<![CDATA[The low mathematical logical thinking ability of Indonesian students is reflected in Indonesia's 2022 PISA score of 366, which falls significantly below the OECD average of 472, indicating weaknesses in students' capacity to draw inferences, recognize patterns, construct arguments, and derive systematic conclusions. This study aims to synthesize empirical evidence regarding the effectiveness of Artificial Intelligence (AI) on both competencies, while identifying the types of technology, implementation designs, and moderating factors that influence its impact. A Systematic Literature Review (SLR) was conducted following the PRISMA 2020 guidelines, with searches performed on Google Scholar using PICOS-based keywords structured with Boolean AND/OR operators, restricted to publications from 2024 to 2026. Study selection was conducted independently by two researchers, with inter-rater reliability maintained through Cohen's Kappa. Of 538 initially identified articles, 15 final articles from eight countries were retained following deduplication, screening, and quality assessment using MMAT, RoB 2.0, and NOS. AI technologies were classified into five categories: generative AI, Intelligent Tutoring Systems, adaptive learning systems, educational chatbots, and AI-based Learning Management Systems. Results indicate that AI consistently exerts a positive influence on self-regulated learning through adaptive feedback and metacognitive support, and on mathematical logical thinking ability through its role as a cognitive partner. However, its effectiveness is conditional, necessitating that mathematics teachers integrate AI within structured instructional designs incorporating formative feedback, graduated scaffolding, and metacognitive reflection facilitation.]]>
Copyrights © 2026
