SRLAgent: Enhancing Self-Regulated Learning Skills through Gamification and LLM Assistance Conference: Preprint; 2025;
Wentao Ge, Yuqing Sun, Ziyan Wang, Haoyue Zheng, Weiyang He, PiaoHong Wang, Qianyu Zhu, Benyou Wang · 2025
Read SRLAgent: Enhancing Self-Regulated Learning Skills through Gamification and LLM Assistance Conference: Preprint; 2025; by Wentao Ge, Yuqing Sun, Ziyan Wang, Haoyue Zheng, Weiyang He, PiaoHong Wang, Qianyu Zhu, Benyou Wang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-09968.txt.
Introduction
Wentao Ge and seven co-authors present SRLAgent, a system to build self-regulated learning skills in college students. Their twenty twenty-five paper starts from a survey of fifty-nine students. Many struggled to set goals, manage time, and reflect on their study, skills that matter once classes demand more independence.
SRLAgent wraps study in a game-like setting guided by a language model. It follows a three-phase frame: set goals, carry out strategies, then reflect. The system gives live feedback and support as students work. In a controlled test, students with SRLAgent gained more in these skills and stayed more engaged than two comparison groups.
Catalog record
| CREATOR | Wentao Ge, Yuqing Sun, Ziyan Wang, Haoyue Zheng, Weiyang He, PiaoHong Wang, Qianyu Zhu, Benyou Wang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 78 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-09968
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
SRLAgent: Enhancing Self-Regulated Learning Skills through Gamification and LLM Assistance Conference: Preprint; 2025; Wentao Ge, Yuqing Sun, Ziyan Wang, Haoyue Zheng, Weiyang He, PiaoHong Wang, Qianyu Zhu, Benyou Wang Abstract . Self-regulated learning (SRL) is crucial for college students navigating increased academic demands and independence. Insufficient SRL skills can lead to disorganized study habits, low motivation, and poor time management, undermining learners’ ability to thrive in challenging environments. Through a formative study involving 59 college students, we identified key challenges students face in developing SRL skills, including difficulties with goal-setting, time management, and reflective learning. To address these challenges, we introduce SRLAgent, an LLM-assisted system that fosters SRL skills through gamification and adaptive support from large language models (LLMs). Grounded in Zimmerman’s three-phase SRL framework, SRLAgent enables students to engage in goal-setting, strategy execution, and self-reflection within an interactive game-based environment. The system offers real-time feedback and scaffolding powered by LLMs to support students’ independent study efforts. We evaluated SRLAgent using a between-subjects design, comparing it to a baseline system (SRL without Agent features) and a traditional multimedia learning condition. Results showed significant improvements in SRL skills within the SRLAgent group ( p < .001, Cohen’s d = 0.234), and higher engagement compared to the baselines. This work highlights the value of embedding SRL scaffolding and real-time AI support within gamified environments, offering design implications for educational technologies that aim to promote deeper learning and metacognitive skill development. 1. Introduction In higher education, students are expected to take ownership of their learning—setting goals, managing their progress, and reflecting on outcomes. This ability, known as self-regulated learning (SRL) ( Zimmerman 2002 ; Pintrich 2004 ; Harris and Graham 1999 ; Schraw et al. 2006 ; Schunk 1996 ) , is a critical competency in college settings, where students face significantly greater demands for autonomy and self-direction than in high school. ( Appleby 2006 ; Song and Vermunt 2021 ) . Unlike the structured and teacher-guided learning environment common in secondary education, college students must proactively plan, manage, and reflect upon their learning activities with minimal external supervision ( Conley 2008 ; Venezia and Jaeger 2013 ; Credé and Kuncel 2008 ) . Prior research has demonstrated that effective SRL practices are strongly connected to students’ academic achievement ( Mega et al. 2014 ; Dent and Koenka 2016 ; Schunk and Zimmerman 2011 ; Broadbent and Poon 2015 ) , study satisfaction ( Liborius et al. 2019 ) , and long-term educational success ( Theobald 2021 ) . Conversely, inadequate SRL skills among college students frequently lead to disorganized study habits, reduced intrinsic motivation, poor time management, lower academic achievement overall ( Credé and Kuncel 2008 ; Mega et al. 2014 ; Zusho 2017 ; Dent and Koenka 2016 ; Schunk and Zimmerman 2011 ; Dunlosky and Rawson 2012 ) , and addictive behaviours ( Wang et al. 2020 ) . Cultivating SRL skills is therefore a critical priority in supporting college students’ academic development ( Credé and Kuncel 2008 ; Zusho 2017 ; Broadbent and Poon 2015 ; Broadbent 