{"title":"母语和二语读者的阅读理解:通过大型语言模型揭示的神经计算机制。","authors":"Chanyuan Gu, Samuel A Nastase, Zaid Zada, Ping Li","doi":"10.1038/s41539-025-00337-y","DOIUrl":null,"url":null,"abstract":"<p><p>While evidence has accumulated to support the argument of shared computational mechanisms underlying language comprehension between humans and large language models (LLMs), few studies have examined this argument beyond native-speaker populations. This study examines whether and how alignment between LLMs and human brains captures the homogeneity and heterogeneity in both first-language (L1) and second-language (L2) readers. We recorded brain responses of L1 and L2 English readers of texts and assessed reading performance against individual difference factors. At the group level, the two groups displayed comparable model-brain alignment in widespread regions, with similar unique contributions from contextual embeddings. At the individual level, multiple regression models revealed the effects of linguistic abilities on alignment for both groups, but effects of attentional ability and language dominance status for L2 readers only. These findings provide evidence that LLMs serve as cognitively plausible models in characterizing homogeneity and heterogeneity in reading across human populations.</p>","PeriodicalId":48503,"journal":{"name":"npj Science of Learning","volume":"10 1","pages":"46"},"PeriodicalIF":3.6000,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12246224/pdf/","citationCount":"0","resultStr":"{\"title\":\"Reading comprehension in L1 and L2 readers: neurocomputational mechanisms revealed through large language models.\",\"authors\":\"Chanyuan Gu, Samuel A Nastase, Zaid Zada, Ping Li\",\"doi\":\"10.1038/s41539-025-00337-y\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>While evidence has accumulated to support the argument of shared computational mechanisms underlying language comprehension between humans and large language models (LLMs), few studies have examined this argument beyond native-speaker populations. This study examines whether and how alignment between LLMs and human brains captures the homogeneity and heterogeneity in both first-language (L1) and second-language (L2) readers. We recorded brain responses of L1 and L2 English readers of texts and assessed reading performance against individual difference factors. At the group level, the two groups displayed comparable model-brain alignment in widespread regions, with similar unique contributions from contextual embeddings. At the individual level, multiple regression models revealed the effects of linguistic abilities on alignment for both groups, but effects of attentional ability and language dominance status for L2 readers only. These findings provide evidence that LLMs serve as cognitively plausible models in characterizing homogeneity and heterogeneity in reading across human populations.</p>\",\"PeriodicalId\":48503,\"journal\":{\"name\":\"npj Science of Learning\",\"volume\":\"10 1\",\"pages\":\"46\"},\"PeriodicalIF\":3.6000,\"publicationDate\":\"2025-07-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12246224/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"npj Science of Learning\",\"FirstCategoryId\":\"102\",\"ListUrlMain\":\"https://doi.org/10.1038/s41539-025-00337-y\",\"RegionNum\":1,\"RegionCategory\":\"心理学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"EDUCATION & EDUCATIONAL RESEARCH\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"npj Science of Learning","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.1038/s41539-025-00337-y","RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"EDUCATION & EDUCATIONAL RESEARCH","Score":null,"Total":0}
Reading comprehension in L1 and L2 readers: neurocomputational mechanisms revealed through large language models.
While evidence has accumulated to support the argument of shared computational mechanisms underlying language comprehension between humans and large language models (LLMs), few studies have examined this argument beyond native-speaker populations. This study examines whether and how alignment between LLMs and human brains captures the homogeneity and heterogeneity in both first-language (L1) and second-language (L2) readers. We recorded brain responses of L1 and L2 English readers of texts and assessed reading performance against individual difference factors. At the group level, the two groups displayed comparable model-brain alignment in widespread regions, with similar unique contributions from contextual embeddings. At the individual level, multiple regression models revealed the effects of linguistic abilities on alignment for both groups, but effects of attentional ability and language dominance status for L2 readers only. These findings provide evidence that LLMs serve as cognitively plausible models in characterizing homogeneity and heterogeneity in reading across human populations.