Personalization is often perceived as a technical problem in the context of digital education. However, it is also a cognitive challenge, requiring an understanding of how learners process information. This study presents a cognitive-based recommendation model designed and tested within the Master's program in E-learning and Intelligent Educational Systems (ELSEI) at the École Normale Supérieure (ENS) of Abdelmalek Essaadi University in Morocco. This model is based on the principles of cognitive load theory and uses artificial intelligence to adapt learning experiences in real time. The model aims to align content presentation with each learner's cognitive profile. This alignment is achieved through clustering, load classification, and hybrid recommendation techniques. This model is tested for six weeks involving 47 graduate students and has shown encouraging results: an improvement in engagement and academic performance, as well as a reduction in learning disruptions related to overload. These results suggest a broader evolution of educational AI, towards systems that are not only adaptive, but also attentive to the cognitive demands of learning. [For the complete proceedings, "Proceedings of the International Association for Development of the Information Society (IADIS) International Conference on Cognition and Exploratory Learning in the Digital Age (CELDA) (22nd, Porto, Portugal, November 1-3, 2025)," see ED677812.]