Virtual reality in the application of heart rate variability biofeedback

https://doi.org/10.1016/j.ijhcs.2019.06.011Get rights and content

Highlights

  • VR provides a feasible way of performing heart rate variability biofeedback.

  • VR improves motivation to engage in heart rate variability biofeedback.

  • VR enhances attentional focus in heart rate variability biofeedback.

Abstract

Heart rate variability biofeedback (HRV-BF) is an established technique for reducing stress and regaining balance of the autonomic nervous system. The present study explored (1) the feasibility and (2) the potential advantages of using a virtual reality (VR) nature environment as an immersive way of providing HRV-BF. A randomized controlled experiment was conducted, comparing (1) a single session of HRV-BF in a VR-based implementation (treated group) with (2) a traditional implementation (treated control group) and (3) a non-treated control group. Both the VR-based and the traditional version of HRV-BF are equally effective in increasing short-term HRV. However, the VR-based implementation was associated with higher motivation and helped users better maintain attentional focus. Future research needs to further clarify the underlying mechanism and address possible long-term effects.

Introduction

Heart rate variability biofeedback (HRV-BF) has been demonstrated to be an effective and efficient prevention and treatment for a wide range of stress related conditions. Evidence has been provided by numerous empirical studies (for a recent meta-analysis, see Goessl et al., 2017), including the treatment of somatic and psychosomatic disorders (for a review, see Gevirtz, 2013 or Wheat and Larkin, 2010) as well as non-clinical and occupational applications (Löllgen et al., 2009, McCraty et al., 2003, McCraty et al., 2009a). Besides stress-reducing effects, HRV-BF has been shown to increase performance and reduce anxiety in sports (Paul and Garg, 2012), educational testing (Lagos et al., 2008, Henriques et al., 2011) and cognitive tasks (Prinsloo et al., 2011, Prinsloo et al., 2013a, Prinsloo et al., 2013b). While biofeedback is typically designed as a long-term training, studies have shown that a single session of HRV-BF reduces state anxiety and boosts relaxation (Prinsloo et al., 2011, Prinsloo et al., 2013a, Prinsloo et al., 2013b, Sherlin et al., 2009).

HRV-BF taps into the phenomenon of respiratory sinus arrhythmia (RSA), which are natural heart rate fluctuations modulated by respiration (Hayano et al., 1996). Inhaling inhibits parasympathetic nervous system activity, which in turn increases heart rate. Exhaling increases parasympathetic nervous system activity, which decreases the heart rate. A greater RSA results in a more efficient gas exchange in the lungs as heart rate is greater when oxygen saturation is higher and vice versa (Lehrer et al., 2003; Vaschillo et al., 2006, Yasuma and Hayano, 2004). At a breathing frequency of about 0.1 Hz (i.e., six breath cycles per minute, so-called resonance frequency; Lehrer et al., 2003) the RSA reaches its maximum, resulting in a high HRV. Breathing and heart rate oscillations are perfectly coupled (0° phase shift), while heart rate and blood pressure oscillations are 180° out of phase (Shaffer et al., 2014, Vaschillo et al., 2002). This state is also referred to as heart coherence (McCraty et al., 2009b, Vaschillo et al., 2006). In biofeedback, the trainee is instructed to maximize RSA through paced breathing at resonance frequency. The intervals between adjacent heartbeats are measured, processed, and fed back in real-time. While there is a range of different feedback implementations, HRV-BF is typically administered visually in the form of numerical indicators, charts, abstract graphical elements or colors on a 2-dimensional screen (Peek, 2017, Yu, 2018). In many cases, this works reasonably well. Nevertheless, these traditional forms of feedback bear two possible hindrances that may reduce or impede the success of the session, especially for novice users: (1) HRV-BF provides feedback over complex features of the user's cardiovascular system (Lehrer and Gevirtz, 2014). For inexperienced users, the feedback may be intimidating at times, especially if the target behavior cannot be shown immediately, as is often the case (Yucha and Montgomery, 2008). Receiving negative feedback may result in a counter-productive stress reaction, which can lead to an unfavorable breathing pattern (cf. Boiten et al., 1994, Masaoka and Homma, 1997, Suess et al., 1980) and in turn could trigger a negative feedback loop. (2) Depending on the intensity of visual appeal, simple forms of biofeedback stimuli carry the risk of being too monotone over time (Huang et al., 2006, Soyka et al., 2016). Users might not experience these forms of feedback as rewarding and motivating for long, possibly resulting in waning adherence throughout the training.

Both potential hindrances may keep users from maintaining an adequate training routine or from even establishing a routine in the first place. In addition to a sought-out measurement implementation, providing motivational instructions and proper psychoeducation, researchers have created novel approaches to biofeedback, utilizing technology to foster engagement and immersion. For example, biofeedback has successfully been administered in the form of immersive installations using sounds of nature (Yu et al., 2018a), soothing ambient light (Yu et al., 2018b) and sound-light combinations (Yu et al., 2018c), animated smartphone games (Dillon et al., 2016, Sonne and Jensen, 2016) and virtual reality (VR) applications (van Rooij et al., 2016). In a similar way, the present exploratory study reports on an immersive implementation of HRV-BF, specifically designed to tackle the abovementioned common problems to biofeedback by putting a focus on comfort and relaxation while still offering an immersive experience. We suggest that realistic virtual nature environments present a suitable setting in which biofeedback can be embedded in an immersive way. We use the term VR-based biofeedback to describe such an immersive feedback in the form of simulated virtual environments (independent of the technical delivery modality). As a feasibility test, the study compares this VR-based HRV-BF with a more traditional implementation using graphical indicators. In light of unknown effect sizes in this exploratory study, we designed the virtual nature environment to be as immersive as possible. Besides aiming for realism in the design of the virtual environment, we delivered the embedded feedback via a head-mounted display (HMD). It should be noted that this keeps the study from disentangling the effects of biofeedback representation (VR nature environment vs. abstract graphical indicators) and delivery modality (HMD vs. 2D monitor). However, in this single-session pilot study, such a contrasting approach seemed favorable in terms of identifying potential subtle differences and generating future research avenues and informed hypotheses.

Lately, VR has received much attention due to the advance in HMD technology, which have stepped up from expensive, high-end equipment to available, consumer-ready and even mobile devices with potential mass adoption. In contrast to traditional ways of providing visual stimuli, (head-mounted) VR offers immersive ways of experiencing virtual environments. VR has been shown to complement or even substitute traditional therapy (e.g., Baños et al., 2011, Powers and Emmelkamp, 2008) as well as stress management (e.g., Anderson et al., 2017, Villani and Riva, 2012). In the following, we will first provide arguments for why we think it is worth investigating virtual nature environments as a setting for biofeedback and will then discuss how such a VR-based HRV-BF could be designed to enable adequate training.

To begin with, virtual nature environments provide a setting that prevents the user from being rattled by the feedback. The Attention Restoration Theory (ART, Kaplan, 1995, Kaplan and Kaplan, 1989) asserts that spending time in as well as looking at environments with restorative qualities (e.g., nature settings) allows the brain to replenish its voluntary attention capacity, thereby improving concentration and relieving mental fatigue and stress (e.g., Ulrich, 1984, Ulrich et al., 1991). In line with ART, nature environments have been shown to lower stress levels and enhance productivity (Berto, 2014, Bowler et al., 2010, Kaplan, 1995), even when simulated (Anderson et al., 2017, Gromala et al., 2015, Hartig et al., 1996, de Kort et al., 2006, Valtchanov et al., 2010, Villani and Riva, 2012; for a recent review, see also Ohly et al., 2016). Virtual nature environments can be delivered via different modalities. The more immersive and realistic the experience, the better the restorative and relaxing effect (de Kort et al., 2006, Slater and Wilbur, 1997, Villani and Riva, 2012). Modern HMDs stand out from other visual media through the higher degree of sensory fidelity (immersion) and hence, capability to induce a sense of presence (Buttussi and Chittaro, 2018, Cummings and Bailenson, 2016, Makransky et al., 2019). Experiencing virtual nature environments via an HMD appears to be a promising way of mimicking actual nature experiences (Anderson et al., 2017, Annerstedt et al., 2013, Freeman et al., 2004, Villani and Riva, 2008, Villani et al., 2007). Providing an immersive virtual nature environment as a setting in which the biofeedback is to take place (cf. Kosunen et al., 2016), can both put the user in a relaxed and confident state and buffer the impact of potentially negative feedback.

Besides adding a restorative effect, feedback in the form of changes in natural scenes such as moving clouds, starting or abating rain and sunshine may also decrease the risk of monotony and hence boost motivation. A fundamental mechanism of biofeedback training is operant conditioning through positive reinforcement (cf. Gaume et al., 2016, Sherlin et al., 2011). Therefore, the choice of a suitable reinforcer is crucial to the learning success as well as to the trainee's motivation to keep engaged in the training. In biofeedback, it is crucial to make the feedback itself as rewarding as possible (Strehl, 2014). This includes providing feedback stimuli that are both meaningful and valued by the user, and can be experienced as real. Immersion has been found to make feedback more diverse, interesting, and attractive (Bohil et al., 2011, Gromala et al., 2015, Kaiser and Othmer, 2000). Altering a realistic environment through physiological changes in biofeedback may have a more metaphorical, intuitive and thus meaningful impact than numerical indicators or charts that need to have a valence assigned through instruction. With abstract, data centered feedback, the user would mostly watch or observe the feedback. Realistic simulated virtual environments, on the contrary, enable the user to experience the feedback, as if it were real. This exerts a stronger impact on the user's emotions (cf. Felnhofer et al., 2015, Riva et al., 2007) and learning experience (cf. Sutcliffe et al., 2005). Again, a more sophisticated and immersive delivery format is expected to strengthen the effect. Thus, the rich and experienceable virtual environment in which the biological feedback parameter is integrated has the potential to maintain the trainees’ motivation and engagement.

Integrating HRV-BF in virtual environments appears to be a promising endeavor. However, design and proper implementation are paramount. While there does not seem to be the one best virtual environment in terms of relaxation or the one best feedback implementation, the literature allows for identifying certain beneficial elements. Following the argumentation in Section 1.2.1, virtual nature environments seem to be well-suited as the basic setting. To derive more detailed implementation recommendations, we apply the reasoning of ART (Kaplan, 1995, Kaplan and Kaplan, 1989) to HRV-BF. Accordingly, a virtual environment should meet four criteria:

  • (1)

    Provide a feeling of being away: The environment should be different from everyday surroundings. This can easily be achieved in VR due to its potential to simulate any environment while providing a feeling of actually being there (cf. research on sense of presence; Buttussi and Chittaro, 2018). Additionally, the possibility to influence the virtual environment through one's own physiological changes represents a break from day-to-day experience.

  • (2)

    Include stimuli that allow for involuntary attention or fascination: Slowly and gradually changing natural elements such as sunsets, clouds, trees or water appear suitable (Kaplan, 1995). Water surfaces with high perceived water quality have been shown to be especially relaxing (White et al., 2010, Wilkie and Stavridou, 2013). Utilizing these elements as feedback stimuli can draw the involuntary attention even more towards those fascinating parts of the environment.

  • (3)

    Have a certain extent: The environment should open up enough space while being perceived as coherent. A large virtual environment can easily be created, whereas a coherent presentation is a greater challenge. Different objects, elements and feedback stimuli need to be sought-out and combined carefully as to create a feeling of coherence. Despite well-picked and -arranged visual details and feedback within the scene, a coherent and relaxing experience also requires an adequate and realistic soundscape (Annerstedt et al., 2013).

