Gabriel Rongyang Lau

Gabriel Rongyang Lau

Auto-generated profile, based on publicly available information

Gabriel Rongyang Lau

National University of Singapore, Singapore

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 May 2026 in Computers in Human Behavior Reports
Verified
Journal Article
Open access
GLGabriel R. Lau
Gabriel R. Lau
WLWei Yan Low
Wei Yan Low
LTLouis Tay
Louis Tay
YGYsabel A. Guevarra
Ysabel A. Guevarra
DGDragan Gašević
Dragan Gašević
Andree HartantoAndree Hartanto
Andree Hartanto

27 June 2026 in Computers in Human Behavior Reports
Verified
Journal Article
Open access
GLGabriel R. Lau
Gabriel R. Lau
KKK.T.A. Sandeeshwara Kasturiratna
K.T.A. Sandeeshwara Kasturiratna
AGAdalia Y.H. Goh
Adalia Y.H. Goh
ETEddie M.W. Tong
Eddie M.W. Tong
Andree HartantoAndree Hartanto
Andree Hartanto

2 June 2026 in PsyArXiv
Verified
Preprint
Open access
GLGabriel Rongyang Lau
Gabriel Rongyang Lau
CLCaresse Jia Min Liow
Caresse Jia Min Liow
DGDragan Gasevic
Dragan Gasevic
Andree HartantoAndree Hartanto
Andree Hartanto
Generative AI can support learning, but it may become counterproductive when learners use it to bypass regulatory processes needed for durable understanding. We conceptualise this pattern as maladaptive AI use for learning, a learning-specific regulatory failure within self-regulated learning comprising Uncritical Reliance, Shallow Processing, and Illusory Competence. Across six studies (N = 909), we advanced a regulatory-failure framework, developed and validated a 12-item measure, mapped its nomological network, and examined links with learning outcomes. Study 1 generated and content-validated the initial item pool. Study 2 identified a three-factor structure through EFA. Studies 3 and 4 confirmed this structure, with robust factor loadings and good internal consistency across validation studies (αs = .86–.90). Study 4 supported sex invariance, convergent and discriminant validity. In Study 4, maladaptive AI use for learning was associated with higher AI attachment, social influence surrounding AI adoption, positive attitudes toward AI, academic procrastination, controlled academic motivation, trait anxiety, and maladaptive coping, as well as lower critical thinking in AI use, open-mindedness, and conscientiousness. Study 5 showed good one-week test–retest reliability (ICC(2,1) = .77). Study 6 supported criterion validity, with greater maladaptive AI use for learning predicting lower GPA, poorer independent writing-task performance after AI-supported learning, and greater overconfidence in perceived understanding relative to actual performance. These findings clarify maladaptive AI use for learning as a distinct learning-specific construct and provide a foundation for future research and instructional interventions that help learners use AI to support rather than replace the regulatory processes required for durable learning.

6 April 2026 in PsyArXiv
Verified
Preprint
Open access
GLGabriel Rongyang Lau
Gabriel Rongyang Lau
CLCaresse Jia Min Liow
Caresse Jia Min Liow
DGDragan Gasevic
Dragan Gasevic
Andree HartantoAndree Hartanto
Andree Hartanto
As AI-powered tools become embedded in learning environments, understanding how learners regulate their interactions with these systems is essential for designing learner-sensitive, adaptive AI-supported instruction. Yet existing self-regulated learning measures do not capture the regulatory demands of learning with generative AI nor the individual differences in how learners navigate them. This research conceptualises AI learning strategies as the strategic ways learners use AI tools to support and regulate their learning, grounded in a self-regulated learning framework spanning forethought, performance, and self-reflection phases. Across five studies (N = 1001), we developed and validated the 12-item AI Learning Strategies Scale and mapped its nomological network. Study 1 generated and content-validated scale items. Study 2 identified a three-factor structure (Forethought, Performance, and Self-Reflection) via exploratory factor analysis. Studies 3 and 4 confirmed this three-factor model, demonstrated strong internal consistency (αs = .94–.95), robust factor loadings, sex invariance, and convergent and discriminant validity. Within the nomological network, greater AI learning strategy use was associated with higher self-efficacy, autonomous motivation, mastery-approach goals, conscientiousness, and extraversion, as well as more positive attitudes toward AI, more frequent and diverse AI use, and lower AI fatigue and neuroticism. Study 5 demonstrated one-week test–retest reliability (ICC(2,1) = .80) and criterion validity, with higher scores predicting richer self-regulated learning content in a writing task. The present work clarifies how learners differ in their strategic use of AI for learning, and provides a validated measure of individual differences to support adaptive, theory-driven, and cross-group research in technology-enhanced learning environments.

25 September 2026 in PsyArXiv
Verified
Preprint
Open access
MHMeilan Hu
Meilan Hu
GLGabriel Rongyang Lau
Gabriel Rongyang Lau
AGAdalia Y. H. Goh
Adalia Y. H. Goh
SCSamuel Rhys Cox
Samuel Rhys Cox
LTLouis Tay
Louis Tay
Andree HartantoAndree Hartanto
Andree Hartanto
Large language model (LLM)-based chatbots offer new methodological opportunities for psychological research by enabling scalable, standardised, and interactive research procedures. However, their usefulness depends on the researchers’ ability to customise, evaluate, and deploy them in ways that align with the methodological requirements of a given study. The current tutorial provides a practical, step-by-step framework for developing customised LLM-based chatbots for psychological research. We distinguish between no-code or low-code development approaches, including provider-hosted chatbot builders and visual or low-code programming platforms, and code-based approaches using proprietary model application programming interfaces (APIs) or locally hosted open-weight models. We compare these approaches in terms of technical expertise, behavioural and workflow control, study integration, cost, data governance, maintenance, and reproducibility. We also outline key components of the process of customising an LLM-based chatbot and illustrate deployment workflows across multiple platforms, including local desktop interfaces, messaging applications and survey platforms. Finally, we discuss key considerations for research ethics, privacy, data governance, security, transparency, and reproducibility, including recommendations for documenting and sharing chatbot configurations and study procedures. Through this tutorial, we aim to support researchers in integrating customised LLM-based chatbots into psychological research and in expanding the methodological possibilities available to them.

23 October 2026 in Computers in Human Behavior: Artificial Humans
Verified
Journal Article
Open access
MRMary K.G.R. Reyes
Mary K.G.R. Reyes
GLGabriel R. Lau
Gabriel R. Lau
MTMing-Hong Tsai
Ming-Hong Tsai
KKK.T.A. Sandeeshwara Kasturiratna
K.T.A. Sandeeshwara Kasturiratna
Andree HartantoAndree Hartanto
Andree Hartanto