Cognitive neuroscience robotics will be employed to develop and establish new diagnostic and therapeutic sensorimotor approaches for NDDs. Converging evidence suggest that NDD patients benefit from early diagnosis and early comprehensive behavioural intervention tailored to their specific needs. Yet, diagnosis of NDDs remains complex and often difficult to obtain. It currently depends on interpretative coding of patient observations, parent interviews, and manual testing by medical experts. Recent quantification of sensorimotor parameters in ASD and other NDDs provides a new target for the development of early assessment and intervention methodologies. When behaviour is tracked observationally, the continuous stream of movement and variable degrees of intent that are inherently present in natural behaviours are lost. However, new technology can capture with high precision the movements that our eyes miss, from facial micro-expressions to rapid and frequent eye motions, to atypicalities in kinematic patterns. In combination with quantitative imagining techniques (e.g., functional magnetic resonance imaging), quantification of movement parameters can bring to the identification of objectively defined bio-behavioural markers more effective in early assessment of NDDs than traditional methods. Additionally, is can also be used for improved therapeutic interventions using robots capable of ‘modularity of control’. Modularity of control means that it is possible to decompose robot movements into elements, thus systematically varying the degree of similarity between the robot and the child’s behaviour. This offers a valuable entry point for intervention to model social skills and reconfigure sensorimotor integration.