Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams
Development of the language network in the brain
Poster E47 in Poster Session E, Saturday, October 8, 3:15 - 5:00 pm EDT, Millennium Hall
Also presenting in Poster Slam E, Saturday, October 8, 3:00 - 3:15 pm EDT, Regency Ballroom
Ola Ozernov-Palchik1, Amanda M. O’Brien*1,2, Rachel Romeo3, Hannah Small1, Benjamin Lipkin1, Jimmy Capella1, John D. E. Gabrieli1,2, Evelina Fedorenko1,2; 1Brain and Cognitive Sciences Department and McGovern Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA, 2Speech and Hearing Bioscience and Technology, Harvard University, Cambridge, Massachusetts, USA, 3Department of Human Development and Quantitative Methodology, University of Maryland, College Park, Maryland, USA
Adult language processing relies on a functionally-specialized left-lateralized fronto-temporal network (e.g., Fedorenko et al., 2011). How this network emerges over the course of development remains poorly understood. Although many prior studies have examined language processing in children (e.g., Blumenfeld et al., 2006; Wood et al., 2004; Friederici et al., 2011), most have relied on traditional group-averaging analyses (cf. Olulade et al., 2020), which assume voxel-wise correspondence across brains. Because the precise locations of functional areas differ across individuals, such analyses suffer from low sensitivity and functional resolution (Nieto-Castañón & Fedorenko, 2012) and may be especially problematic when comparing between age groups due to developmental changes in inter-individual variability. We use individual-subject fMRI analyses to investigate developmental differences in selectivity and lateralization of the language network. All participants (Children: N=171, 4-14 years, mean=10.9, SD=3.3; Adults: N=91, 19-45 years, mean=27.4, SD=5.5) were presented with a robust and extensively validated functional ‘localizer’ task (Fedorenko et al., 2010; Scott et al., 2017). During the task, participants listened to a language condition (engaging passages) and an incomprehensible auditory control condition (acoustically degraded or backward speech). We defined functional regions of interest (fROIs) in each individual based on the language>control contrast. Using pre-existing ‘masks’ (six per hemisphere, covering lateral frontal and temporal cortex) derived from a large independent sample of adults performing a similar task, we defined individual fROIs as the top 10% of most localizer-responsive voxels within each mask based on half of the data. Response magnitudes for the two conditions were extracted from these fROIs using the other half of the data. Two primary findings emerged. First, a reliable language>control effect was observed in each left-hemisphere language region in adults and children, including the youngest children, in the 4-5 age range. This result suggests that by age 4, both temporal and frontal language areas support language comprehension. However, the size of the language>control effect in the frontal, but not temporal areas, showed a developmental increase, consistent with slower functional maturation of the frontal cortex (Fuster, 2002). Secondly, across all ages, the magnitude of the language>control effect was larger in the left hemisphere than the right hemisphere, and unlike Olulade et al. (2020) and others (Holland et al., 2007), the degree of lateralization was not modulated by age. This result suggests that by age 4, the language network is strongly left-lateralized, as in adult brains. These findings suggest that some aspects of the language network’s architecture—including reliable responses to language in both temporal and frontal areas and left-hemisphere lateralization—are already in place by age 4, whereas other aspects—like the language-selective response in the frontal areas—continue to mature into mid/late childhood. Understanding whether/how these slower-maturing aspects affect linguistic and cognitive processing remains an important question for future work. Methodologically, this work establishes that functional localization is effective for identifying language-selective areas in children, and will likely confer similar benefits, relative to the traditional group-averaging approach, as it has in adult language processing research (e.g., Nieto-Castañón & Fedorenko, 2012; Gratton & Braga, 2021).
Topic Areas: Development, Methods