A story for Pepper and Nao: applying the learning by teaching paradigm in children with patterns of neuromotor impairment

BackgroundSocial robots are increasingly used in educational and clinical contexts to enhance engagement and learning, particularly in children with neurological and neurodevelopmental disorders. The Learning by Teaching paradigm, in which children actively teach social robots, promotes cognitive and metacognitive processes such as planning, self-efficacy, and executive functioning; however, controlled and systematic evidence supporting its effectiveness remains limited. Children with neuromotor disorders frequently present with closely intertwined motor and cognitive planning difficulties, which significantly impact learning and everyday functioning. In this context, social robot–mediated interventions based on the Learning by Teaching paradigm may represent a promising approach. The current multicenter randomized controlled trial (RCT) explores the feasibility and effectiveness of this approach to enhance cognitive and motor planning in children with neuromotor impairments, and to examine potential generalization effects on other cognitive domains and engagement.MethodsThe main aim of this RCT study is to evaluate the feasibility and effectiveness of a social robot-mediated Learning by Teaching intervention in children aged 7–16 years with neuromotor impairments. Eighty participants will be randomly assigned to an Experimental Group (EG) receiving social robot training or to a Control Group (CG) continuing standard care. The intervention consists of 20–25 sessions in which children teach a humanoid social robot (Pepper or NAO) to order picture-based stories of increasing complexity. Assessments will be performed at baseline (T0), post-intervention or control period (T1), and follow-up/post-treatment (T2). Primary outcomes include planning abilities assessed through picture arrangement and an ad hoc motor planning task. Secondary outcomes encompass executive functions, visuo-spatial memory, attention, inhibition, narrative production, visual-perceptual skills, and reading comprehension. Engagement will be measured using a multimodal approach integrating behavioral, cognitive, emotional, and physiological indicators derived from video, audio, and wearable sensors.ResultsThe study will examine improvements in planning and motor programming abilities and potential generalization effects on executive functions and reading comprehension. Additionally, the predictive and mediating role of engagement in intervention effectiveness will be reported.ConclusionsThis research protocol proposes an innovative, engagement-focused rehabilitation approach integrating cognitive, motor, and narrative processes through social robotics. Findings are expected to inform the development of evidence-based, personalized interventions for children with neuromotor disorders and to advance methodological approaches for assessing engagement in human–robot interaction.Clinical Trial RegistrationClinicalTrials.gov, identifier NCT06797050.