Unpacking AI-supported Chinese as a foreign language learning: How beginner-level learners' cognitive and motivational factors predict speaking proficiency.
{"title":"Unpacking AI-supported Chinese as a foreign language learning: How beginner-level learners' cognitive and motivational factors predict speaking proficiency.","authors":"Jie Zhang, Xiaosheng Zhou, Ying Soon Goh","doi":"10.1016/j.actpsy.2025.105703","DOIUrl":null,"url":null,"abstract":"<p><p>This study examines the cognitive and motivational factors that influence Mandarin speaking proficiency among learners in an artificial intelligence (AI)-supported, project-based Chinese as a Foreign Language (CFL) learning environment. It specifically investigates how learners' psychological engagement with AI tools relates to language performance. A mixed-methods explanatory design was adopted. Quantitative data were collected from 120 beginner-level CFL university students in Southeast Asia and analyzed using correlation and multiple regression techniques. Speaking proficiency was assessed through both script writing and oral project presentations. To enrich interpretation, qualitative data were obtained from interviews, learner-generated artifacts, and instructor observations with a purposive subsample of 30 learners. Results indicated that technical proficiency, perceived usefulness, and positive attitudes toward AI were significant positive predictors of speaking performance. In contrast, high levels of trust in AI were negatively associated with outcomes, suggesting diminished learner autonomy. A strong correlation emerged between AI usage and script writing scores, while a weaker but significant relationship was observed for oral presentation. Notably, written and oral performance were not significantly correlated, highlighting distinct developmental trajectories in these modalities. Qualitative results revealed that learners who engaged critically and strategically with AI achieved better outcomes than those who relied on it passively. This study contributes to understanding the psychological mechanisms underlying AI-assisted language learning. It highlights the importance of fostering mindful and autonomous engagement with technology to support speaking proficiency, and underscores the need for skill-specific instructional strategies in AI-integrated educational settings.</p>","PeriodicalId":7141,"journal":{"name":"Acta Psychologica","volume":"260 ","pages":"105703"},"PeriodicalIF":2.7000,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Acta Psychologica","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.1016/j.actpsy.2025.105703","RegionNum":4,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PSYCHOLOGY, EXPERIMENTAL","Score":null,"Total":0}
引用次数: 0
Abstract
This study examines the cognitive and motivational factors that influence Mandarin speaking proficiency among learners in an artificial intelligence (AI)-supported, project-based Chinese as a Foreign Language (CFL) learning environment. It specifically investigates how learners' psychological engagement with AI tools relates to language performance. A mixed-methods explanatory design was adopted. Quantitative data were collected from 120 beginner-level CFL university students in Southeast Asia and analyzed using correlation and multiple regression techniques. Speaking proficiency was assessed through both script writing and oral project presentations. To enrich interpretation, qualitative data were obtained from interviews, learner-generated artifacts, and instructor observations with a purposive subsample of 30 learners. Results indicated that technical proficiency, perceived usefulness, and positive attitudes toward AI were significant positive predictors of speaking performance. In contrast, high levels of trust in AI were negatively associated with outcomes, suggesting diminished learner autonomy. A strong correlation emerged between AI usage and script writing scores, while a weaker but significant relationship was observed for oral presentation. Notably, written and oral performance were not significantly correlated, highlighting distinct developmental trajectories in these modalities. Qualitative results revealed that learners who engaged critically and strategically with AI achieved better outcomes than those who relied on it passively. This study contributes to understanding the psychological mechanisms underlying AI-assisted language learning. It highlights the importance of fostering mindful and autonomous engagement with technology to support speaking proficiency, and underscores the need for skill-specific instructional strategies in AI-integrated educational settings.
期刊介绍:
Acta Psychologica publishes original articles and extended reviews on selected books in any area of experimental psychology. The focus of the Journal is on empirical studies and evaluative review articles that increase the theoretical understanding of human capabilities.