Using a Split-belt Treadmill to Evaluate Generalization of Human Locomotor Adaptation

The overall goal os this procedure is to induce locomotor adaptation or long term changes in walking coordination, using a split-belt treadmill which has two belts that can drive each leg at different speeds. This method can help address key questions in the motor learning field, such as how people learn to walk differently and how they can adapt to different conditions. The main advantage of this technique, is that most people have never encountered a split-belt treadmill before, thus, the technique provides an insight into how people adapt to a novel pertubation.

The implications of this technique extend toward gate rehabilitation following stroke because we can use the split-belt treadmill to train somebody with stroke to walk with more symmetric gate coordination. This part of the video gives a general overview of the set up. Conduct all the testing in a quiet environment with minimal peripheral activity, prepare the participant with all the materials needed.

Such as, the motion tracking markers and electromyography sensors. The baseline gate measurement is critical for evaluating normal walking coordination. For overground baseline trials, instruct the participant to walk on a walkway where motion capture data can be collected.

Collect a minimum of 10 stride cycles to establish the baseline. If the motion capture system only allows for capture within a limited space, have the participant perform several passes through that space, after each pass through, the participant should pause, turn and wait for the investigator’s queue to go again. Within limited space, each pass through must have at least two stride cycles, not including the first and last stride cycles which have acceleration changes.

Generally, the speed of the overground baseline trial is the participant’s choice, but if a specific speed is desired, have the participant walk at that speed on the treadmill for one or two minutes to learn the desired speed. Then, during the overground walking, give verbal feedback about their pace between trials. Good and stay at that same speed for another one.

For safety, protect the participant from falls with a harness and when stopping and starting the treadmill, the handrails should always be held. Since the effects of holding handrails are unknown, it’s important to maintain consistency and our movements during the experiment for all participants. Your protocol may require participants to always hold on to handrails during the experiment.

For the treadmill baseline trials, use tight belts and allow the participant a little time to get used to walking on the treadmill. Then, collect data for one to five minutes at each treadmill speed that will be tested post adaptation. While the participant is standing on the stationary treadmill belts, instruct the participant to look straight ahead, not down at their feet.

Do not warn the participant of the split belt difference. The speed difference between the belts will depend on the experiment, typically, use a two to four full difference between belt speeds. Start the split-belt treadmill with one belt running faster than the other and allow the participant to walk for 10 to 15 minutes.

The goal is to collect data of the participant adapting to unanticipated terrain. Once the participant has fully adapted to the split-belt, stop the treadmill briefly, then, briefly restart with both belts running at the same speed. This period is called the catch trial and the purpose is to measure the after effect as it develops over the adaptation period.

The largest after effects occur when the speed of the two belts during the catch trial are the same speed as the slower belt during adaptation. Stop this trial after five strides, then, readapt the participant of the split-belt speeds for at least two minutes and repeat the catch trial if needed. In the current protocol, the second catch trial was conducted at a different speed than the first to evaluate the speed dependency of after effects.

Following the final catch trial, stop the treadmill and restart the treadmill with split-belts for five minutes to allow the participant to readapt. To test after effects on ground walking, transfer the participant to the overground walkway via wheelchair so that their first step over the ground can be recorded. Then, repeat the stride evaluation along the walkway, if desired, instruct the participant to replicate their baseline walking speed and give feedback between trials.

Next, wheel the participant back to the treadmill for a second after effect test. The treadmill after effect test shows how much of the learning was extinguished by the overground walking. Restart the treadmill with tight belts at the slower speed and allow the participant to walk for about 30 seconds.

In order to access the rate of de-adaptation, record continuous tight belt walking for at least 10 minutes to ensure a complete wash-out of the after effects. Two adults under 40, without neurological or orthopedic conditions were subjected to the described paradigm. The experiment began with a baseline overground and a tight belt treadmill measurement at the two speeds of the split-belt.

Then, the split-belt test was performed, followed by two tight belt catch trails at the slow speed and the fast speed, presented in random order. Then, after readapting to the split belt, overground post adaptation effects were measured at the fast or slow speed. The data was plotted by step, perfect gate cemeteries denoted by the zero line, indicating that the step length or double support duration was equal between the two sides.

Both participants walked with near-symmetric spatial and temporal inter-limb coordination during these baseline trials. The split-belt treadmill disrupted normal walking coordination, but people were able to adapt and restore symmetry between one to two minutes of walking or after about 100 strides. Adaptation is demonstrated by the presence of after effects or asymmetric coordination in treadmill catch trials and in post-adaptation overground trials.

Note that the after effect size can be modulated by the speed of walking during treadmill catch trials. After effects during post adaptation over ground trials, are smaller than treadmill after effects and they are less affected by the speed of walking. While attempting this procedure, it is important to control for the effects of potential confounding variables, such as, arm movement, the length and duration of resting breaks and the speed of overground in treadmill walking.

Many aspects of this procedure can be modified, depending on the hypothesis being tested. For example, the rate of acceleration and the variability of split-bell speeds can be altered. Don’t forget that working with gate perturbations can be destabilizing to people, so take the appropriate precautions to protect people from falling.

Thanks for watching and good luck with your experiments.