Abstract: This thesis examines the influence of design characteristics in multimedia learning environments on learners' engagement, motivation, attention, and learning performance. Building upon Mayer's cognitive theory of multimedia learning, it proposes an integrative model articulating the concepts of engagement and attention, and tests it through three empirical studies.The first study investigates the impact of anthropomorphism in e-learning content about immunotherapy. It reveals that a design rich in humanized visual elements improves learners' engagement, identified motivation, and memorization performance, without increasing cognitive load. The second study examines the effects of the presence and degree of expressiveness of an animated pedagogical agent. It shows differentiated effects on comprehension and memorization, modulated by learners' individual attentional capacities. The third study compares a classical pedagogical agent to an interactive agent based on artificial intelligence. While the interactive agent promotes greater engagement, it does not significantly improve learning performance.These results enrich existing models by emphasizing the importance of considering engagement as a specific attentional and experiential state, characterized notably by an altered perception of time. They highlight the key role of individual attentional capacities in regulating engagement and managing cognitive resources in response to multimedia design choices.Methodologically, the thesis demonstrates the value of a multidimensional approach to assessing engagement and attention, combining objective and subjective measures. It underlines the relevance of indicators such as subjective time perception to capture states of deep engagement.This work opens up perspectives for designing more engaging digital learning environments adapted to learners' needs. It advocates for the development of "attentive" interfaces, capable of identifying engagement dynamics in real-time and adapting pedagogical guidance accordingly. More ...
No Comments.