Reaching to recover balance in unpredictable circumstances: is online visual control of the reach-to-grasp reaction necessary or sufficient?

Reaching to grasp an object for support is a common and functionally important response to sudden balance perturbation. The need to react very rapidly (to prevent falling) imposes temporal constraints on acquisition and processing of the visuospatial information (VSI) needed to guide the reaching movement. Previous results suggested that the CNS may deal with these constraints by using VSI stored in memory proactively, prior to perturbation onset; however, the extent to which online visual control is necessary or sufficient to guide these reactions has not been established. This study examined the speed, accuracy, and effectiveness of perturbation-evoked reach-to-grasp reactions when forced to rely entirely on either online- or stored-VSI by using liquid-crystal goggles to occlude vision either before or after perturbation onset. The reactions were evoked, in twelve healthy young adults, via sudden unpredictable antero-posterior platform translation (barriers deterred stepping reactions). Prior to perturbation onset, a small cylindrical handhold was positioned unpredictably (by a motor-driven device) at one of four locations in front of the subject. Results indicated that equilibrium could be recovered successfully by grasping the handhold using either online-VSI or stored-VSI to guide the arm reaction; however, both sources of VSI were required for optimal performance. Reach initiation and arm movement were slowed when dependent on online-VSI, whereas reach accuracy and grip formation were impaired when dependent on stored-VSI. Comparison with normal-VSI trials suggests that both sources of VSI are utilized when grasping a small handhold for support under normal visual conditions, with stored-VSI predominating during initiation/transport and online-VSI contributing primarily to final target acquisition/prehension.