Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Cooperative and Instructor-Led Learning of Chinese Idioms Engage Distinct Neural Coordination Patterns among Autistic Children — an fNIRS hyperscanning study
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
Xin Zhou1, Hao Huang1, Jim S. F. Lau1,2, Pearl C. K. Lam2, Patrick C. M. Wong1,2; 1Brain and Mind Institute, The Chinese University of Hong Kong, 2Department of Linguistics and Modern Languages, the Chinese University of Hong Kong, Hong Kong SAR, China
Introduction: Idioms are one type of figurative language that often conveys meaning beyond their literal interpretation, and are frequently used in our daily communications and literacy. Idiom learning and comprehension can be particularly challenging for children with autism spectrum disorder (ASD) and global developmental delay (DL), due to their deficits in linguistic abilities, inferencing abilities, and executive functioning. Limited research has focused on Chinese idioms, most of which are quadrasyllabic, highly synthetic, compact, and not decomposable based on the semantics, and may impose additional challenges on young autistic and DL children, compared to their typically developing (TD) peers. Adhering to inclusive education guidelines, most autistic and DL children attend mainstream schools. Cooperative learning (CL) has been reported to have a small-to-medium effect on the academic achievement of school-aged children, compared to instructor-led learning (IL). However, no research has examined whether CL would facilitate idiom learning among autistic and DL children, and how their presence would affect the classroom dynamics. To address these gaps, we conducted an fNIRS hyperscanning study, by simultaneously measuring the brain activity of children in small groups to measure their inter-brain coherence (IBC) during social interactions. Methods: Sixty-three quartets of 4 children (n = 252; each with two boys/girls aged 6-7), including 22 TD, 21 ASD, and 20 DL quartets, completed data acquisition. To simulate a real-world small group with the presence of ASD or DL children, each ASD/DL quartet included one child with ASD/DL and three TD children. A 2*2*2 design was implemented to examine the effect of teaching mode (CL versus IL), decomposability (decomposable versus nondecomposable), and emotion (positive versus negative) on their learning outcome, with a total of 8 idioms being taught per quartet. fNIRS hyperscanning data were recorded in four regions of interest (ROIs) within the left and right frontoparietal networks to compute their IBC. Twelve models were constructed using a support vector machine with recursive feature elimination (SVM-RFE) to classify TD, ASD, and DL groups, separately for the CL and IL modes, using behavioral and IBC measures combined, and behavioral measures alone. Results: Models including behavioral and IBC measures outperformed behavioral-only models, achieving AUCs ranging from 0.794 to 0.907. Across three child group contrasts, the SVM-RFE models under the CL and IL modes did not share any significant neural patterns. However, the unique features within the CL and IL models accounted for 37-59% of model significance. Specifically, when comparing ASD vs. TD, and ASD vs. DL, more same-ROI connections showed significance in the CL mode and when learning positive and decomposable idioms, and more cross-ROI connections showed significance in the IL mode and when learning nondecomposable idioms, suggesting that the two teaching modes engage distinguished neural networks and favor the learning of different idioms. Conclusion: This is the first fNIRS hyperscanning study that investigates small-group classroom learning involving children with ASD and DL. Our findings shed light on the presence of an ASD and DL child in small groups on group dynamics and how two teaching modes favor idioms of varying decomposability.
Topic Areas: Language Development/Acquisition, Disorders: Acquired