The most popular way to gather quantitative data is to have users fill out a questionnaire. Researchers have designed many questionnaires over the years [172]; the most widely known and utilized is the simulator sickness questionnaire (SSQ) [146]. It was designed for simulator sickness studies for the US military, but has been used much more broadly. The users are asked to score each of $ 16$ standard symptoms on a four-point scale: 0 none, 1 slight, 2 moderate, and 3 severe. The results are often aggregated by summing the scores for a selection of the questions. To determine onset or decay rates, the SSQ must be administered multiple times, such as before, after 10 minutes, after $ 30$ minutes, immediately after the experiment, and then 60 minutes afterwards.

Questionnaires suffer from four main drawbacks. The first is that the answers are subjective. For example, there is no clear way to calibrate what it means across the users to feel nausea at level ``1'' versus level ``2''. A single user might even give different ratings based on emotion or even the onset of other symptoms. The second drawback is that users are asked pay attention to their symptoms, which could bias their perceived onset (they may accidentally become like perceptually trained evaluators, as discussed in Section 12.1). The third drawback is that users must be interrupted so that they can provide scores during a session. The final drawback is that the intensity over time must be sampled coarsely because a new questionnaire must be filled out at each time instant of interest.

Steven M LaValle 2020-01-06