Physiological Presence Measures

Physiological measures for presence are in principle very attractive, as at least some of them can be recorded fairly unobtrusively while the subject is participating in the virtual environment. This would potentially allow for a real-time response to the subject's level of presence. A list of possible physiological measures for presence is given by Barfield and Weghorst [8] and includes posture, muscle tension and cardiovascular and ocular responses to virtual events. Neurological measures might also be considered.

Held and Durlach [38] suggest a measure for presence based on the ability of an environment to produce a ``startle response'' to unexpected stimuli. More generally, to what extent can a virtual environment produce responses which imply that observers interpret it as the real environment?

Unfortunately, there is currently little evidence that physiological measures correlate well with presence. Sheridan [93] (p. 209) mentions physiological measures, stating that ``It is natural to seek an objective measure or criterion that can be used to say that telepresence or virtual presence have been achieved. However telepresence (or virtual presence) is a subjective sensation, much like mental workload, and it is a mental model -- it is not so amenable to objective physiological definition and measurement.''

Nevertheless, the use of postural measures have received attention. Ohmi [69] claims a good correlation between the angle of body tilt and the subjective impression of acceleration. Hoshino et al. [43] used body sway to measure the presence of observers exposed to various frequencies of a rolling stimulus displayed on an HMD; on a comsumer 3D television; and on a large 3D display. Body sway was measured in terms of head displacements. They reported highest body sway in the case of the HMD with a .33 Hz stimulus.