Adaptive robotic training strategies can be used to cognitively and physically assist or challenge patients based on patient's needs (e.g., long-lasting cognitive impairments post-stroke) to enhance motor neurorehabilitation. Yet, the reality is that robotic neurorehabilitation systems are failing to outperform conventional therapy. Why?

Little research effort has been invested in investigating the neural effects of keeping a certain level of cognitive or physical challenge during motor training. In this project, we employ electroencephalography (EEG) to investigate the effect of training strategies that support cognitive functions (e.g., instructions) or physical limitations (e.g., robotic assistance) on motor learning and brain neural networks, in order to shed light on the neural correlates of motor learning.

Little is known about the effect of task instructions on participant’s attentional networks. However, our preliminary results on 36 healthy participants show that training with verbal instructions or visual cues (instead of training without instructions) results in improved motor performance of a virtual sailing motor task. Also, we observed that the task instructions modulate the attentional status during motor learning: visual cues support participants’ visual attention and verbal cues gear verbal analytical networks to succeed in the task.

The next step is to investigate whether robotic interventions (e.g., physically assisting or challenging patients) may negatively interact with neural processes responsible for motor planning and execution, limiting motor neurorehabilitation outcomes.

Our vision is that adaptive robotic training strategies that account for patients’ neural activity over areas involved in motor or cognitive aspects of motor learning, could be used to adapt the level of physical and cognitive assistance/challenge based on patient's cognitive and motor deficits to support motor learning.