Heart Rate Variability (HRV) Biofeedback and Athletic Performance
Science and practical application.
Recently, two new studies highlighted the effectiveness of Heart Rate Variability (HRV) biofeedback, and in particular resonant frequency breathing, in increasing acutely HRV. In the words of Sylvain Laborde, one of the leading experts in the field, “Slow-paced breathing is the most acceptable, scalable, and cost-effective intervention for enhancing HRV” (paper here). Additionally, the recent paper of Joshua Marchant et. al. (including Inna Khazan among the co-authors) shows that not only does resonant frequency breathing increase (acutely) HRV more than other breathing protocols but that slightly changing the inhale-exhale ratio, leads to an extra increase.
Coincidentally (or not), resonant frequency breathing with a 4:6 inhale-exhale ratio is the protocol I use and recommend for our HRV4Biofeedback app, and therefore I’d like to elaborate a bit further on this work.
In this blog, I provide an overview of HRV Biofeedback, including the underlying physiological principles, protocols, and expected physiological, psychological, and performance outcomes.
The Physiology of HRV Biofeedback
HRV Biofeedback is a technique that consists of providing an individual with real-time feedback on instantaneous heart rate and respiration changes while being instructed to breathe at low frequencies (Lehrer and Gevirtz, 2014).
Breathing at low frequencies (or slow breathing) causes large oscillations in the instantaneous heart rate, which synchronizes with the breathing rate. The influence of breathing on heart rate is called Respiratory Sinus Arrhythmia (RSA) and is mostly modulated by the parasympathetic branch of the nervous system (Lehrer and Gevirtz, 2014). Hence, we can see slow breathing as a form of training of the parasympathetic system, which might explain at least part of the positive effects of HRV Biofeedback reported in the literature in the context of reducing stress and anxiety (Goessl, Curtiss, and Hofmann, 2017). Strengthening the parasympathetic nervous system could also motivate using HRV Biofeedback in athletes, with the potential to improve emotional self-regulation, coping mechanisms, and performance (Khazan, 2016; Pusenjak et al., 2015).
Facing stress
Stressors associated with life events and intense physical training can result in negative physiological responses such as stress hormone perturbation, immunosuppression, and impaired skeletal muscle repair (Williams and Andersen, 2007; Appaneal and Perna, 2014). All of these aspects can act as mediators for negative outcomes such as increased injury risk or training maladaptation, in both cases resulting in reduced health and performance (Prinsloo, Rauch, and Derman, 2014).
Upon facing a stressor we have an activation of the sympathetic nervous system which is directly innervating most organs. Secondly, we have hormonal responses through the hypothalamic-pituitary-adrenal axis which results in cortisol release. Depending on an individual’s ability to cope with a stressor, these responses can last longer and have a stronger negative effect on an individual’s physiology.
This is especially true when combined with high-intensity and high-volume training typical of elite sports (Clow and Huckle-Bridge, 2004). For example, in Perna and McDowell (1995) the authors showed how athletes that reported being more stressed, had a long-lasting negative response including increased cortisol levels for several hours after exercise, with respect to athletes that did not report high levels of life stress.
Given the physiological and psychological factors just discussed, HRV Biofeedback could be an ideal strategy to help us self-regulate and better cope with stressful situations. HRV Biofeedback can directly affect autonomic activity through slow breathing exercises that stimulate the parasympathetic system.
Therefore, HRV Biofeedback might directly provide a positive impact on the physiological and psychological factors that mediate health and performance.
