Language learning is central to the social inclusion pathways of adult migrants. In particular, the composition of the foreign population in Italy and its educational needs introduce important challenges for language education, such as adult literacy and education, which go hand in hand with the learning of communicative competence. In this context, a conversational agent, on mobile devices, can play a significant role of integration between formal and informal learning, leveraging interactivity and flexibility of use. In the contribution, we will illustrate how methodological approaches tailored to the needs of low literate adults have been translated into the functional specifications of a chatbot (CPIAbot) supporting classroom and out-of-class teaching, within the CPIA[1]. The interweaving of specific needs, didactic-methodological approaches and bot architecture will be analysed, highlighting its strengths and critical points in view of a forthcoming experimentation aimed at verifying the effectiveness of CPIAbot in supporting poorly educated migrants in learning Italian L2. [1] http://www.retecpialiguria.it/

Every day life tasks can present significant challenges for neurodivergent individuals, particularly those with Autism Spectrum Disorders (ASD) who are characterized by specific sensitivities. This contribution describes a virtual reality system that allows neurodivergent individuals to experience everyday situations in order to practice and implement strategies for overcoming their daily challenges. The key strength of the proposed system is the automatic personalization of the virtual environment, based on both the individual's abilities and their specific training needs. The proposed method has been evaluated on four synthetic user profiles, also proposing a metric able to evaluate the variance of the features within the same difficulty level. The results show that the method can produce a significant number of scenarios for the various difficulty levels. Furthermore, within the same difficulty, there is a wide variance of the non-constrained features for the specific profile.