Vection reduction strategies

The main problem with locomotion is vection, which leads to VR sickness. Recall from Section 8.4 that six different kinds of vection occur, one for each DOF. Furthermore, numerous factors were given that affect the sensitivity to vection. Reducing the intensity of these factors should reduce vection and, hopefully, VR sickness.

Several strategies for reducing vection-based VR sickness are:

  1. If the field of view for the optical flow is reduced, then the vection is weakened. A common example is to make a cockpit or car interior that blocks most of the optical flow.
  2. If the viewpoint is too close to the ground, then the magnitudes of velocity and acceleration vectors of moving features are higher. This is why you might feel as if you are traveling faster in a small car that is low to the ground in comparison to riding at the same speed in a truck or minivan.

    Figure 10.7: (a) Applying constant acceleration over a time interval to bring the stopped avatar up to a speed limit. The upper plot shows the speed over time. The lower plot shows the acceleration. The interval of time over which there is nonzero acceleration corresponds to a mismatch with the vestibular sense. (b) In this case, an acceleration impulse is applied, resulting in the desired speed limit being immediately achieved. In this case, the mismatch occurs over a time interval that is effectively zero length. In practice, the perceived speed changes in a single pair of consecutive frames. Surprisingly, case (b) is much more comfortable than (a). It seems the brain prefers an outlier for a very short time interval, as supposed to a smaller, sustained mismatch over a longer time interval (such as 5 seconds).
    \begin{figure}\begin{center}
\begin{tabular}{cc}
\psfig{file=figs/vectionacc.eps...
...nacc2.eps,width=2.7truein} \\
(a) & (b)
\end{tabular}\end{center}
\end{figure}

  3. Surprisingly, a larger mismatch for a short period of time may be preferable to a smaller mismatch over a long period of time; see Figure 10.7.
  4. Having high spatial frequency will yield more features for the human vision system to track. Therefore, if the passing environment is smoother, with less detail, then vection should be reduced. Consider the case of traveling up a staircase. If the steps are clearly visible so that they appear as moving horizontal stripes, then the user may quickly come nauseated by the strong vertical vection signal.
  5. Reducing contrast, such as making the world seem hazy or foggy while accelerating, may help.
  6. Providing other sensory cues such as blowing wind or moving audio sources might provide stronger evidence of motion. Including vestibular stimulation in the form of a rumble or vibration may also help lower the confidence of the vestibular signal. Even using head tilts to induce changes in virtual-world motion may help because it would cause distracting vestibular signals.
  7. If the world is supposed to be moving, rather than the user, then making it clear through cues or special instructions can help.
  8. Providing specific tasks, such as firing a laser at flying insects, may provide enough distraction from the vestibular conflict. If the user is instead focused entirely on the motion, then she might become sick more quickly.
  9. The adverse effects of vection may decrease through repeated practice. People who regularly play FPS games in front of a large screen already seem to have reduced sensitivity to vection in VR. Requiring users to practice before sickness is reduced might not be a wise strategy for companies hoping to introduce new products. Imagine trying some new food that makes you nauseated after the first $ 20$ times of eating it, but then gradually becomes more acceptable. Who would keep trying it?
A final suggestion is to avoid locomotion wherever possible! Try to design experiences that do not critically depend on it.

Steven M LaValle 2016-12-31