CARLOS VOLTER BUENAÑO PESÁNTEZ , KELLY DEYSI HERNÁNDEZ MITE , ROSSY ELVIRA MACÍAS ACOSTA , ANGÉLICA DEL ROCIO TOMALÁ DE LA CRUZ, LUIS MARQUÉS MOLIAS
DOI: https://doi.org/The integration of Artificial Intelligence (AI) in neuroeducation is transforming the understanding of the affective and cognitive processes involved in scientific learning. This systematic review analyzes research published between 2020 and 2025 in Scopus, Web of Science, IEEE Xplore, ACM Digital Library, ERIC, and PsycINFO, in order to examine how AI is used to recognize, support, or intervene in emotion, motivation, and attention during science teaching. The study followed the PRISMA 2020 guidelines and included 31 empirical and review investigations. The findings were grouped into three thematic axes: (1) affective computing and emotional regulation, (2) intelligent tutor systems and adaptive gamification as drivers of intrinsic motivation, and (3) attentional analytics and monitoring of cognitive engagement. Evidence indicates that AI enables personalized, emotionally sensitive, and self-regulated learning experiences, provided that ethical frameworks of transparency, fairness, and data protection are in place. However, challenges associated with algorithmic bias, reductionist interpretation of affective states, and scarce contextualized production in Latin America remain. It is concluded that the pedagogically grounded integration of AI can strengthen science teaching, in particular through culturally sensitive models that articulate emotion, motivation, and attention from a neuroeducational perspective.
