AI-supported English public speaking in the Chinese EFL context: Insights from self-regulated learning and positive psychology

IF 4.4 4区心理学 Q3 PSYCHOLOGY, BIOLOGICAL

Learning and Motivation Pub Date : 2025-11-01 Epub Date: 2025-11-03 DOI:10.1016/j.lmot.2025.102211

Hui Xun, Changyin Zhou, Yixuan Li, Yukun Wu

{"title":"AI-supported English public speaking in the Chinese EFL context: Insights from self-regulated learning and positive psychology","authors":"Hui Xun, Changyin Zhou, Yixuan Li, Yukun Wu","doi":"10.1016/j.lmot.2025.102211","DOIUrl":null,"url":null,"abstract":"<div><div>Positive affective factors, including enjoyment, engagement and self-efficacy etc., have been established as correlates of enhanced academic performance. While both self-regulated learning (SRL) and artificial intelligence (AI) demonstrate individual merits in promoting cognitive and affective outcomes, their synergistic potential within English public speaking (EPS) remains inadequately investigated. Grounded in SRL theory, this study aims to develop and examine an AI-Supported Self-Regulated EPS (AI-SREPS) model, thereby addressing this gap. A total of 88 first-year undergraduate students participated in the study. Two experimental groups received training based on the AI-SREPS model, while the control group received instruction of equal duration that focused on general introductions to AI tools for English language learning but did not involve the AI-SREPS model. Data were collected at three time points: pre-test, post-test, and a six-week follow-up, including questionnaires and English speech performances. To examine the effects of the AI-SREPS model, a series of statistical analyses were performed on Chinese EFL learners’ speech performance and positive affective factors, including a mixed-design repeated measures ANOVA, a Hierarchical Linear Model (HLM), and Bonferroni-adjusted post hoc comparisons with simple effect analyses. The results indicate that students in the AI-SREPS groups demonstrated significant improvements in speech content and language quality. They also reported higher emotional engagement, personal enjoyment, self-efficacy, and willingness to communicate (WTC). Nonetheless, challenges remained in fostering creativity in speech content, supporting behavioral delivery, and maintaining social support. This study makes contributions through the development of the AI-SREPS model, a longitudinal multi-group design that captures both immediate and sustained outcomes, and a multidimensional analysis cross-validating speech performance with questionnaire data. The results offer important implications for promoting the integrated development of cognitive, emotional, and social dimensions in EPS pedagogy.</div></div>","PeriodicalId":47305,"journal":{"name":"Learning and Motivation","volume":"92 ","pages":"Article 102211"},"PeriodicalIF":4.4000,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Learning and Motivation","FirstCategoryId":"102","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0023969025001183","RegionNum":4,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/11/3 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"PSYCHOLOGY, BIOLOGICAL","Score":null,"Total":0}

引用次数: 0

Abstract

Positive affective factors, including enjoyment, engagement and self-efficacy etc., have been established as correlates of enhanced academic performance. While both self-regulated learning (SRL) and artificial intelligence (AI) demonstrate individual merits in promoting cognitive and affective outcomes, their synergistic potential within English public speaking (EPS) remains inadequately investigated. Grounded in SRL theory, this study aims to develop and examine an AI-Supported Self-Regulated EPS (AI-SREPS) model, thereby addressing this gap. A total of 88 first-year undergraduate students participated in the study. Two experimental groups received training based on the AI-SREPS model, while the control group received instruction of equal duration that focused on general introductions to AI tools for English language learning but did not involve the AI-SREPS model. Data were collected at three time points: pre-test, post-test, and a six-week follow-up, including questionnaires and English speech performances. To examine the effects of the AI-SREPS model, a series of statistical analyses were performed on Chinese EFL learners’ speech performance and positive affective factors, including a mixed-design repeated measures ANOVA, a Hierarchical Linear Model (HLM), and Bonferroni-adjusted post hoc comparisons with simple effect analyses. The results indicate that students in the AI-SREPS groups demonstrated significant improvements in speech content and language quality. They also reported higher emotional engagement, personal enjoyment, self-efficacy, and willingness to communicate (WTC). Nonetheless, challenges remained in fostering creativity in speech content, supporting behavioral delivery, and maintaining social support. This study makes contributions through the development of the AI-SREPS model, a longitudinal multi-group design that captures both immediate and sustained outcomes, and a multidimensional analysis cross-validating speech performance with questionnaire data. The results offer important implications for promoting the integrated development of cognitive, emotional, and social dimensions in EPS pedagogy.

查看原文本刊更多论文

中国英语语境下人工智能支持的英语公开演讲：来自自主学习和积极心理学的见解

积极的情感因素，包括享受、投入和自我效能等，已被确立为提高学习成绩的相关因素。虽然自我调节学习（SRL）和人工智能（AI）在促进认知和情感结果方面都表现出各自的优点，但它们在英语公开演讲（EPS）中的协同潜力仍未得到充分研究。基于SRL理论，本研究旨在开发和检验人工智能支持的自我调节EPS （AI-SREPS）模型，从而解决这一空白。共有88名一年级本科生参与了这项研究。两个实验组接受基于AI- sreps模型的培训，而对照组接受同样时长的指导，重点介绍用于英语语言学习的AI工具，但不涉及AI- sreps模型。在三个时间点收集数据：测试前、测试后和六周的随访，包括问卷调查和英语演讲表现。为了检验AI-SREPS模型的效果，我们对中国英语学习者的语言表现和积极情感因素进行了一系列的统计分析，包括混合设计重复测量方差分析、层次线性模型（HLM）和bonferroni调整后的简单效应分析的事后比较。结果表明，AI-SREPS组的学生在语音内容和语言质量方面表现出显著的改善。他们还报告了更高的情感投入、个人享受、自我效能和沟通意愿（WTC）。然而，在培养演讲内容的创造性、支持行为传递和维持社会支持方面，挑战仍然存在。本研究通过开发AI-SREPS模型做出贡献，该模型是一种纵向多组设计，可捕获即时和持续的结果，并通过问卷数据交叉验证语音表现的多维分析。研究结果对促进EPS教学法中认知、情感和社会维度的整合发展具有重要意义。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Learning and Motivation Multiple-

CiteScore

2.90

自引率

0.00%

发文量

期刊介绍： Learning and Motivation features original experimental research devoted to the analysis of basic phenomena and mechanisms of learning, memory, and motivation. These studies, involving either animal or human subjects, examine behavioral, biological, and evolutionary influences on the learning and motivation processes, and often report on an integrated series of experiments that advance knowledge in this field. Theoretical papers and shorter reports are also considered.