areas in childhood (Romeo et al., 2018) but is not known from how early language input impacts brain structure. Here, we investigate the relationship between language exposure and brain development at 6 and 30 months of age, while taking SES into account. We focus on the white matter tracts most associated with language processing and cognitive control, the superior longitudinal fasciculus (SLF) and the arcuate fasciculus (AF); we also consider whole-brain myelination. Our pre-registered hypotheses were that at both ages, amount of adult input and measures of conversational experience would be positively related to white matter concentrations along SLF and AF (H1). Also, we expected measures of conversational experience to be more relevant at older ages as language production increases, which would strengthen the relationship between turns and white matter in SLF and AF (H2).
To measure language exposure, we used the LENA software, focusing on the maximum hourly adult words, child vocalisations, and turns across several days of home recordings. Myelination in the brain was assessed using myelin water fraction values following the MRI protocol mcDESPOT (Deoni, 2015). We collected LENA and MRI data from 84 children from the two age groups.
For our priori analyses we ran linear models on language input measures and myelination in SLF and AF controlling for age and SES (z-score that included income and mean parental education). Analyses show relationships between adult input and age in the left AF, the right AF and the right SLF. Those relationships were positive for 30-month-old children, with more adult words relating to more myelin. However, 6-month-old children showed a negative trend, where more adult input was associated with less myelin (H1 is partially confirmed and H2 is not confirmed although positive links to language measures were stronger at 30mo). We also found negative relationships for child vocalisations in the left SLF, as well as SES interactions in the right hemisphere areas, with older children exposed to more adult input and from higher SES backgrounds, showing greater myelin concentrations than their peers.
Exploratory analysis investigated brain myelination in areas highlighted in prior studies. After correcting for multiple comparisons, only the left frontal prior showed a positive effect with the same pattern as before: more adult words were associated with more myelin at older ages and less myelin in younger children.
This study highlights the role of adult input early in life in relation to brain structure. We did not find any effects of conversational turns in our analyses; turns might be more influential later in life as infants’ language skills improve. We found surprising patterns for younger infants; this will require further investigation of our longitudinal dataset. Understanding the relationship between environmental factors and brain development in infancy is crucial, since neural plasticity in early childhood may be central to early intervention programs.

Authors

• Laia Fibla Reixachs, University of East Anglia
  Presenting Author

• Samuel H Forbes, Durham University
  Non-Presenting Author

• Larissa K Samuelson, University of East Anglia
  Non-Presenting Author

• Vincent A Magnotta, University of Iowa
  Non-Presenting Author

• Sean CL Deoni, Brown University
  Non-Presenting Author

• John P Spencer, University of East Anglia
  Non-Presenting Author