Psychophysiological Measures

Psychophysiological techniques address physiological processes such as heartbeat, blood flow, electrodermal processes, reactions of the eyes, and muscular responses. In line with the response similarity paradigm, it can be assumed that as the sense of presence in a VE increases, the physiological responses to the environment will become increasingly similar to those exhibited in a similar real environment (IJsselsteijn, 2004).
Apart from their objectivity, an advantage of physiological measures is that they are continuous, allowing for the assessment of time-varying qualities of presence. As noted before, a disadvantage of physiological measures is that can be difficult to determine what is being measured. Other difficulties are orienting effects which may overwhelm effects of experimental manipulations, individual differences in physiological level (which can be solved by establishing a baseline for each participant), and the fact that some physiological measures, such as skin temperature, change rather slowly (Insko, 2003). Finally, measurement equipment is expensive, and the wearing of sensors may interfere with participants’ experience.

Cardiovascular Measures

Description

Cardiovascular measures are associated with heart rate and blood pressure. An electrocardiogram (ECG) is a graphic produced by an electrocardiograph, which records the electrical current in the heart in the form of a continuous graph. Cardiovascular activity is associated with emotional experience, hedonic valence, orienting response to novelty, defensive responses (Dillon. Keogh, Freeman, & Davidoff, 2000).
Laarni, Ravaja, & Saari (2003) think that both automatic and controlled attention play an important role in presence, for which cardiac measures can be an indicator. Phasic heart rate deceleration is suggested as a measure of automatic attention, and respiratory sinus arrhythmia as a measure of controlled attention.

Research

Meehan, Insko, Whitton, & Brooks (2001) investigated physiological reactions, amongst which was heart rate (HR), in 3 experiments using a stressful VE including a virtual pit room. The first study (n=10, within-subjects design) investigated the hypothesis that presence would decline over multiple exposures on separate days. The second study (n=52, within-subjects design) investigated the hypothesis that passive haptics (the presence of a wooden ledge) would increase presence. The third study (n=33, within-subjects design) investigated the hypothesis that presence would increase when frame rate increases. In all three studies, HR was significantly higher in the Pit Room (as compared to the other 2 rooms of the VE). HR did not decrease significantly after the first exposure, which is attributed to an orienting effect. HR was significantly higher in the condition with the passive haptics. HR was found to correlate with reported presence and reported behavioural presence as measures by a questionnaire. The authors conclude that HR is a promising between-subjects measure.

Dillon, Keogh, & Freeman (2002) investigated the effects of content and visual angle of the display on presence. In their study (n=24, mixed design), the content of a videoclip (amusement, sadness, neutral) was varied within-subjects and the horizontal visual angle (21 or 42 degrees) was varied between-subjects. A significant effect of content was found on HR (greater lowering of HR for Amusement and Sadness material than for Neutral). No effect of angle on HR was found.

Slater, Brogni, & Steed (2003) investigated whether breaks in presence (BIPs) correspond to changes in physiological state, such as HR. In an experiment which examined the effect of 6 different VE scenes on presence, subjects (n=60, between-subjects design) visited urban VE’s inside a CAVE-like system, and were instructed to report BIPs by pressing a button. Heart Rate was measured during this experience. Mean HR was found to increase and reach a peak approx. 1 second before a BIP is signalled. Additional data was gathered to make sure that this was not merely a result of (the intention of) pressing the button.

In an experiment of Wiederhold et al. (2001), participants (n=72) took part in a 6 minute VR airplane flight. Subjective measures included HR and a questionnaire on Presence and Realism (taken from Parent, 1998). HR showed significant correlations with both Presence and Realism.

Ravaja (2002) used respiratory sinus arrhythmia (RSA) as a measure for attention/engagement. In his experiment, participants (n=36, within-subjects design) saw financial news messages presented on a very small display, accompanied by either a static or a moving face. RSA scores were lower in the moving face condition.

Sensitivity:Heart rate discriminated between stressful/non-stressful conditions and the absence/presence of passive haptics in Meehan’s experiment, and between different types of content in Dillon et al. RSA scores discriminated between conditions in Ravaja's experiment.
Reliability:Not reported
Validity:Correlations between heart rate and reported presence and reported behavioural presence in Meehan’s experiment. Correspondence to reported Breaks in presence (Slater et al.). Correlations between heart rate and subjective presence and realism (Wiederhold et al.).

