Language Environment and Infants' Brain Structure

The quality and quantity of language experienced during early childhood is a known predictor of brain development, long-term language achievement, and later cognitive and academic ability (Cartmill et al., 2013; Romeo et al., 2018). In early childhood, parental language input is the most frequent and important form of interaction (Barnes et al., 1983; Huttenlocher et al., 1991; Shneidman et al., 2013). In recent years, the quality of parental language input and parent–child interactions rather than word quantity has been put forth as the driving factor behind language achievement (Cartmill et al., 2013; Hirsh-Pasek et al., 2015). Recent work has found that early-childhood intervention of parental language input may be beneficial for long-term language achievement (Ferjan Ramírez et al., 2020). This “parental coaching” is focused on increasing the quality of parental language input through increased use of parentese (speaking with a higher pitch and slower tempo) as well as the number of parent–child conversational turns, in turn increasing overall interaction quality. Ferjan Ramírez et al. (2020) specifically reported that parental language input intervention increased the use of both parentese and conversational turn frequency. These behavioral changes in language were significantly correlated with language growth and overall language achievement at age 18 months. Other work has shown that parental language intervention in early childhood may be able to offset disadvantages that stem from lower familial socioeconomic status (SES), one of the most important determinants of overall health (Jensen et al., 2017; Romeo et al., 2018).

Conversational contingency is a key feature of positive parent–child interactions. Contingency is present when responses within a conversation are on topic and drive the interaction forward while stimulating sustained attention (Masek et al., 2021). Parental engagement is an important factor in conversational contingency and also improves child language ability (Hart and Risley, 1995; Hoff, 2003). Similarly, child engagement is related to higher-quality conversational output and overall language comprehension and production (Masek et al., 2021; Pagmar et al., 2022). Overall, behavioral research shows that in early childhood, language ability is positively correlated with responsive parenting wherein parent–child conversations are topically relevant and reliant on input from both parties (Bornstein et al., 2008; Masek et al., 2021).

Behavioral studies of parental language interventions and the importance of parental input have built a strong literature base in the field. Yet, little research has investigated the effects of parental language input and these interventions on structural and functional brain development in children <3 years of age (Romeo et al., 2018). Although parental language coaching has shown promise as an effective tool to augment childhood language ability, exactly how interventions augment language skills and when during development they could be most beneficial remains unclear. The recent work published by Huber et al. (2023) in The Journal of Neuroscience attempted to fill the existing literature gap by describing the effects of parental input on the brain between the ages of 6 and 24 months.

Using participants from an existing intervention study, Huber et al. (2023) used quantitative MRI data to analyze estimates of white matter myelination in children aged 6, 10, 14, 18, and 24 months (Ferjan Ramírez et al., 2020). The study focused on the relationship between measures of conversational quality and myelination in the dorsal language pathways of the brain. This study focused specifically on the arcuate fasciculus (AF) and superior longitudinal fasciculus (SLF)—frontotemporal and frontoparietal white matter tracts, respectively (Janelle et al., 2022). These are key tracts involved in language acquisition and production including connectivity between Broca's and Wernicke's areas (Janelle et al., 2022; Shekari and Nozari, 2023). Conversational data were collected using the home Language Environment Analysis System (LENA), which recorded parent–child conversations during 2 d when children were at the aforementioned ages. The data were separated into 100 30 s intervals of conversation per participant at each age. In total, 22 children took part in the study. Seventeen of the participants were part of the intervention study while 5 were members of a control group.

MRI acquisition included both diffusion-weighted imaging (DWI) and macromolecular proton fraction (MPF) mapping. In addition to the well established techniques and uses associated with DWI, MPF mapping offers added sensitivity to changes in myelin density (Kisel et al., 2022). MRI scans were performed after the completion of the intervention study at age 24 months (mean age, 26.30 months; SD, 1.62). Previous work in children 4–6 years of age has shown that myelination of the dorsal language pathway, namely white matter connectivity in the left AF and SLF, specifically adjacent to Broca's area are related to language ability independent of adult word count (Romeo et al., 2018). The current study attempted to build on these findings by identifying the relationship between the AF and SLF and language development and ability in younger children.

The MRI scan MPF and DWI data were compared with behavioral data from ages 6 to 24 months. All participants (intervention and control) were treated as a single group during analysis. Conversational turns were found to be correlated with MPF (left anterior AF) from 6 to 18 months and with MPF (left posterior SLF) from 6 to 14 months, but only the correlation between MPF (left anterior AF) and CT results at 18 months remained significant after multiple-comparison correction. In addition, MPF (left anterior AF) was predicted by individual growth in conversational turns from 6 to 24 months. Although no significant correlation was observed for the left arcuate or SLF, vocabulary was correlated with MPF percentage in the right inferior longitudinal fasciculus (ILF) and right inferior fronto-occipital longitudinal fasciculus (IFOF) at 24 months.