  • (4)

    Be compatible with the user's preferences and goals: In HRV-BF, the user's goal is to calm down and breathe deeply and rhythmically. Immersive virtual environments, on the contrary, bear the risk of being arousing (Cobb et al., 1999, Keshavarz et al., 2014, Makransky et al., 2019). Consequently, it is important to design the virtual environment to be as calming as possible through soothing and pleasant color scheme and lighting (Jerald, 2015; Naz et al., 2017), a stationary experience without vection-inducing locomotion (Davis et al., 2014, Keshavarz et al., 2014), as balanced surroundings that offer fascination and yet do not overwhelm the user. Furthermore, poorly designed virtual experiences might distract the user from the desired breathing technique. Task-irrelevant stimuli in learning contexts are known to distract from the task at hand (cf. seductive details effect; Rey, 2012), which also seems to hold true in virtual environments (Makransky et al., 2019). Therefore, all parts of the virtual environment need to serve as feedback elements, so that all perceivable stimuli are relevant to the task and attention is drawn to the regulation of the desired behavior.

The present feasibility study is designed as a proof-of-concept pilot study. To the best of our knowledge, this is the first empirical study to implement an HRV-BF protocol in an immersive virtual nature environment. It is meant to examine whether such an embedded implementation is a practicable way of performing HRV-BF in novice users. We compare the VR-based to a traditional HRV-BF implementation to test, (1) whether VR-based HRV-BF is feasible in that it can hold up to the well-established traditional HRV-BF and (2) whether it has advantages over traditional HRV-BF. We chose an exploratory approach, due to a lack of prior evidence in the field of VR and HRV-BF that would allow for formulating hypotheses. Besides the two groups that performed a single session of short duration HRV-BF either using VR (treatment group) or a traditional setting (treated control group), we included an untreated control group as a baseline.

Section snippets

Participants and design

Participants were recruited via social media as well as an online database. There was no monetary incentive for participation. In total, 68 healthy participants took part in the experiment. The sample consisted of 41 (60.3%) women and 27 (39.7%) men; 55 (80.9%) were students and 13 (19.1%) were part- or full-time employees. The average age was 22.9 years (SD = 4.0). All participants had experienced VR no more than two times before, with the majority (64.7%) having had no previous experience.

Results

To ensure that there were no between-condition differences at the outset of the experiment, we performed a range of between-subjects analyses of variances (ANOVA) for demographic variables as well as pre-manipulation measures of the dependent variables. As can be seen in Table 3, the three experimental conditions showed comparable mean-values and did not differ in any measure prior to the manipulation, neither in age, gender or work status (all p ≥ .072), nor in any dependent variable (all p

Discussion

The present study implemented a VR-based HRV-BF embedded in a virtual nature environment. We used a supposedly calming beach scenery as the virtual setting and compared this immersive HRV-BF to a traditional HRV-BF implementation with geometrical indicators as feedback stimuli as well as to a non-treated control condition. The study explored (1) whether a single-session VR-based HRV-BF implementation is feasible in that the participants succeed in performing a biofeedback session and (2)

Conclusion

Our study explored the feasibility of utilizing VR for HRV-BF. We designed a virtual environment capable of providing a quality biofeedback experience on the one hand, while improving some of the issues related to traditional HRV-BF on the other hand. Our study shows that, in novice users, a VR-based HRV-BF implementation is feasible in that it holds up with traditional HRV-BF. Furthermore, the study indicates advantages of a VR-based implementation, such as the capability to improve training

Funding source

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest

The authors declare no conflict of interest.

References (83)

  • A. Sutcliffe et al.

    Presence, memory and interaction in virtual environments

    Int. J. Hum. Comput. Stud.

    (2005)
  • M.P. Tarvainen et al.

    Kubios HRV - heart rate variability analysis software

    Comput. Methods Programs Biomed.

    (2014)
  • R.S. Ulrich et al.

    Stress recovery during exposure to natural and urban environments

    J. Environ. Psychol.

    (1991)
  • M. White et al.

    Blue space: the importance of water for preference, affect, and restorativeness ratings of natural and built scenes

    J. Environ. Psychol.

    (2010)
  • S. Wilkie et al.

    Influence of environmental preference and environment type congruence on judgments of restoration potential

    Urban For. Urban Green.

    (2013)
  • F. Yasuma et al.

    Respiratory sinus arrhythmia: why does the heartbeat synchronize with respiratory rhythm?

    Chest

    (2004)
  • A.P. Anderson et al.

    Relaxation with immersive natural scenes presented using virtual reality

    Aerosp. Med. Hum. Perform.

