Investigation of the processing of proprioception and vision for motor actions at the behavioural and neural levels
Moving in a complex and dynamic environment requires continuous processing of different sources of information. Optimal feedback control (OFC) is a theory of motor control that predicts how information updated by sensory feedback should be integrated into motor commands for voluntary actions. This theory emphasizes that sensory feedback is critical for motor actions and should be flexibly used in accordance with the behavioural goal. In this thesis we tested several key predictions from OFC about how sensory feedback should be integrated. Our first study (Chapter 2) examined how corrective actions to visual feedback of the limb (cursor) were impacted by properties of goal and the environment. Our results indicate that the motor system processes visual feedback using two processes, an early process starting 90ms after a disturbance reflecting spatial redundancies of the goal, and a later process starting 120ms after a disturbance reflecting environmental obstacles. Our second study (Chapter 3) examined if goal shape influences how monkeys correct for sensory errors. We found goal shape influenced corrective responses to mechanical and visual disturbances of the limb starting 70ms and 90ms after the disturbance, respectively. Our third study (Chapter 4) examined the organization of different sources of sensory feedback to primary motor cortex (M1), an area involved with performing feedback control. Vision typically provides the only source of sensory input about behavioural goals whereas proprioception and vision provide sensory inputs about the limb. As predicted from OFC we found limb and goal feedback targeted virtually the same neurons in M1 and generated similar patterns of activity. Our fourth study (Chapter 5) tried to unify two competing views about M1. OFC views M1 as a feedback controller relying heavily on sensory feedback for control. In contrast, a recent theory suggests M1 behaves like an autonomous dynamical system with little influence from sensory input based on observations of rotational dynamics in M1. We show that rotational dynamics are present in feedback control networks indicating rotational dynamics are not unique to autonomous dynamical systems. Collectively, these studies provide insight into how the motor system integrates sensory feedback at the behavioural and neural circuit levels.
URI for this recordhttp://hdl.handle.net/1974/28779
Request an alternative formatIf you require this document in an alternate, accessible format, please contact the Queen's Adaptive Technology Centre
The following license files are associated with this item: