Mutual modelling ability for a humanoid robot how can it improve my learning as we solve a problem together?

Robot-mediated learning activities are often designed as collaborative exercises where two or more children work together to achieve the activity objectives. Witnessing such activities allows for making an interesting observation: although miscommunications and misunderstandings occur frequently, humans, even at young ages, are very good at understanding each other. Humans, unlike robots, are highly skilled in detecting and addressing misunderstandings. We represent whether the others understood what we said or did by using the complex cognitive ability of mutual modelling1, i.e. the reciprocal ability to construct a mental representation of the other, by attributing beliefs, desires and other mental states to the other [10]. This ability is critical in order for humans to comprehend each other and react appropriately in their interactions. Thus, the main goal of my PhD is to equip a robot with mutual modelling ability, and use this ability in an educational activity in order to improve the quality of the interactions between the robot and a learner and (hopefully) the learning outcomes.