    (2017)
  • R. Berto

    The role of nature in coping with psycho-physiological stress: a literature review on restorativeness

    Behav. Sci.

    (2014)
  • C.J. Bohil et al.

    Virtual reality in neuroscience research and therapy

    Nat. Rev. Neurosci.

    (2011)
  • D.E. Bowler et al.

    A systematic review of evidence for the added benefits to health of exposure to natural environments

    BMC Public Health

    (2010)
  • F. Buttussi et al.

    Effects of different types of virtual reality display on presence and learning in a safety training scenario

    IEEE Trans. Vis. Comput. Graph.

    (2018)
  • S.V. Cobb et al.

    Virtual reality-induced symptoms and effects (VRISE)

    Presence Teleoper. Virtual Environ.

    (1999)
  • J.J. Cummings et al.

    How immersive is enough? A meta-analysis of the effect of immersive technology on user presence

    Media Psychol.

    (2016)
  • S. Davis et al.

    A systematic review of cybersickness

  • A. Dillon et al.

    Smartphone applications utilizing biofeedback can aid stress reduction

    Front. Psychol.

    (2016)
  • K. Druschky et al.

    Mobile biofeedback of heart rate variability in patients with diabetic polyneuropathy: a preliminary study

    Clin. Physiol. Funct. Imaging

    (2015)
  • J. Freeman et al.

    Relaxation Island: virtual, and really relaxing

  • R. Gevirtz

    The promise of heart rate variability biofeedback: evidence-based applications

    Biofeedback

    (2013)
  • A.M. Gillinov et al.

    Variable accuracy of commercially available wearable heart rate monitors

    J. Am. Coll. Cardiol.

    (2017)
  • V.C. Goessl et al.

    The effect of heart rate variability biofeedback training on stress and anxiety: a meta-analysis

    Psychol. Med.

    (2017)
  • D. Gromala et al.

    The virtual meditative walk: virtual reality therapy for chronic pain management

  • T. Hartig et al.

    Environmental influences on psychological restoration

    Scand. J. Psychol.

    (1996)
  • J. Hayano et al.

    Respiratory sinus arrhythmia. A phenomenon improving pulmonary gas exchange and circulatory efficiency

    Circulation

    (1996)
  • G. Henriques et al.

    Exploring the effectiveness of a computer-based heart rate variability biofeedback program in reducing anxiety in college students

    Appl. Psychophysiol. Biofeedback

    (2011)
  • H. Huang et al.

    Recent developments in biofeedback for neuromotor rehabilitation

    J. Neuroeng. Rehabil.

    (2006)
  • J. Jerald

    The VR Book: Human-Centered Design for Virtual Reality

    (2015)
  • D.A. Kaiser et al.

    Effect of neurofeedback on variables of attention in a large multi-center trial

    J. Neurother.

    (2000)
  • R. Kaplan et al.

    The Experience of Nature: A Psychological Perspective

    (1989)
  • B. Keshavarz et al.

    Combined effects of auditory and visual cues on the perception of vection

    Exp. Brain Res.

    (2014)
  • I. Kosunen et al.

    RelaWorld: neuroadaptive and immersive virtual reality meditation system

  • L. Lagos et al.

    Heart rate variability biofeedback as a strategy for dealing with competitive anxiety. A case study.

    Biofeedback

    (2008)
  • Cited by (62)

    • Virtual reality-supported biofeedback for stress management: Beneficial effects on heart rate variability and user experience

      2023, Computers in Human Behavior
      Citation Excerpt :

      Simplified and intuitive feedback might also reduce cognitive load (Sun, Cao, & Ma, 2017; Wollmann et al., 2016). Several studies found that experiencing BF in VR via HMD leads to greater attention and less mind-wandering than practising BF on a two-dimensional screen (Blum et al., 2019, 2020; Rockstroh et al., 2019). This may be attributed to the strong sense of presence induced by VR (Cummings & Bailenson, 2016; Rockstroh, Blum, & Göritz, 2020) and the elimination of external distractions through the use of an HMD.

    View all citing articles on Scopus
    1

    These authors contributed equally and therefore share first authorship.

    View full text