A Systematic Review of Artificial Intelligence-Based Computer Adaptive Testing (CAT) and Item Response Theory for Enhancing the Effectiveness of Science Learning Assessment

Muhammad Gibran Alif  Prasetya; Arif  Widiyatmoko; Ani  Rusilowati

doi:10.54783/ijsoc.v7i4.1581

Muhammad Gibran Alif Prasetya Universitas Negeri Semarang, Semarang, Indonesia
Arif Widiyatmoko Universitas Negeri Semarang, Semarang, Indonesia
Ani Rusilowati Universitas Negeri Semarang, Semarang, Indonesia

DOI: https://doi.org/10.54783/ijsoc.v7i4.1581

Keywords: Computer Adaptive Testing, Item Response Theory, Artificial Intelligence, Fuzzy Logic, Science Learning.

Abstract

The advancement of technology has accelerated the adoption of Computerized Adaptive Testing (CAT) in educational assessment due to its ability to dynamically adjust item difficulty levels, thereby producing more precise, efficient, and valid measurements compared to conventional tests. While Item Response Theory (IRT) serves as the primary psychometric foundation of CAT, traditional IRT implementation faces computational challenges because ability estimation requires lengthy iterative processes, resulting in reduced system responsiveness. To address this issue, Artificial Intelligence (AI), particularly Fuzzy Logic, offers a promising solution through rapid inference mechanisms and monotonic reasoning that can adaptively map students’ cognitive abilities to corresponding item difficulty levels. This study aims to develop a hybrid CAT system that integrates Fuzzy Logic for fast inference with IRT as a robust and valid psychometric framework in the context of science learning. The research employs a systematic literature review using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework, encompassing the stages of Identification, Screening, and Inclusion of relevant studies. The findings indicate that the integration of AI/ML with IRT in CAT consistently enhances assessment accuracy and efficiency. Algorithms such as Maximum Information (MI) and Expected a Posteriori (EAP) effectively reduce test length without compromising reliability, while Fast Adaptive Cognitive Diagnosis (FACD) improves early-stage ability prediction. Furthermore, Fuzzy Logic demonstrates strong effectiveness in selecting adaptive test items aligned with students’ ability levels. The study concludes that developing CAT systems based on AI and IRT yields adaptive, personalized, efficient, and diagnostic evaluation mechanisms that support personalized science learning.

References

Albalushi, N., & Awad, W. S. (2025). Generating Questions Bank for Adaptive Assessment Using Machine Learning Techniques: Review.

Cheng, S.-C., Cheng, Y.-P., & Huang, Y.-M. (2021). To implement computerized adaptive testing by automatically adjusting item difficulty index on adaptive English learning platform. Journal of Internet Technology, 22(7), 1599–1607.

El Msayer, M., Aoula, E.-S., & Bouihi, B. (2024). Artificial intelligence in computerized adaptive testing to assess the cognitive performance of students: A Systematic Review.

Frick, S., Krivosija, A., & Munteanu, A. (2024). Scalable learning of item response theory models. International Conference on Artificial Intelligence and Statistics, 1234–1242.

Ghosh, A., & Lan, A. (2021). BOBCAT: Bilevel Optimization-Based Computerized Adaptive Testing. IJCAI International Joint Conference on Artificial Intelligence, 2410–2417.

Göktepe Körpeoğlu, S., Filiz, A., & Göktepe Yıldız, S. (2025). AI-driven predictions of mathematical problem-solving beliefs: Fuzzy logic, adaptive neuro-fuzzy inference systems, and artificial neural networks. Applied Sciences, 15(2), 494.

Harrison, C. J., & Trickett, R. W. (2025). Patient reported outcome measures: from the classics to AI. Journal of Hand Surgery: European Volume, 50(6), 807–813.

Haryanto, H. (2011). Pengembangan Computerized Adaptive Testing (CAT) dengan Algoritma Logika Fuzzy. Jurnal Penelitian Dan Evaluasi Pendidikan, 15(1).

Huda, A., Irfan, D., Hendriyani, Y., & Sukmawati, M. (2024). Optimizing Educational Assessment : The Practicality of Computer Adaptive Testing (CAT) with an Item Response Theory (IRT) Approach. International Journal on Informatics Visualization, 8(1), 473–480.