The Resonant Frequency
As previously introduced, HRV Biofeedback requires the individual to breathe at low frequencies. Experimental studies have found the highest amplitudes in instantaneous heart rate oscillations when breathing at approximately 0.1 Hz. The frequencies at which amplitude is maximal is often called the resonant frequency and can vary by 0.5–1 breath/minute between individuals. The resonant frequency of an individual can be established with a protocol that consists of breathing at different frequencies for a few minutes until the frequency that elicits the maximal amplitude is found (Lehrer et al., 2003). At the resonant frequency, heart rate and breathing rate are perfectly synchronized. Additionally, the resonant frequency results in a 180-degree phase relationship between heart rate and blood pressure, suggesting that the high variations in the instantaneous heart rate are due to the baroreflex. Further research into these mechanisms has shown how biofeedback practice could indeed increase baroreflex gain, which might be a causal pathway explaining why hypertensive disorders can improve using HRV Biofeedback (Vaschillo et al., 2002).
Other pathways have also been proposed to explain the relationship between HRV Biofeedback and positive outcomes. First, we have a potential pathway between the baroreflex and neural control, in particular the amygdala, which could explain why improvements are seen in patients with anxiety and depression when using biofeedback interventions (Lehrer et al., 2003; Karavidas et al., 2007). Secondly, another pathway could involve a strengthening of the parasympathetic nervous system, as shown using electrical vagal stimulation in the context of treating depression (Brown, Gerbarg, and Muench, 2013), which might also occur during HRV Biofeedback.
Acute and baseline changes in HRV
During an HRV Biofeedback session, large changes in HRV are elicited as a result of the session itself, mainly because respiratory sinus arrhythmia is maximized while breathing at a low frequency.
This is not the same as looking at baseline changes. HRV analysis can be used for various applications. For example, what we do at HRV4Training is to quantify baseline physiological stress (what we could call "chronic" stress), and how this changes in response to training and lifestyle over periods of weeks or longer. To quantify baseline physiological stress, our measurements need to be taken at a precise moment, which is first thing in the morning so that we can avoid the effect of confounding factors. By capturing changes in resting physiology, we can provide useful feedback that helps individuals make meaningful adjustments to better balance training and lifestyle. This is particularly relevant as we all respond differently even to the same stressors depending on various aspects (how novel is the stressor, how much of that stressor we are used to taking, what other stressors are present).
While our morning measurements should be done while resting and breathing naturally, during biofeedback we use slow breathing to elicit higher parasympathetic activity. You can see your biofeedback session the same way you see your other training sessions, this is something you do so that in the longer term, there can be beneficial changes in health and performance. Biofeedback is just a positive stressor. However, this also means that the acute change in HRV during biofeedback does not necessarily mean that there will be any change in our baseline (resting) HRV.
Note that this does not mean that the acute change is not important. On the contrary, the large oscillations due to breathing at the resonant frequency are most likely what causes the positive outcomes reported by countless studies. However, it should be made clear that baseline physiology might still be unaffected and needs to be measured differently, outside of the biofeedback session.
Normally I would recommend doing biofeedback exercises as an add on the regular morning measurement done with HRV4Training. Combining biofeedback with morning measurements taken with HRV4Training, you could see potential changes in baseline chronic physiological stress as measured in a known context (first thing in the morning), as a result of your biofeedback sessions.
Expected physiological, psychological, and performance outcomes
What does the research say?
Most early studies reported positive outcomes in terms of psychological measures, and mixed findings in terms of physiological measures and performance improvements (Tanis, 2008). However, none of these studies included a control group, and some of the reported improvements in performance were either anecdotal (Beauchamp, Harvey, and Beauchamp, 2012) or involved a single athlete (Lagos et al., 2008).
In recent years, more and more studies have started including control groups to better account for the effect of HRV Biofeedback independently of other changes that might be occurring as training progresses.
Finally, in very recent studies, a variety of different approaches and methods have been employed in order to make HRV Biofeedback more practical, for example using mobile apps and mostly home-based practice as well as sessions as short as 3 minutes. Let’s look at some of the results.
