Biomedical Engineering Reference
In-Depth Information
However, there is no real consensus about differences between ground and tread-
mill walking. When walking in virtual environments it seems positive to offer natural
interfaces and multisensory feedback, although it is not always possible to physically
walk in as large an environment as displayed in the virtual environment. Hence, all of
the reported differences may lead to perceptual perturbations that could affect pres-
ence. Durgin et al. [ 20 ] found that perceived speed in a richly structured near environ-
ment differ by only about 10% in treadmill compared to wide-areawalking. They also
found that trial-to-trial variations in step frequency predicted changes in perceived
locomotor speed. It has been reported that traveled distance is under-estimated in
virtual environments. However, some authors were able to separate the component
of real walking from other sources that possibly affect distance estimation [ 54 ].
As belt speed variation seems to be a key point to explain these differences,
some authors proposed an innovative treadmill speed controller that compensates
for possible perturbations in real-time [ 66 ]. It seems that this kind of controller does
not disturb immersiveness, but the results are very sensitive to the gains of the belt
speed controller.
To summarize, it seems that treadmill walking may lead to similar kinematic
data to ground walking. This is especially true for treadmills that offer accurate belt
speed control and when users are trained to use treadmills (4-6min training should be
enough to reach this objective). However, perceptual studies seem to demonstrate that
treadmill walking affects distance and speed estimation that could lead to instability.
This is particularly true in immersive environments where other authors reported
that distance evaluation is affected in any type of immersive application. One has to
notice that kinematic data in treadmill walking may result in values similar to those in
ground walking, but dynamic parameters seem to remain different (especially joint
torques and muscle activation patterns). Hence, proprioception in both situations
may be different. This problem is still open and further studies will be necessary to
clearly understand the real advantages and limitations of using treadmills in VR.
3.5 Conclusion
In virtual environments we potentially know everything about the environment and
it is thus possible to imagine infinite possibilities to navigate. However, if we wish
to improve the quality of immersion it is important to notice how people pick-up
information and select the most appropriate action. This perception-action coupling
seems to be very relevant to allow for realistic navigation in VR. One must acknowl-
edge that any additional cognitive load could significantly change things for the user;
using metaphors may add cognitive load that may affect gait quality.
Hence, biomechanical knowledge reported in this chapter could be used in three
main directions. Firstly, it could help to measure the user's navigation wishes us-
ing few biomechanical signals and more natural metaphors. Treadmills have been
widely studied to directly measure speed and direction, but could lead to some gait
perturbations that could affect many dynamic parameters including balance, ground
reaction force and muscle activation. Walk-in-place metaphors and adaptations could
Search WWH ::




Custom Search