El uso de ChatGPT para personalizar el aprendizaje en estudiantes universitarios: una Revisión Sistemática de la Literatura con el modelo PRISMA

Karla Karina Ruiz Mendoza; Rubi Surema Peniche Cetzal

doi:10.54674/ess.v37i2.1121

The use of ChatGPT to personalize learning for university students: A PRISMA-guided systematic literature review.

Karla Karina Ruiz Mendoza Autonomous University of Baja California, Tijuana and Ensenada, Mexico https://orcid.org/0000-0001-8978-8364
Rubi Surema Peniche Cetzal Autonomous University of Baja California, Tijuana and Ensenada, Mexico https://orcid.org/0000-0003-0105-6471

DOI: https://doi.org/10.54674/ess.v37i2.1121

Keywords: Artificial intelligence, Educational technology, Higher education, University students, Individualized instruction, Artificial Intelligence; Educational Technology; Higher Education; University Student; Individualized Instruction

Abstract

This PRISMA 2020-guided systematic review examines how ChatGPT is used to personalize learning in higher education (2022–2025) and what conditions enable its adoption. Searches across Web of Science, Scopus, and ERIC were complemented with backward and forward citation chasing. We included 38 peer-reviewed empirical studies featuring explicit ChatGPT use and evidence of personalization. Publication output has accelerated since 2023 and spans multiple disciplines. Findings indicate that the most consistent pathway is individualized, timely feedback and data-triggered support, with reported gains in achievement, comprehension, self-regulation, and engagement. Partial modalities—tone/persona adjustment, learning itineraries, and personalized explanations—further refine the experience, albeit with heterogeneous evidence. Technical transparency remains limited, as model versions and key parameters are seldom reported, constraining replicability and cross-study comparison. Governance-aware personalization is therefore recommended: human oversight, transparency and integrity policies, GAI literacy, and equitable infrastructure. Instructionally, prioritize feedback-first designs, goal-linked sequences and pacing, higher-order thinking activation, and technical traceability. Future research should employ longitudinal and experimental designs and standardized reporting of model configurations and versions.

Author Biographies

Karla Karina Ruiz Mendoza, Autonomous University of Baja California, Tijuana and Ensenada, Mexico

She holds a Ph.D. in Educational Sciences from the Institute for Educational Research and Development (IIDE) at the Autonomous University of Baja California, and she earned a Bachelor’s degree in Language and Literature Education and a Master’s degree in History from the same university. She also holds a Bachelor’s degree in Computer Systems, a Master’s degree in Digital Education, E-learning, and Social Media, as well as certificates and courses in Technology, Data Analysis, Programming, and Machine Learning. She is currently pursuing a Master’s degree in Artificial Intelligence. Her interests focus on topics related to education, technology, and their intersections with the humanities and social sciences.

Rubi Surema Peniche Cetzal, Autonomous University of Baja California, Tijuana and Ensenada, Mexico

Bachelor’s degree in Education and Master’s degree in Educational Research from the School of Education at the Autonomous University of Yucatán. Ph.D. in Educational Research for Curriculum Development and School Organizations from the University of Granada, Spain. Teaching experience since 2006 at the high school, undergraduate, and graduate levels. Member of the SNII since 2015. Her research interests include school effectiveness, assessment, high school, and the transition to higher education. She is a full member of the Mexican Council for Educational Research (COMIE), the Ibero-American Network for Leadership and Educational Practices (RILPE), and the Women United for Education Network (MUxED).

HTML (Español (España))

Published

2026-05-30

How to Cite

Ruiz Mendoza, K. K., & Peniche Cetzal, R. S. (2026). The use of ChatGPT to personalize learning for university students: A PRISMA-guided systematic literature review . Higher Education and Society Journal (ESS), 37(2), 165-184. https://doi.org/10.54674/ess.v37i2.1121

Download Citation

Issue

Vol 37 No 2 (2025): Artificial Generative Intelligence (GenAI) in teaching and research: best practices, challenges, and prospects

Section

Articles Thematic Dossier

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Copyright notice

Copyright allows the protection of original material, and curbs the use of others' work without permission. UNESCO IESALC adheres to Creative Commons licenses in the open access publication of ESS. Specifically, texts published in this journal are subject to a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license: ESS is an open access journal, which means that all content is freely available to the user or their institution. Users may read, download, copy, distribute, print, search or link to the full text of the articles, or use them for any other lawful purpose, without asking prior permission from the publisher or the author, always making sure to cite the author. Commercial use is not permitted. ESS requires authors to accept the Copyright Notice as part of the submission process. Authors retain all rights.

The full license can be found at https://creativecommons.org/licenses/by-nc/4.0/

Attribution - NonCommercial (CC BY-NC 4.0)

This journal does not charge authors for the submission or processing of articles. The authors of the contributions will receive acknowledgment of receipt that the work has reached the Editorial Team of the Journal.