Methods

There were 15 adult volunteers (10 male, 5 female). Twelve reported more than 10 minutes prior experience with virtual environments. The participants are summarized in Table 6.2.

A pre-exposure SSQ was administered to check for pre-existing nausea or related symptoms. No participants or data were removed on this basis.

**Table 6.2:** Participant Overview for Experiment AIIIE1
Id	Gn	Age	Prev
72	M	23	1
73	F	25	1
74	M	22	1
75	M	24	1
76	M	32	1
77	M	41	0
78	F	33	0
79	M	22	1
80	F	44	1
81	M	24	0
82	F	41	1
83	F	25	1
84	M	41	1
85	M	42	1
86	M	29	1

Id is the participant number. Gn gives the gender, Age the age in years, and Prev is ``1'' if the participant had more than 10 minutes of prior experience in virtual environments, ``0'' otherwise.

See Section 3.7 for a description of the equipment. The experiment was conducted as a within-subjects design. The order of conditions was counter-balanced across participants.

A circular motion visual stimulus was created by placing a videocamera on a tripod in an open plaza on the University of Washington campus and continuously rotating the camera through 360 $^{\circ}$ in yaw with a period of six seconds. The location was picked for visual salience, having a variety of sharp edges and both vertical and horizontal features. The angular velocity of 60 $^{\circ}$ /sec was picked in accordance with literature indicating that motion sickness peaks at this value [30]. Other sources suggest that vection increases with stimulus velocity up to about 90 $^{\circ}$ /sec [20], at least for stimuli consisting of vertical stripes.

The head-mounted display (HMD) used to display the videotape was a Virtual i-O i-glasses! VTV/VPC (see Section 3.7.3 for HMD specifications). The HMD was masked so that no visual cues were available except those on or through the display. (That is, the edges of the optics and HMD were occluded.)

For each participant, there were two separate three-minute sessions in which he/she was exposed to the circular motion stimulus, one session in each of the see-through (IVB) and occluded (no IVB) modes (see Figure 3.2 on page ). The sessions were separated by 5 minutes. During the first and third minutes of each session, participants were asked to stand in the ``Sharpened Romberg'' stance. This stance consists of placing one foot in front of the other, heel touching toe, weight evenly distributed between the legs, arms folded across the chest and chin up [32]. In the second minute, to avoid fatigue, participants were allowed to stand in a relaxed posture, but were instructed to continue to look forward. Half-way through the first, second and third minutes, participants were asked to roll their heads down to their right shoulder, then to their left shoulder, then return their head to the upright position. This action was intended to increase motion sickness by introducing a visual-vestibular interaction.

Participants were instructed to keep their eyes open and to look forward. They were told that they could break the Sharpened Romberg stance if necessary, but that they should get back into it as quickly as possible. Other than maintaining postural stability and looking forward, participants were not given a task in Experiment AIIIE1.

A verbal rating of the relative visibility of the background in the see-through condition was recorded in the post-test questionnaire. This was intended to make sure that the IVB in the see-through condition was not simply ``washing-out'' the circular vection stimulus provided by the HMD. A 1-5 scale was used, with the low end indicating that only the foreground scale was visible, and the high end indicating that only the background was visible.

Three dependent measures were recorded in Experiment AIIIE1 to measure the effect of the IVB. The first was a measure of per-exposure ataxia: the total number of stance breaks in the first (SB1) and third (SB3) minutes of each session. A break was defined as a translation of either foot and/or an uncrossing of the arms. Participants were asked to maintain the Sharpened Romberg stance, but if a break was necessary they were to asked to get back into Sharpened Romberg as soon as possible.

The second dependent measure was the post-exposure SSQ, to record self-reported symptoms of simulator sickness.

The third dependent measure was a post-exposure rating of vection, defined in terms of the following question. ``While in the virtual environment, did you get the feeling of motion (i.e., did you experience a compelling sensation of self-motion as though you were actually moving)?''. The endpoints of the 1-7 scale were anchored as ``not at all'' and ``very much so'', respectively.