2017 ) . Previous research has explored various interventions and tools aimed at enhancing college students’ self-regulated learning (SRL) skills. Educational interventions such as structured training sessions, workshops, and courses explicitly targeting SRL strategies have demonstrated improvements in student self-regulation and academic outcomes ( Bail et al. 2008 ; Paris and Paris 2003 ) . Additionally, technology-driven solutions including mobile applications ( Broadbent and Poon 2015 ) , web-based platforms ( Bellhäuser et al. 2016 ) , and AI-driven tutoring systems ( Ng et al. 2024 ; Hsu et al. 2024 ; Molenaar 2022 ) have emerged to provide personalized feedback, task management, and performance analytics aimed at fostering SRL. However, despite these advances, existing interventions exhibit several limitations. Explicit instructional programs often lack sustained engagement or fail to effectively motivate students, leading to limited transfer and application of SRL strategies beyond initial training contexts ( Järvelä and Hadwin 2013 ) . Similarly, technology-based platforms frequently provide generic feedback or overly rigid scaffolds, lacking the adaptability necessary to address individual learner differences and emerging needs during real-time learning activities. Furthermore, many existing solutions do not adequately integrate motivational and engagement elements, such as gamification features, which could enhance learners’ intrinsic motivation and sustained participation ( Landers 2014 ; Koivisto and Hamari 2019 ) . Consequently, there is a critical need for innovative approaches that combine personalized, adaptive feedback mechanisms with engaging, motivationally supportive elements to effectively cultivate college students’ SRL skills. Building on these limitations, we conducted a formative study to investigate the specific design needs and better understand the challenges college students encounter while developing SRL skills. A total of 59 undergraduate students from diverse majors participated in this study, which incorporated a mixed-methods approach involving semi-structured interviews and surveys. Based on findings from the formative study, we developed SRLAgent, a Minecraft-based interactive system designed to foster students’ SRL skills with large language model(LLM) assistance. Grounded in Zimmerman’s widely accepted three-phase model of SRL ( Zimmerman 2002 ) , SRLAgent guides students through goal-setting (forethought), strategic monitoring and active learning (performance), and reflective evaluation of their learning experiences (reflection). By embedding SRL strategies directly within engaging gameplay, SRLAgent aims to seamlessly integrate skill practice into students’ authentic learning experiences. To assess the effectiveness of our proposed system, we conducted a between-subjects study comparing SRLAgent with two baseline conditions: a baseline version of SRLAgent without SRL support features and traditional multimedia learning. The study results showed that SRLAgent effectively promotes learners’ abilities in forethought and performance stages of Zimmerman’s SRL model, highlighting the value of explicit goal-setting and real-time feedback mechanisms. Descriptive trends indicate that SRLAgent may positively influence learner engagement and trust in AI systems, suggesting promising directions for future research and design improvements. This paper contributes to the field of educational technology and human-computer interaction (HCI) as follows: 1) Development of SRLAgent , an LLM-assisted interactive system designed to foster students’ SRL skills; 2) Implementation of a between-subjects user study to evaluate SRLAgent’s effectiveness and usefulness; 3) Design implications derived from the design process and user studies, offering guidelines for designing interactive systems that leverage generative AI and gamification. 2. Related Work 2.1. Self - Regulated Learning Strategies in Higher Education Empirical research has consistently shown the positive effect of SRL on academic achievement ( Azevedo and Cromley 2004 ; Chen and Huang 2014 ; Ergen and Kanadlı 2017 ; Broadbent and Poon 2015 ; Dent and Koenka 2016 ; Mega et al. 2014 ; Schunk and Zimmerman 2011 ) . Particularly, SRL is widely regarded as a foundational competency in higher education, where students are expected to take increasing responsibility for their learning and external guidance is reduced compared to secondary education ( Xu et al. 2023 ; Song and Vermunt 2021 ; Conley 2008 ; Credé and Kuncel 2008 ) . To foster this shift toward autonomy, numerous tools have been developed. Previous research has demonstrated the effectiveness of explicit instructional methods, such as structured workshops ( Arredondo and Rucinski 1994 ) , training courses ( Bail et al. 2008 ; Theobald 2021 ; Dignath and B"uttner 2008 ) , and classroom-based programs ( Paris and Paris 2003