Huda, C. (2024). Paradigma Pembelajaran IPA Berbasis Proyek Berdiferensiasi: Menyukseskan Kurikulum Merdeka Belajar Kampus Merdeka. Penerbit NEM.

Imawan, O. R., Retnawati, H., Haryanto, H., & Ismail, R. (2025). Innovations in Assessment Methods: Computerized Adaptive Testing (CAT) for Sustainable Energy Efficiency. Lecture Notes in Civil Engineering, 557 LNCE, 161–168.

Imawan, O. R., Retnawati, H., & Ismail, R. (2025). The Challenges of Implementing Computerized Adaptive Testing in Indonesia. Journal of Education and E-Learning Research, 12(2), 124–144.

Iwintolu, R. O., Opesemowo, O. A. G., & Adetutu, P. O. (2024). Effect of 2-PL and 3-PL Models on the Ability Estimate in Mathematics Binary Items. Journal on Efficiency and Responsibility in Education and Science, 17(3), 257–272.

Kane, L. T., Abboud, J. A., Plummer, O. R., & Beredjiklian, P. T. (2021). Improving Efficiency of Patient-Reported Outcome Collection: Application of Computerized Adaptive Testing to DASH and QuickDASH Outcome Scores. Journal of Hand Surgery, 46(4), 278–286.

Kishida, W., Fuchimoto, K., Miyazawa, Y., & Ueno, M. (2023). Item Difficulty Constrained Uniform Adaptive Testing. Communications in Computer and Information Science, 1831 CCIS, 568–573.

Klein, B., & Kovács, K. (2024). The performance of ChatGPT and Bing on a computerized adaptive test of verbal intelligence. PLOS ONE, 19(7 July).

Kwong, H. Y., & Mohammadi, G. (2025). Recommender Methods for Computerised Adaptive Testing. 13–15.

Liu, Y., You, Y., Liu, S., Qian, H., Qian, Y., & Zhou, A. (2025). A Fast-Adaptive Cognitive Diagnosis Framework for Computerized Adaptive Testing Systems. IJCAI International Joint Conference on Artificial Intelligence, 5824–5832.

Ma, W. A., Richie-Halford, A., Burkhardt, A. K., Kanopka, K., Chou, C., Domingue, B. W., & Yeatman, J. D. (2025). ROAR-CAT: Rapid Online Assessment of Reading ability with Computerized Adaptive Testing. Behavior Research Methods, 57(1), 56.

Maji, S., & Ganguli, S. (2025). Fuzzy Logic Control for Industrial Applications. Controller Design for Industrial Applications, 1–20.

Papadimitriou, S., & Virvou, M. (2025). Fuzzy Logic and Applications in Education and Games: Theory, Practical Implementations and a Literature Review. Artificial Intelligence-Based Games as Novel Holistic Educational Environments to Teach 21st Century Skills, 95–127.

Sathya, D., Saravanan, G., & Thangamani, R. (2024). Fuzzy logic and its applications in mechatronic control systems. Computational Intelligent Techniques in Mechatronics, 211–241.

Suzuki, A., & Negishi, E. (2024). Fuzzy logic systems for healthcare applications. Journal of Biomedical and Sustainable Healthcare Applications, 4(1).

Tian, X., & Dai, B. (2020). Developing a computerized adaptive test to assess stress in Chinese college students. Frontiers in Psychology, 11, 7.

Tsaousis, I., Sideridis, G. D., & AlGhamdi, H. M. (2021). Evaluating a computerized adaptive testing version of a cognitive ability test using a simulation study. Journal of Psychoeducational Assessment, 39(8), 954–968.

Wanniarachchi, V. U., Greenhalgh, C., Choi, A., & Warren, J. R. (2025). Personalization variables in digital mental health interventions for depression and anxiety in adolescents and youth: a scoping review. Frontiers in Digital Health, 7.

Wulansari, A. D., & Kirana, D. P. (2023). Pengukuran English Vocabulary Size dengan Computerized Adaptive Testing. Thalibul Ilmi Publishing & Education.

Yang, H., Li, Z., & Pedrycz, W. (2025). Adaptive Deep Learning from Crowds. IJCAI International Joint Conference on Artificial Intelligence, 4263–4272.

Zhu, H. (2022). Research on intelligent analysis strategies to improve athletes' psychological experience in the era of artificial intelligence. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 119, 110597.