The right inferior frontal cortex has been implicated in word learning and early literacy (Bosseler et al., 2021; Shekari and Nozari, 2023). Specifically, the ILF is involved in the processing of complex visual information tied to word and emotion perception, and the IFOF has been noted as a key factor in reading, facial recognition, and spatial awareness (Catani, 2022). The exploratory findings of Huber et al. (2023) indicate early that markers of top-down control and visual/emotional processing are related to language ability, namely vocabulary size.

These findings alongside those of Fibla et al. (2023) suggest that fostering contingent conversations throughout infancy, particularly those with quality language, may enhance the development of myelinated circuits (Fibla et al., 2023; Huber et al., 2023). These increases in myelin density during early development may optimize or facilitate language-related cognitive abilities and learning in childhood development (Corrigan et al., 2022). This may be in part because of myelin growth being shaped as a result of neuronal function and vice versa, whereby refinement in circuitry because of white matter-induced neuroplasticity improves synchrony and speed of synaptic transmission (Wake et al., 2011; Pajevic et al., 2014; Noori et al., 2020). Myelin density may also determine individual differences in propensity for future language growth (Zuk et al., 2021). Notably, increases in myelin density observed in language-effective pathways during early development relative to other association white matter tracts are potentially enhanced by the social aspect of communication (Makinodan et al., 2012). This simultaneous maturation in the arcuate fasciculus and the IFOF may be a result of the receptive and productive language skills encompassing the learning experience throughout infancy (Tierney and Nelson, 2009).

The findings by Huber et al. (2023) support the notion that myelination increases are coupled to cognitive development. These myelination increases are not exclusive to the neonatal and infancy period, but may also shape adulthood learning (Chang et al., 2015). The current paradigm of myelination holds that the third trimester to the end of infancy (2 years) is the most sensitive period for alterations in myelin development; particularly of importance is how disruption of this critical period may lead to cognitive and learning issues in adulthood as seen in preterm births (Kennedy et al., 2021). A study by O'Muircheartaigh et al. (2014) noted that individual differences in cognitive abilities, particularly those of expressive and receptive language skills, showed a significant relationship to myelin volume fraction. This coupling may be facilitated by the seemingly programmed course of myelination of particular sensory regions from birth. Specifically, it has been suggested that the maturation of language comprehension and production regions shows parallel maturation rates along the temporofrontal language network throughout infancy (Pujol et al., 2006). This is in accordance with the evidence seen with regard to the theories of myelination, gradients from anterior to posterior regions (Grotheer et al., 2022). Myelination of the brain is nonuniform and dynamic; developmentally, it occurs in the posterior to anterior direction, whereby changes in functionally related white matter tracts are synchronized (Buyanova and Arsalidou, 2021).

The study by Huber et al. (2023) is also consistent with previous studies that have linked fractional anisotropy of SLF and AF with home language and literacy environment (HLLE) in infancy. HLLE is found to be a predictor of white matter tract organization specific to prereading and linguistic development; as well as a predictor of preliteracy skill development (Turesky et al., 2022). Interestingly, Turesky et al. (2022) measured SES using maternal education level in contrast to Huber et al. (2023), who measured SES using the Hollingshead index. Although measures of SES can vary in the accuracy and factors considered, the Hollingshead index used by Huber et al. (2023) is considered to be a more complete measure as it considers several additional factors, including occupation (Conway et al., 2019). Nevertheless, the two studies showed similar results underscoring the importance of the home language environment in infancy.

To conclude, this work by Huber et al. (2023) investigates the role of parental language input on cortical myelination at 24 months of age. The study found compelling evidence that the quality of conversation and, specifically, the number of conversational turns, are predictors of myelin levels in the dorsal language stream before 2 years of age. This study brings the field one step closer to elucidating whether there is a sensitive period for intervention to facilitate white matter development; implications may include early interventions for reducing speech impediments or to counter disadvantages because of SES (Jensen et al., 2017; Stephens et al., 2020). Future work that examines time frames and cortical structures that are particularly sensitive to the social–interactional variables of conversation are thus an important avenue to explore. Moreover, larger and more diverse samples (including language disorders; e.g., autism) should be considered as well as the use of participants from outside of parental input interventions to expand current findings. Finally, these findings provide compelling information for all parents and early childhood caregivers. Increased conversational turn-taking, child-directed speech, and use of parentese are documented predictors of later cognitive and academic achievement, and should be implemented as early as possible in both home and educational settings.