Performance outcomes
Results in terms of performance outcomes have been conflicting so far. Part of the issue has to do with the difficulties of measuring performance. While for certain sports, in particular individual endurance sports, performance testing is often done routinely, this is not the case for team settings where performance testing is hardly performed or standardized. Additionally, when performance testing is carried out with a test that aims at simulating certain skills required during a game, it is debatable how such tests relate to in-game performance (Den Hartigh et al., 2018; Bergkamp et al., 2019). This being said, in elite settings where a control group was lacking (e.g. Canadian Olympic athletes), anecdotal evidence has been positive, but it is difficult to determine if HRV Biofeedback was particularly beneficial in the athlete’s quest for a gold medal. Other studies that have shown improvements also relied on subjective assessment of athletic performance, or on metrics that might have little to do with performance, for example time spent training in an elite triathlete (Mueller et al., 2019). Similar criticisms can be made for studies that showed no improvements (Tanis, 2008; Rijken et al., 2016).
From the performance outcomes reported in HRV Biofeedback interventions so far, it seems difficult to determine if there is any direct positive effect of HRV Biofeedback on performance. While some of the aspects just covered can be better accounted for (e.g. including an active control group, using standard performance tests or preferably in-game performance, as well as using standard protocols for interventions), in certain sports it can still be very challenging to effectively determine performance changes following an intervention, for example in team sports or sports with a strong tactical component (Cannon-Bowers and Salas, 1997; Richard et al., 1999; Wiseman et al., 2014; Den Hartigh et al., 2018; Bergkamp et al., 2019).
For these reasons, and considering the important physiological and psychological changes that can be impacted by HRV Biofeedback, and how in turn physiological and psychological changes can mediate performance, it can be particularly interesting to determine if there are more consistent findings when it comes to such measures.
Psychological and physiological outcomes
Psychological measures following HRV Biofeedback interventions are probably the most consistent in terms of positive outcomes. In particular, the various studies investigating effects on anxiety (both trait and state) as well as on self-esteem and self-efficacy, often found improvements (Lagos et al., 2008; Paul, Garg, and Sandhu, 2012; Paul and Garg, 2012; Dziembowska et al., 2016) with only one study reporting no changes in a small sample of younger athletes (Perry, 2018).
In general, psychological parameters seem to benefit from HRV Biofeedback interventions, especially for what concerns anxiety, similarly to what has been reported in literature outside of sports.
In terms of physiological measures, results are also quite consistent across studies. However, an important caveat here needs to be considered. In particular, I have stressed before how physiology can be measured at different times, and in the case of HRV Biofeedback, how measurements taken during the session are reflective of acute changes as a result of slow breathing and respiratory sinus arrhythmia, and how these changes are not necessarily linked to baseline physiological changes in parasympathetic activity. These are important aspects to consider as results differ greatly depending on when HRV was measured, as we can see from the few studies that measured both before and during HRV Biofeedback, and often found no improvement in HRV before the session (a resting measurement), but increases during and right after each session (Tanis, 2008). Only one study has measured changes in resting HRV from data collected first thing in the morning, showing an increase over time. However, this was a case study with only one participant, no control, and the change in HRV was associated with a change in training load, hence it is not possible to determine causality.
While these acute changes are unrelated to baseline effects, they are still very important. In particular, consistent results from a variety of studies show that breathing at resonant frequency synchronizes breathing and heart rate in a way that is quantifiable and can have beneficial effects. I have introduced earlier the various pathways that try to explain the relationship between HRV Biofeedback and positive changes in physiological and psychological variables, and we have seen how certain psychological aspects are consistently improved following HRV Biofeedback interventions, for example, anxiety. Reduced anxiety could be a beneficial effect of HRV Biofeedback due to slow breathing and changes in baroreflexes regardless of baseline changes in HRV.
Yet, the relationship between acute changes in HRV, baseline changes in HRV, and psychological measures following an intervention is complex and requires further investigation.