Sources

Skin Measures

Description

There are two psychophysiological measures related to the skin: skin temperature (ST) and skin conductance (SC). Measures of skin conductance, which is also referred to as electrodermal activity (EDA) or galvanic skin response (GSR), record changes in the electrical conductance of the skin, generally from the fingers or palms. SC is associated with emotional arousal, memory effects, and orienting response to novelty (Dillon et al., 2000).

Research

Meehan et al (2001) used SC and ST in 3 experiments using a stressful VE including a virtual pit room (the studies are described in the section cardiovascular measures). In all three studies, SC was significantly lower in the Pit Room as compared to the other 2 rooms of the VE. SC decreased significantly after the first exposure. SC was significantly higher in the condition with the passive haptics, whereas ST showed the opposite effect than was expected. SC was found to correlate significantly with reported presence and reported behavioural presence as measured by a questionnaire. No correlations with ST were found.

Dillon et al (2002) used SC in a study investigating the effects of content and visual angle of the display on presence (the study is described in the section cardiovascular measures). Measures also included the SUS questionnaire, and items relating to Engagement and Negative Effects taken from the ITC-SOPI questionnaire. SC was generally (significance is not reported) higher in the Amusement category. A larger angle was associated with a lower negative SC deviation from baseline. This angle was also rated higher on the second SUS item.

In their study investigated the relation between breaks in presence (BIPs) and physiological measures (this study was described in the section cardiovascular measures), Slater et al (2003) measured subjects’ SC. Mean SC was found to increase and reach a peak approximately 1.8 seconds after a BIP is signalled. Additional data was gathered to make sure that this was not merely a result of (the intention of) pressing the button. In a control group where anxiety was induced, SC showed a similar response, suggesting that a BIP may be a stress-inducing event.

In Wiederhold et al.’s (2001), VR airplane flight experiment (study is described in the section cardiovascular measures), ST showed significant correlations with both Presence and Realism questionnaire scores.

Dillon et al (2000) intend to compare electrodermal activity to post-test subjective ratings (ITC-SOPI) in high and low immersion conditions (realized by manipulation of stereoscopic cues). Sensitivity:SC measures discriminated between stressful/non-stressful conditions and the absence/presence of passive haptics in Meehan’s experiment, and between different content in Dillon et al. Sensitivity of ST was not supported.
Reliability:Not reported
Validity:Correlations between SC and reported presence and reported behavioural presence in Meehan’s experiment, and with SUS item in Dillon et al. Correspondence to reported Breaks in presence (Slater). Correlations between ST and subjective presence and realism (Wiederhold).

Sources

Ocular Measures

Description

There are many psychophysiological measures associated with the eyes. Two have been suggested for presence research: eye tracking and pupil response. Eye tracking measures fall into two categories: spatial and temporal (Goldberg & Kotval, 1999). Spatial measures are, for example, amplitude of saccades and scanpath length. Temporal measures include fixation duration, fixation number, and scanpath duration. Laarni, Ravaja, & Saari (2003) discuss the use of eye tracking measures for presence research. Three methods suitable for researching spatial presence are identified:

  1. measure to what degree the user’s attention is distracted away from media
  2. analyse what aspects of the mediated info the user looks at and the order in which different areas of media stimuli are processed
  3. distinguish whether user uses focused attention or distributed attention strategy
Pupilometry is the study of how a pupil reacts to different emotions and stimuli. Whereas facial expressions can be inhibited, the dilation and contraction of the pupils is an involuntary response not controllable by cognitive means. Huang & Alessi (1999) suggest that this approach could be used in presence research.

Research

Not reported. Laarni et al. (2003) propose that eye tracking should be combined with other continuous measures such as phasic heart rate deceleration (as a measure of automatic attention) and respiratory sinus arrhythmia (as a measure of controlled attention).

Sources

Facial Electromyography (EMG)

Description

Facial EMG is recorded by surface electrodes placed on the skin of the face. The potential difference measured at the electrode can be used as an indication for emotion.

Research

Facial EMG was used in an experiment by Ravaja (2002) (study was previously described in the section cardiovascular measures). Participants viewed financial news messages presented on a very small display, accompanied by either a static or a moving face. EMG scores were higher in the moving face condition, as were subjectively rated emotional reactions and arousal. Sensitivity:EMG discriminated between different conditions
Reliability:Not reported
Validity:similar results of subjective emotional measures

Sources
  • Ravaja, N. (2002). Presence-related influences of a small talking facial image on psychophysiological measures of emotion and attention. In Proceedings of the 5th International Workshop on Presence. Porto, Portugal, October 9-11 2002.