Children’s Engagement during Cognitive Assessments with Minimal Robot Intervention

Standard pediatric cognitive and projective assessments can be monotonous and provoke anxiety. This reduces children's engagement and compliance and diminishes the quality of observable behavior. In this work, we present a workflow that transfers routine paper-and-pencil tasks to a tablet application and adds a NAO humanoid robot as a social partner. Our goal is to provide a robot-and-tablet system for standardized assessments. In this exploratory study, we focused on quantifying the added contribution of the robot to engagement compared to a tablet-only system. To this end, we evaluated the workflow in a pediatric clinic with 24 participants, each with a distinct neurodevelopmental disorder diagnosis, using a controlled, between-subjects design. The tablet-only condition delivered the same activities and pacing without robot cues. This served as a control to isolate the effect of social scaffolding. Licensed therapists rated engagement using the Pediatric Assessment of Rehabilitation Engagement (PARE) scale. Our preliminary findings suggest that the robot's presence supported higher observable engagement and sustained participation without altering the rapport within the child–therapist–robot triad. Moreover, results in the PARE scale were significantly different between the two conditions when considering the age of the little patients (with mean age 10 vs. 8, respectively). While the small sample size and age distribution warrant cautious interpretation, this work highlights a robot's potential contribution to children's engagement during standardized assessment.