Manipulating presence influences the magnitude of virtual reality analgesia
Hoffman, et. al
Introduction: Excessive pain during medical procedures performed in unanesthetized patients is frequently reported (Gilron and Bailey, 2003; Karling et al., 2002; Melzack, 1990; Schechter, 1989 and Shang and Gan, 2003) despite the widespread use of analgesic therapies. In clinical settings, side effects of opioid analgesia (e.g. nausea, post-procedure sedation, cognitive dysfunction, and constipation) limit dosage. In contrast, non-pharmacologic techniques typically produce minimal and short-lived side effects, and may serve as valuable adjuncts to traditional pharmacologies. One such non-pharmacologic technique is distraction, which has been shown to help reduce procedural pain in several settings ( Fernandez and Turk, 1989 and Tan, 1982).
Researchers have recently explored the use of immersive virtual reality (VR) as a pain control technique that can be used in combination with traditional pharmacologic therapies. Subjective reports of pain during a variety of painful medical procedures in the clinical setting have been shown to drop approximately 40–50% when patients are distracted by immersive VR (Hoffman et al., 2000a; Hoffman et al., 2000b; Hoffman et al., 2001a; Hoffman et al., 2001b; Hoffman et al., 2004a and Steele et al., 2003).
We theorize that VR analgesia works via an attentional mechanism. Humans have a limited amount of conscious attention available (Kahneman, 1973). Pain requires conscious attention ( Chapman and Nakamura, 1999 and Eccleston and Crombez, 1999). VR systems provide computer-generated multi-sensory input (sight, sound, and more rarely touch, taste and/or smell). Such converging sensory input, and the interactive nature of the experience help give patients the illusion of going into the virtual environment, which can make the virtual world presented difficult for the user's brain to ignore. We theorize that the more intense the patient's illusion of going inside the virtual environment, the more his/her attention will be drawn into the virtual world ( Hoffman, 1998 and Hoffman et al., 2003a), leaving less attention available to focus on pain.
In the present study, some subjects (High Tech VR) used VR hardware (VR helmet, headphones and headtracking system) designed to elicit a strong illusion of VR presence. Others (Low Tech VR) used VR hardware designed to elicit a less compelling illusion of VR presence (see-through VR glasses, no headphones, no headtracking). Regardless of the mechanism of VR analgesia, we predicted that (1) subjects' illusion of ‘going into’ the 3D virtual world (i.e. VR presence) would be greater for the High Tech VR group, and (2) the High Tech VR group would experience more pain reduction than the Low Tech VR group. And we predicted (3) the amount of VR presence reported would be positively and significantly correlated with the amount of pain reduction in VR. In essence, we predicted a measurable dose (increasing VR presence) response (pain reduction) relationship.
In the current study, we compared the relative effectiveness of Low Tech VR vs. High Tech VR distraction on pain ratings during brief thermal pain stimuli. Subjects showed the predicted dose–response relationship: higher VR presence and more pain reduction in the High Tech VR group than in the Low Tech VR group, and a significant positive correlation between subjective presence ratings and amount of VR pain reduction. The results of the present study and preliminary clinical results (Hoffman et al., 2000b and Hoffman et al., 2001a) are consistent with the notion that pain and VR compete for attention. Although the present study does not specifically identify the mechanism of VR analgesia, we speculate that the more attention is directed towards VR, the less attentional resources are available to process incoming nociceptive signals, and the less pain is consciously experienced.
To date, research exploring VR analgesia has used a within-subjects design (Hoffman et al., 2000a; Hoffman et al., 2000b; Hoffman et al., 2001a; Hoffman et al., 2001b; Hoffman et al., 2004a and Hoffman et al., 2004b), such as comparing pain during 3 min of physical therapy without VR to pain during 3 min of physical therapy with VR within the same physical therapy session ( Hoffman et al., 2000b). Potential nuisance variables such as plasma opioid level or how much sleep the patient had the night prior to the study were all controlled using such a within-subject design. One potential limitation of the within-subjects design is that subjects receive (and are thus aware of) both the experimental and control conditions. In the current study a double-blind, between-groups design was used to help reduce demand characteristics.
Eccelston and Crombez (1999) claim that pain is unusually attention grabbing, making it difficult to distract attention away from pain. Similarly, McCaul and Malott (1984) have proposed that distraction works for mild to moderate pain, but is much less likely to reduce extreme pain. In contrast, preliminary clinical results show that VR is able to distract severe burn patients experiencing extreme pain during wound care ( Hoffman et al., 2000a) suggesting that in comparison to VR, pain does not appear to have privileged access to attentional resources. Why VR is able to compete with extreme pain for attentional resources is an important research question. The present results suggest that the illusion of going into the virtual environment may help explain why VR is so effective for reducing various components of the pain experience.
In a previous VR study not involving pain, Hoffman et al. (2003a) tested the fundamental assumption that VR requires conscious attention. Healthy volunteers monitored a string of numbers from a tape recorder for three odd numbers in a row while in VR (helmet worn and turned on) and without VR (helmet worn but turned off). Participants showed a significant reduction in performance on a divided attention task (accuracy in identifying the consecutive odd numbers) while in VR (74% correct) compared to the control condition (95% correct), and they also estimated that the amount of time they were able to attend to the task of monitoring the numbers was significantly higher with no VR than with VR (96 vs. 65%, respectively).
In the present study, compared to the Low Tech VR group, subjects in the High Tech VR group reported a significant increase in how much fun they had during VR. Pain reduction in VR was correlated with how much fun subjects reported having, and is consistent with severe burn patients who report having fun during wound care and physical therapy in High Tech VR (e.g. Hoffman et al., 2004a). In the present study, increasing the ‘immersiveness’ of the VR hardware also led to higher VR presence ratings and was correlated with pain reduction. Studies exploring medical applications of VR exposure therapy for treating anxiety disorders have also described manipulations of the immersiveness of the VR hardware that increased the illusion of presence and increased treatment effectiveness/clinical outcome ( Hoffman et al., 2003b). We predict that further increasing the immersiveness of VR systems in future studies will further increase the participant's illusion of presence in VR, and may increase the magnitude of VR analgesia. Future laboratory and clinical studies should systematically explore (1) the addition of converging sensory input from visual, sound, tactile, smell and vibrotactile (e.g. surround sound) stimulation, (2) increased interactivity between the participants and the virtual world, and (3) which components of the VR environment (including both hardware and software, and individual differences) contribute to the sense of presence and analgesia. Some manipulations that increase presence may also increase simulator sickness (e.g. going faster through the virtual canyon). Care should be taken to minimize simulator sickness in these more immersive VR systems, especially when used adjunctively in clinical studies in patients at risk for nausea from pharmacologic (opioid) analgesics.
VR is a promising non-pharmacologic analgesic, especially for patients who must undergo brief painful procedures. Results from the current study suggest design guidelines for VR analgesia systems. Specifically, highly immersive combinations of VR hardware and software that maximize the user's illusion of presence in the VR environment will likely enhance the effectiveness of virtual reality as a non-pharmacologic analgesic. Selecting participants who have a pre-disposition to feel high presence in VR may also be possible in some applications. Furthermore, we speculate that patients may respond better to some virtual worlds than others. Since excessive procedural pain is a widespread problem for the medical community, and these preliminary results provide additional support for the notion that VR might prove valuable for pain control, additional research on this topic is warranted.