Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams

Greater learning-induced plasticity in children than adults during linguistic SL

Poster C27 in Poster Session C, Friday, October 7, 10:15 am - 12:00 pm EDT, Millennium Hall

Anqi Hu¹, Katherine Trice², Yi-Lun Weng¹, Zhenghan Qi²; ¹University of Delaware, ²Northeastern University

Statistical learning (SL), the ability to extract regularities from inputs, plays a key role in the development of language and literacy skills (Aslin & Newport, 2014; Erickson & Thiessen, 2015). The theoretical frameworks of SL have emphasized the contribution of both modality-specific and domain-general cognitive processes in SL (Conway, 2020), which raises important questions regarding how SL skills development across age. Previous behavioral study has not been able to differentiate between the age-invariant account and the less-is-more account. Even though linguistic SL has shown much less age effect measured by offline recognition accuracy, compared to nonlinguistic SL (Shufaniya & Arnon, 2018; Authors, 2022), children learned more quickly than adults only in the linguistic domain (Authors, 2022). In the current fMRI study, we aim to compare the engagement of linguistic-specific and domain-general attention networks in SL between children and adults. The less-is-more account predicts greater learning-induced changes in the developing language network in children, who have fewer years of language experiences and less mature attention network than adults. In contrast, the age-invariant account predicts similar learning-induced changes between children and adults. 18 children (Mean Age=7.76, SD=3.58, 11 Males) and 27 adults (9 Males) completed two auditory SL tasks in the scanner. In the linguistic Syllable SL task, participants listened to syllable sequences containing triplet patterns, intermixed with random tone sequences. In the nonlinguistic Tone SL task, participants listened to tone sequences containing triplet patterns, intermixed with random syllable sequences. Participants alternated between responding to a target syllable and responding to a target tone during learning. Participants’ BOLD responses were analyzed within the language network (LN; Fedorenko et al., 2010) and the dorsal attention network (DAN), which was isolated using a map generated by Neurosynth via an automated meta-analysis of 99 studies. In the LN, there was a greater learning-induced changes in the linguistic than the non-linguistic domain across both adults and children (b=-1.35, t=-8.63, p<0.000). The changes were greater in children than adults in the linguistic SL but not in the nonlinguistic SL task (Syllable: t(31.5)=-1.838, p <0.05). In both the LN and DAN, adults showed an overall greater activation than children across tasks and sequences (LN: b=-1.35, t=-8.63, p<0.001. DAN: b=-0.47, t=-9.33, p<0.001). Finally, stronger activation in the DAN was associated with better responses to the structured sequences in the LN across adults and children for both tasks (b=0.32, t=4.45, p<0.001). Together, these results provide the initial neural evidence supporting the less-is-more account of language acquisition. While adults showed greater overall engagements in both networks than children across both tasks, children’s LN showed a greater degree of learning-induced plasticity only for the linguistic domain. Moreover, greater attention engagement during the task is associated with greater learning-induced functional malleability in LN for both age groups, supporting a tight interplay between domain-general and domain-specific learning mechanisms underlying SL.

Topic Areas: Development, Perception: Auditory