Fig 1.
The particpant walked over 3 overground forceplates. The participant wore 8 surface electromyography (EMG) sensors on each leg. We used motion capture markers to determine kinematics. The participant walked through infrared timing gates before and after the force plates. We used the timing gates to measure walking speed. The bodyweight support system transferred support force to the participant via a modified rock climbing harness.
Fig 2.
Average ground reaction forces of 12 participants walking overground with bodyweight support.
Gait cycle was defined as starting with right heel strike (0%) and ending at the next right heel strike (100%). Forces were normalized to bodyweight at 1 G.
Fig 3.
Saggital plane ground reaction force vectors from walking at 1.2 and 1.6 ms-1 overground with bodyweight support.
The timings of the vectors are: Leftmost vectors found at initial peak of vertical ground reaction force, middle vectors found at the lowest vertical ground reaction force after initial peak, and the rightmost vetors were found at the second peak of vertical ground reaction force. The vectors make the typical ‘m’ shape associated with the vertical ground reaction force pattern of healthy walking. The dotted horizontal line shows vertical ground reaction force equal to bodyweight. The scale of both plots is equal.
Fig 4.
Ankle joint angle, moment, and power from walking with bodyweight support at 4 speeds.
Data were averaged from 12 participants, and gait cycle was right heel strike to consecutive right heel strike. The grey lines represent right toe off at different levels of simulated gravity. Bodyweight support is shown as gravity, with normal walking at 1 G, and bodyweight support of 69% bodyweight as 0.31 G. Positive angle and moment is plantarflexion, negative angle and moment is dorsiflexion. The ankle angle in neutral standing position is 0°. Positive power is power generated, negative power is power absorbed. Magnitude of moment and power decrease with simulated gravity.
Fig 5.
Knee joint angle, moment, and power from walking with bodyweight support at 4 speeds.
Data were averaged from 12 participants, and gait cycle was right heel strike to consecutive right heel strike. The grey lines represent right toe off at different levels of simulated gravity. Bodyweight support is shown as gravity, with normal walking at 1 G, and bodyweight support of 69% bodyweight as 0.31 G. Positive angle and moment is knee extension, negative angle and moment is knee flexion. Positive power is power generated, negative power is power absorbed. Magnitude of moment and power tends to decrease with simulated gravity.
Fig 6.
Hip joint angle, moment, and power from walking with bodyweight support at 4 speeds.
Data were averaged from 12 participants, and gait cycle was right heel strike to consecutive right heel strike. The grey lines represent right toe off at different levels of simulated gravity. Bodyweight support is shown as gravity, with normal walking at 1 G, and bodyweight support of 69% bodyweight as 0.31 G. Positive angle and moment is hip extension, negative angle and moment is hip flexion. Hip angle in neutral standing position is 0°. Positive power is power generated, negative power is power absorbed. Magnitude of moment and power decrease with simulated gravity.
Fig 7.
Peak joint moment and powers from walking overground with bodyweight support.
Error bars are 1 standard deviation. Early-mid stance knee extension moment is the initial peak of knee extension moment. Transition knee extension moment is the peak extension moment that occurs around push-off (the second peak). Bodyweight support had a signigicant effect on hip, early-mid stance knee, and ankle moments (p < 0.01). All peak joint powers were also reduced with decreasing gravity (p < 0.001).
Table 1.
Peak ground reaction force descriptive data from overground walking with bodyweight support.
Table 2.
Statistical results of significance of gravity level on dependent variables.
Fig 8.
Time series muscle activation in response to bodyweight support for overground walking at 4 speeds.
The linear envelopes are data averaged from 12 participants. Muscle activation was normalized at the individual participant level to maximum activation across all walking conditions. Gait cycle was right heel strike to consecutive right heel strike. The grey lines represent right toe off at different levels of simulated gravity All plots have the same axes.
Fig 9.
Response of RMS muscle activity to bodyweight support at 4 different overground walking speeds.
Data is presented as a spider plot. Each spoke on the compass represents a muscle. The circular bands represent level of normalized activation, with the smallest activation at the innermost band and a maximum activation of 1 at the outer-most band. RMS data was normalized to maximum RMS activation of each muscle at the participant level. The top row represent the RMS data in stance phase, and the bottom row of spider plots represent the data in swing phase. The shortans for the muscles are: Rectus femoris (RecFem), Vastus lateralis (VasLat), Vastus medialis (VasMed), Biceps femoris (BicFem), Medial gastrocnemius (MedGas), Lateral gastrocnemius (LatGas), Soleus (Sol), and Tibialis anterior (TibAnt).
Fig 10.
Average and standard deviation of right stride length across conditions.
The stride length is normalized to leg length.
Table 3.
Effects of bodyweight support on muscle activity found in previous studies.