It is feasible to use a functional virtual environment for investigation of dual tasking. Different gait strategies, including an increase or decrease in gait speed, can be used to cope with the increase in cognitive demands required for dual tasking.

The aim of the present study was to investigate the relationship between proactive and reactive components of postural control. We contrasted the kinematic and electromyographic (EMG) responses to multidirectional voluntary leg lifts with those elicited by unexpected surface tilts. In particular, we addressed the role of trunk stabilization following either a voluntary or forced weight shift from double to single limb support. Nine young female subjects stood with a standing posture of 45 degrees toe-out and their arms abducted to shoulder level. On the experimenter's signal, subjects either (1) lifted one leg as fast as possible in one of six directions (R/L side, R/L diagonal front, R/L diagonal back) to a height of 45 degrees or (2) maintained standing as the support surface tilted at a rate of 53 degrees /s to a height of 10 degrees in one of six directions (R/L-up, R/L diagonal toes-up, R/L diagonal toes-down). For both tasks, our results showed that the center of pressure (COP) displacement began before or in conjunction with displacement of the center of mass (COM), after which the COP oscillated about the horizontal projection of the COM. In addition, the muscles were recruited in a distal-to-proximal sequence, either in anticipation of the voluntary leg lift or in response to the sudden surface tilt. Thus, the COP was being used dynamically to control displacement of the COM. The axial postural strategy comprising head, trunk, and pelvis movements was quantified by means of principal component analysis. More than 95% of the variance in the data could be described by the first two eigenvectors, which revealed specific coordination patterns dominated by pelvis rotation in one direction and head/trunk rotation in the opposite direction. Unexpected surface tilting elicited an automatic response strategy that focused on controlling the orientation of the head and trunk with respect to the vertical gravity vector while trunk verticality was compromised for movement generation and the recovery of postural equilibrium during leg lifting. In conclusion, regardless of the type (voluntary versus involuntary) or direction of perturbation, the strategy employed by the central nervous system to control the body COM displacement concerns mainly trunk stabilization.

Virtual reality (VR) technology offers a new and safe way to increase practice time and provide the varied environments and constraints needed to optimize locomotor training. Our specific objectives are (1) to create a virtual environment (VE) coupled with a self-paced treadmill for locomotor training; (2) to compare temporal and distance measurements of gait during treadmill walking while looking at different scenarios of VE; and (3) to develop a protocol optimized for the training of locomotor disorders after stroke. A motorized treadmill was mounted on a six-degree-of-freedom motion platform. VEs were created using commercial software (SoftImage) and projected on a large screen, while system control was administered through the CAREN software (Motek BV). The instantaneous treadmill speed and scene progression were servo-controlled. Preliminary results show that healthy subjects are able to have full control of their own walking speed both on the treadmill and within the virtual scene, while experiencing a strong sense of presence. A street crossing training protocol has been developed for locomotor training. It is expected that locomotor training with increasingly complex VEs will allow persons with stroke to increase progressively their locomotor capacity, as required and entrained by the VE.

No abstract