Designing a new Smart, Adaptive and Embodied Learning Environment

The aim of this paper is to describe a newly design of a smart learning environment (SLE) (Spector 2012, 2014) where students can experience multi-modal and multi- sensory learning paths in order to learn in different related learning contexts in a more entertaining fashion within an inclusive edutainment (educational entertainment) paradigm. This new SLE exploiting deep technology integration can be originally identified as a "Smart Adaptive and Embodied Learning Environment". It wants to achieve these goals: o to establish a continuum of education by integrating formal settings to real-life informal learning situations; o to support specific students' needs by setting up personalized learning experiences; o to meet the students' longing to have educational contents conveyed through simpler and more winsome approaches; This new SLE, therefore, offers adaptive learning paths after a preliminary student profiling realized from correlation of information acquired by means of an original integration, across four learning contexts (museum, scientific laboratory, school, households), of existing and ad hoc developed technologies that support students' engagement, achievement and motivation (Chakraborty et al. 2010). Conceiving cognition as complex and embedded in body-world relations (Smith 2005), the adaptive learning paths in the SLE are developed taking into account both the outcomes from learning activities and their connection with linguistic, behavioural and physiological outcomes. Therefore, an innovative strategy of this new SLE is to elaborate multimodal outputs starting from multimodal inputs acquired in different learning contexts. Such a Smart, Adaptive and Embodied Learning Environment is managed by a smart intelligence of a cognitive architecture (Augello et al. 2013) that is able to manage interactions and tasks involving its two embodiments: robots and robot's avatar on mobile platforms as storytellers and catalyst of learning processes (Mubin et al. 2013; Tahriri et al. 2015). Moreover, the creation of natural language, behavioural, physiological and personalized learning modules is planned in order to equip this cognitive architecture of a complex knowledge that allows a physical and linguistic interaction between the robot/avatar and child (Byrge et al. 2014; Nacke et al. 2011). The cognitive architecture, thanks to its embodiments and thanks to its cognitive modules, creates adaptive learning paths (Garber-Barron e Si 2013) starting from relations and activities inside the four sub-SLEs (museum, laboratory, school, household). In particular, in this creation process, it endorses various technological integrations fitting the different learning contexts: i.e. the robot's avatar and physiological sensors in household; the robot and the robot's avatar at school; the robot, its avatar and sensors for several behavioural and physiological measurements in museum and laboratory (Benitti 2012). Furthermore in order to achieve its goals the design of the new SLE integrates two different approaches, the use of robots/virtual robots and use of visual elements connected to learning activities (artworks), both effective for the increase of cognitive skills not only for typically developed students but also for students with special needs, especially for students with autism (Losh e Gordon 2014). This SLE is able to take care of students with special needs implementing learning paths by means of digital interactive tools that will allow them (students) to partake in activities from which they could be potentially excluded. The whole of the design process takes into account the principles of embodied cognition in order to create a context within the learning process involves the core relation between body and world and allows a multi-modal and multi-sensory learning