Split-second decisions on a split belt: does simulated limping affect obstacle avoidance?

During normal gait a suddenly appearing obstacle is avoided either by making a large crossing step (long-step strategy, LSS) or by interrupting the swing phase (short-step strategy, SSS) depending on the time of appearance of the obstacle in the step cycle. Limping changes the proportion of time spent in the swing phase and the question arises whether this could affect the ability to avoid obstacles. This was investigated using a split-belt treadmill that induces behavior that is similar to limping even in normal adults. Subjects walked on a split-belt treadmill with speed ratios between left and right of 2:2 up to 2:8 km/h in combination with obstacle avoidance (OA) on the slow belt. The failure rate of obstacle avoidance increased markedly as speed differences between legs increased. This increment was paralleled by augmented use of the SSS, related to an increase in time pressure. In all split-belt walking conditions, the alternative strategy (LSS) yielded less OA failures but it required much longer preparation time than the SSS. In addition, the prolonged stance phases prior to crossing in the LSS required a concomitant prolongation of the contralateral swing phase. This was difficult to achieve at times and as a result the swing phase was sometimes interrupted, giving rise to a contralateral SSS (and a 2:1 coordination pattern). It is concluded that simulated limping greatly increases the risk of failing to avoid suddenly appearing obstacles.