International Journal of Computer Applications (0975 – 8887) Volume 168 – No.6, June 2017 46 Animated Pedagogical Agents to Assist Learners and to keep them motivated on Online Learning Environments (LMS or MOOC) K. Bendou IRF-SIC, Laboratory, IbnZohr B.P.37/S – Agadir – Morocco El. Megder IRF-SIC, Laboratory, IbnZohr B.P.37/S – Agadir – Morocco C. Cherkaoui IRF-SIC, Laboratory, IbnZohr B.P.37/S – Agadir – Morocco ABSTRACT This paper gives an overview of the first specifications of current research on an animated pedagogical agent to assist learners and to keep them motivate on an online learning environment (LMS or MOOC). It combines characteristics of intelligent agents like: autonomy, ability to perceive, to interact, to reason and to act; and some other characteristics of pedagogical agents as: observing, evaluating, adapting content, recommending, engaging, motivating, etc. The design of this agent is based on a new concept which we have called the Pedagogical Intervention. An intervention may be of different kinds, but it is more precisely used to overcome the current problem of abandonment of learners. We therefore propose to show, through this paper which is a summary of our recent work, the interest and importance of the analysis of the limitations of the online learning environments, in particular the causes of the drop-out problem in order to define adapted pedagogical interventions strategies. Keywords Online Learning; Environment; Adaptivity; Recommendation; Feedback; Pedagogical Intervention; Pedagogical agent. 1. INTRODUCTION Over the last decade, the interest in Online Learning Environments (OLE) is growing exponentially. In some ways, this is not a new phenomenon. Researchers working in the online learning sector have been constructing and designing a large amount of models and frameworks which aim to provide teaching or learning to individuals who are distant, but also to those with individual constraints to access to knowledge (spatial, temporal, technological, psychosocial and socio- economic constraints). As a result, several online learning systems have been developed. Among them we can mention: Adaptive Hypermedia Systems (AHS), Learning Management Systems (LMSs), Virtuel Learning Environments (VLE), Knowledge Management Systems (KMS) and very recently MOOCs [1, 2]. We can also highlight the different ways of learning in informal settings using social networks for learning, discussion forums, blogs, Wikis and access to free resources, etc. In principle, such environments include course content delivery tools, synchronous and asynchronous activities, exercises and quiz modules, projects, games, workspaces for sharing resources. They also incorporate different types of multimedia learning resources: Videos, Webinars, Podcasts, Apps, etc; but also various nomadic means of access to information: tablets, Smartphone, etc. [3]. These environments, both rich with open varied learning content as well as technologies for interacting and collaborating about this content; offers today new opportunities to learn to each connected resident of the planet [4]. They without hesitation, revolutionize learning. However, perhaps unsurprisingly, the pervading discourses around the possible contributions of such environments on effective learning point out a number of questions, particularly the issue of dropout. These are imperative and urgent issues that the OLE community cannot avoid in order to stabilize existing models. If the issue is the construction of OLE that can contribute to a transformation of traditional relationships to knowledge, the learning and teaching required must be achieved under constraints acceptable to the main actors of learning [4]. For the learner, for example, what is sought is a tool that adapts delivery in different dimensions and levels, namely the adaptation of content, presentation, navigation; but also through individual dimensions and/or collaborative, the fun and the massive one [3, 4], The learner is also looking for tools that allow for a great openness (even massive and therefore very social), and that takes into account their different daily practices. Today’s learners also prefer instruction highly connected to their occupations and interests. In this sense, learners from different part of the world can collaborate or interact with global peers and mentors. In other hand, educators in online learning feel desperately an urgent need for automatic ways to motivate and retain online learners. This perspective can be fully integrated into current research on the issue of dropout, motivation and recent solutions implemented through the new concept of adaptive learning. In the different types of online learning environments, the drop- out issue remains the most researched problem over the past ten years. In the particular case of MOOCs, for example, several studies show a large dropout rate estimated at 90% by most authors of the field [3, 9]. The causes of abandonment are related in particular to the lack of motivation and commitment of the learners. Other reasons may be due to occupation, lack of time, isolation, etc. We believe that scripting and non- adaptation of content is a secondary cause. This is also due, in our view to difficulties to keep in OLE a sufficient number of tutors, because of their massive dimension. The current challenge in OLE is to keep learners motivated [4]. Some authors show that this is possible simply by adding some encouragement phrases above the statement of Mathematics exercises such as "Remember, the more you practice, the more you become intelligent, "or" This could be a difficult problem, but we know that you can get there (do it) ", etc. Learner