Abstract:

Autism spectrum disorder (ASD) is connected to two main deficits (DSM-V, 2013), that is, communication and social-interaction and, the one related to the presence of repetitive, stereotyped and restricted behavioral patterns. Within the first category we can refer to deficit of Theory of Mind (ToM) as the human ability to mindread , that is, deduct and take into account the others’ thoughts, desires, feelings and emotions (Etel, Slaughter, 2019). In ASD we can have different levels of ToM deficiencies (Matthews, Goldberg, 2018), but this is generally due to a lack of comprehension, at cognitive level, of the others’ behaviours as “inner states” (Baron Cohen, 1997). Interventions to support people with ASD in the comprehension of social situations can imply the use of social stories and of a set of prompts that the educator (or any training/research staff) provides in order to strenghten the chance to receive a proper reply/reaction. The prompts should be gradually reduced to help the person with ASD gain autonomy. Social stories are normally read by human interlocutors, but, since the typical difficulties of people with ASD is to keep the attention during the interaction some studies have demonstrated positive results when the story is told by a social robot (Vanderborght et. al., 2012). In fact the literature in the field underlines how the robotic device lets the educator open an alternative communication channel and direct the person with ASD’s attention and eye contact and finally activate new social behaviours (Scassellati et.al., 2018).

The University of Macerata (Italy) with the collaboration of the Roboticss lab of the University of Milan Bicocca (Italy) has developed a research design, consistent with the literature in the field of robotics and ASD, with the objective to improve the social skills connected to the ability of mentalizing of children with ASD. The research, based on a nonconcurrent multiple baseline method, is structured into two steps, in the first intervention the researchers use the humanoid robot NAO and in the latter step the child is introduced to an immersive 3D virtual world to interact with an avatar robot in order to generalize the behaviours trained with NAO in the physical setting (Pennazio, Fedeli, 2019). Both systems are connected to the use of social stories. The experimentation is anticipated by a preparatory session in which the robot NAO is used to help the researcher observe and identify the specific need of the child and analyse the setting of a system of prompts to be provided to the child to be, then, gradually avoided. The present contribution is the description of the session and of the collected results (pilot study).

In order to have a more detailed overview of the child’s ability of comprehension of the emotional states the Test of Emotion Comprehension (TEC) was used. The analysis of the ability of mentalizing was integrated with the one of collateral behaviours (e.g. presence of self-stimulatory behaviors, etc.). Such analysis was developed thanks to the observation of the interaction between the child and the robot in different direction (the child as a spectator and as active interlocutor). A grid was, moreover, structured in order to analyse the typology of prompts as dependent variables (Vandeborght et.al., 2012) offered to the child to activate a proper and successful social interaction based on the target ability that showed a deficit in the previous analysis.

Keywords:

Autism spectrum disorder, robotics, pilot, social abilities.