Fast responses to stepping-target displacements when walking

Abstract

Key points: Goal-directed arm movements can be adjusted at short latency to target shifts. We tested whether similar adjustments are present during walking on a treadmill with shifting stepping targets. Participants responded at short latency with an adequate gain to small shifts of the stepping targets. Movements of the feet during walking are controlled in a similar way to goal-directed arm movements if balance is not violated.

Abstract: It is well-known that goal-directed hand movements can be adjusted to small changes in target location with a latency of about 100 ms. We tested whether people make similar fast adjustments when a target location for foot placement changes slightly as they walk over a flat surface. Participants walked at 3 km/h on a treadmill on which stepping stones were projected. The stones were 50 cm apart in the walking direction. Every 5-8 steps, a stepping stone was unexpectedly displaced by 2.5 cm in the medio-lateral direction. The displacement took place during the first half of the swing phase. We found fast adjustments of the foot trajectory, with a latency of about 155 ms, initiated by changes in muscle activation 123 ms after the perturbation. The responses corrected for about 80% of the perturbation. We conclude that goal-directed movements of the foot are controlled in a similar way to those of the hand, thus also giving very fast adjustments.

Keywords: adjustment; gait; target jump; visual perturbation.

Figure 1. Set‐up

A, side view of a participant who is standing on the left leg (orange). The right leg (red) is swinging to a green stepping‐target that shifts laterally. The belts of the treadmill are black and cover similarly sized force‐plates. For clarity, the projector and the motion capture cameras surrounding the measurement field are omitted from this picture. The high‐speed camera was used to determine the exact timing of the perturbation relative to the gait. B, top view of the treadmill with the same stimulus as in panel A. The red arrow indicates a 2.5‐cm lateral displacement of a stepping target for the right leg. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 2. Example of a single participant's behaviour

A, C and E, various actual values when the right leg was confronted with a lateral shift (red), in the matching steps during which it was not (black) and in all steps during which it was not (cyan). B, D and F, difference between values (responses) for steps with and without the lateral shift. This isolates the response (red) from which we can determine the latency (using the black lines; latency itself indicated by vertical red lines). A and B, lateral foot velocity. C and D, lateral velocity of change in centre of pressure. E and F, EMG of gluteus medius of the right (swing) leg. A single perturbed step is presented in panel A (red). All other red and purple traces show the means of the 60 perturbed steps. The curves in the lower parts of panels A and B represent the standard deviation (SD) of the mean across steps (709 steps for cyan, 20 for black and 60 for red and purple). The continuous vertical line indicates the moment at which the target shifts (time zero). The dotted and dashed vertical lines show the average moments of toe‐off and heel‐strike, respectively. The latencies in B, D and F are 142 ms, 128 ms and 115 ms, respectively, for this participant. Using all 709 unperturbed steps as the reference would have resulted in similar responses to using the selected reference at the overall level (purple vs. red, panel D and F), but it is not as accurate as using the selected reference at the single‐trial level (cyan vs. black, panel A). [Color figure can be viewed at wileyonlinelibrary.com]

Figure 3. Data points for single steps and trace illustrating the trend

The data (SMART method) are for the 60 perturbed steps that were analysed for Fig. 2 (one participant, right leg, lateral perturbation). The left part of the trace is dashed because in the further analysis (Fig. 8) we will only plot the remaining time from 400 to 200 ms as 95% of perturbed steps of all participants fell within this time window. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4. Medio‐lateral responses to target shifts as a function of the time after the perturbation

A, foot responses. B, centre of pressure responses. Positive responses correspond to a change in velocity in the same direction as the perturbation. Curves in the upper part are the average responses; the ones in the lower part show the SD across the 20 participants. Vertical lines show the average onset (toe‐off, dotted) and offset (heel‐strike, dashed) of the leg‐swing, and the moment of perturbation (continuous line at time zero). [Color figure can be viewed at wileyonlinelibrary.com]

Figure 5. Average responses of the 20 participants in the anterior‐posterior direction, perpendicular to the medio‐lateral target shifts

A, foot responses. B, centre of pressure responses. Positive responses are in the anterior direction. Other details as in Fig. 4. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 6. The EMG of all 8 muscles for steps with medial and lateral target shifts

Responses are averages across 19 participants for VL and VM, and across 20 participants for the other 6 muscles. EMG magnitude is shown for both the swing leg and stance leg as a function of time after the target shift. The EMG in the corresponding reference steps is indicated by the two black traces, sometimes hidden behind the coloured traces. VL, vastus lateralis; VM, vastus medialis; BF, biceps femoris; ST, semitendinosus; GlM, gluteus medius; TA, tibialis anterior; GaL, gastrocnemius lateralis; GaM, gastrocnemius medialis. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 7. Main muscle activity changes in response to target shifts as a function of the time after the perturbation

Curves are averages; shaded areas represent the SD across the 20 participants (19 participants for i‐VM and i‐VL). Muscles are ordered by the latency of their response (indicated by a coloured vertical line on the time axis). Left column: lateral target shift. Right column: medial target shift. The ‘i’ and ‘c’ in front of the muscle names indicate ipsilateral (the swing leg) and contralateral (the stance leg), respectively. Muscle abbreviations are detailed in Fig. 6. Note that the latencies indicated are based on the average response rather than on individual responses as in Table 1. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 8. How the medio‐lateral response to a medial or lateral target shift depends on the remaining time until heel‐strike

A, correction. B, peak velocity. Shaded areas represent the SD across the 20 participants. [Color figure can be viewed at wileyonlinelibrary.com]