Considerations for Practitioners
To date, results in terms of the effectiveness of HRV Biofeedback interventions for enhancing athletic performance have been inconsistent. However, for practitioners working with athletes, HRV Biofeedback approaches can be beneficial from various points of view. Most studies were able for example to show reduced anxiety and increased self-efficacy and self-esteem, following the intervention. Additionally, measuring performance under realistic conditions can be a complex task in most sports. As a result, most studies evaluated performance using simple tasks that might not reflect real-life scenarios. For example, a passing task in soccer during practice might have little to do with the athlete’s ability to deal with a stressful situation during an important game, a situation that might be positively affected by an HRV Biofeedback intervention.
For these reasons, despite the limited evidence supporting HRV Biofeedback interventions in terms of positive effects on performance, my recommendation would be to evaluate the possible usefulness of HRV Biofeedback techniques on a case-by-case basis. Depending on an athlete’s sport (precision, highly skilled, etc.), psychological traits (anxiety, tendency to choke under pressure, etc.), and limiters, HRV Biofeedback could be a useful tool to explore.
Practical Guide
Alright, you’ve got up to this point and are still interested in trying HRV biofeedback. Let’s get a bit more practical.
HRV Biofeedback has been first formally defined by Lehrer, Vaschillo, and Vaschillo (2000). The authors proposed a protocol consisting of a 10−session program. During the first session, the resonant frequency of an individual needs to be established. The resonant frequency is typically around 6 breaths/minute but can change by 0.1–1.5 breaths/minute between individuals. Hence, each participant should try a range of frequencies and the practitioner should determine the optimal frequency by analyzing the power spectrum of each different test.
In HRV4Biofeedback, we have introduced an automated resonant frequency protocol that works similarly and guides you through a series of breathing frequencies to determine which one is optimal for you, as described here.
At this point, the participant is instructed to practice at home for two 20-minute periods each day. The participant can be instructed about abdominal breathing and pursed lips breathing and is asked to practice for two 20-minute sessions per day.
Over the years, and considering the demands of modern-day life, small variations of the original protocol that require less time from the athlete (from 3 to 20 minutes instead of two sessions of 20 minutes per day) were often considered by researchers and practitioners working in applied settings (Rijken et al., 2016; Perry, 2018; Deschodt-Arsac et al., 2018; Mueller et al., 2019).
If you use HRV4Biofeedback, my recommendation is to proceed as follows:
Use the resonant frequency test in the app to find the optimal breathing frequency for you.
Set the inhale-exhale ratio to 4:6, in the app Settings.
Start practicing 5-10 minutes per day, and see if you can over time increase the amount of practice to meet the recommended protocols specified in the scientific literature (20-40 minutes).
If you feel more comfortable breathing with a 50:50 ratio, i.e. with the same inhale and exhale duration, that is fine too.
On a side note, I am working on linking up HRV4Biofeedback with the new HRV4Training Pro, so that you can look at your biofeedback sessions together with your resting physiology. No promises, but I thought that could be useful and easier to track this way.
Wrap up
Combining insights from biopsychosocial models and basic physiology, we can see how HRV Biofeedback has been proposed as a technique that can help athletes improve emotional self-regulation and coping mechanisms via a strengthening of homeostasis, with the potential of resulting in better health and performance.
In particular, apart from the potential for direct improvements in health and performance, other pathways could be positively impacted, from a psychological (e.g. anxiety) and physiological (e.g. hormonal response, strengthening of the parasympathetic system) point of view.
Hopefully, the availability of tools such as HRV4Biofeedback will make it easier for scientists, athletes, and coaches to further investigate these aspects.
I hope this was informative, and thank you for reading.
See also:
Marco holds a PhD cum laude in applied machine learning, a M.Sc. cum laude in computer science engineering, and a M.Sc. cum laude in human movement sciences and high-performance coaching.
He has published more than 50 papers and patents at the intersection between physiology, health, technology, and human performance.
He is co-founder of HRV4Training, advisor at Oura, guest lecturer at VU Amsterdam, and editor for IEEE Pervasive Computing Magazine. He loves running.
Social:
