The Effects of Music on HRV Biofeedback Training
A self-study, and initial considerations on sound design for resonance breathing
[TLDR: I designed soundscapes specifically for resonance breathing, and tested them on myself, showing on average a 25% increase in HRV compared to silence. I have also released the soundscapes for a range of different breathing frequencies on all major music platforms.]
Sounds and music can have a profound impact on us.
This is probably clear to everyone on an emotional level. What might be less obvious is that sound can also trigger a physiological response, either indirectly through the mind/body connection or — potentially — through some more direct physical processes.
To some extent, I’m sure everyone has felt these effects.
Just think of the physical discomfort and stress you feel when exposed to some loud and unpleasant noise. Or the goosebumps and energy boost when your favourite track comes on.
With Yudemon, one of my key goals is to explore and use this effect of auditory stimuli to enhance human happiness, performance, and overall wellbeing. And, at least for the moment, I am specifically focusing on its application to resonance breathing and HRV biofeedback training.
In general, I see three potential pathways through which sound and music could enhance this practice:
- Through physiological responses, which directly enhance the response of the autonomic nervous system (ANS)
- Through emotional responses, where the sound guides the mind to indirectly enhance resonance
- Through increased incentive and attention, making the practice more fun, engaging and novel, and improving how often and consistently people practice
I’m convinced that each of these directions contains viable and possible paths, but some will certainly be easier to discover and study than others. How well they generalise across different people may also vary considerably, especially with the emotional responses.
But even if the first and second pathways, the physiological and emotional responses, remain difficult to pin down for a while, I am sure that the increased engagement provided by music can prove immensely valuable for many practitioners, leading to more frequent and longer sessions and less drop out.
There are many different ideas spread across the three paths above that I want to explore with Yudemon. In this article, I will share with you my very first results of using soundscapes that were specifically designed for HRV biofeedback training.
I will share the actual results with you first. After that, in the second half of this piece, I will explain my reasoning behind the sound design.
The Result: Music Can Increase HRV
In my previous two self-studies I determined that my resonance frequency, the breathing rate that maximises my HRV response, is around 5.4 breaths per minute, and that my optimal breathing cycle is made up of about 0.35 inhale and 0.65 exhale.
Given this knowledge, I fixed my breathing rate at 5.4 breaths per minute and the inhale/exhale ratio at 0.375/0.625 for this experiment. The latter choice, which is close to but not exactly the value I had determine, was made for musical reasons since it perfectly splits an 8 bar segment into 3 bars for the inhale and 5 bars for the exhale.
Given these values, I designed a musical soundcape based on some assumptions/guesses of what might improve the resonance response (more on this later).
I further split this into two separate components, which I called Resonant and Temporal, to have three distinct musical soundscapes to test (the combined one, as well as each component individually).
In addition, I also tested against the benchmark of silence, as well as a noise soundscape I created to test the opposite response (which is almost comically unpleasant to listen to for any extended period of time).
You can listen to excerpts of the three musical soundscapes and the noise below (headphones recommended).
With these soundscapes and a modified version of the Yudemon HRV Biofeedback app, I then tested my response to each of them across ten individual sessions, where in each session I would listen to each of the five soundscapes for 3 minutes in a randomly chosen order.
Below the results of combining all the data from the 2.5 hours of resonance breathing.
The large points (and dashed lines) show the mean HRV for each soundscape, and the box plots show the spread of the data.
Impressively, the full musical soundscape performs best, with a mean HRV that’s over 20% higher compared to silence.
The noise, unsurprisingly, performed the worst.
As a side note: It would however be interesting to explore if this might not actually provide the best (or at least a different kind of) training effect and long term progress, since it trains the ANS response given an external stressor. But that’s a question to explore another time (and might require a larger trial).
Looking at the relative HRV, normalized by the maximum HRV of each individual session, shows even more clearly the seemingly strong effect of the full musical soundscape.
In all but two sessions, the full soundscape led to the highest HRV.
The resonant soundscape had the biggest variance, both in terms of data and my subjective response to it.
Both the temporal and full soundscapes, which show the highest HRV response, contain audio cues that tell me when each cycle starts and ends and when to switch from inhale to exhale. As a result, for those two soundscapes I practiced with my eyes closed, not looking at the visual timer.
Also subjectively, I felt like these temporal audio cues helped me time my breathing much better and naturally led to deeper and more consistent inhales and exhales.
Overall, these initial results certainly look very promising.
I can’t, however, get around mentioning one HUGE caveat: This self-study is highly biased.
While I tried to be as objective as possible and don’t consciously change my breathing in any way, this is exactly the result I was hoping for. The mental effect of that on my HRV can absolutely not be ignored. And this is only one of the many mental/emotional biases that might have crept into this experiment with me as its designer.
I would love to see this experiment repeated by other people. In fact, I will be releasing an album next week with a unique version of the full soundscape for 16 breathing frequency from 5 to 6.5 breaths per minute, and I encourage you to try it for yourself.
The Sound Design
As mentioned above, the sound design of the different soundscapes was based on certain guesses and assumptions I had made.
My first idea was to consider the breathing rate as a musical pitch, a fundamental frequency of my personal soundscape.
In my case, 5.4 breaths per minute, which is equivalent to 0.09 Hz, corresponds to a musical note of a slightly detuned F♯ (specifically, detuned by -6 cents, where 100 cents corresponds to one semitone).
This note is so low that it’s way below human hearing, but I can hear the F♯ at 0.09*(2⁹) Hz = 0.09 * 512 Hz = 46.08Hz, 9 octaves above my resonance frequency, as well as all the higher octaves.
So my (admittedly very ambitious and tentative) assumption was that maybe composing pieces in this key of slightly detuned F♯ might be able to trigger some genuine physical resonance that might increase the response of my ANS.
It certainly triggers resonances at these frequencies within my ear’s cochlea, which then get passed on as electrical signals to my brain. Whether this also causes any resonances elsewhere, at these or especially at the much lower fundamental frequency, is an entirely different question though.
In addition to musical pitches, I can also use the lower octaves and harmonics to modulate sounds — like panning a sound from left to right speaker or adjusting a filter cutoff — at an integer multiple (or fraction) of my resonance frequency.
Even if the above guesses prove to be completely wrong, which they very well might, it felt like a good creative constraint and starting point for musical exploration.
What I called the Resonant soundscape above is made up of mostly low and held notes of F♯-6ct at various octaves, as well as some modulation and arpeggiation at power-of-two multiples of my resonance frequency.
On top of that I layered the Temporal soundscape, which contains more clear time cues related to inhale and exhale.
First of all, there are two ride cymbals at a higher and lower pitch which signal the beginning of the inhale and exhale respectively.
In addition to these clear time markers there is a melody (mostly) in F♯ minor, which is ascending in pitch and faster during the inhale, and descending in pitch and slower during the exhale — both to mimic (and maybe enhance) the speeding up and slowing down of the heart rate in these two phases.
Layered together, the two give the full soundscape which has both the resonant drones as well as the clear time cues of melody and cymbals.
These were the assumptions and guesstimates that went into this very first soundscape I designed for resonance breathing.
The results, albeit highly biased, look very promising, whatever the actual mechanisms by which they enhance HRV might be (and again, at this stage I can’t rule out they aren’t purely emotional and specific to me).
I will certainly continue exploring this direction, as well as dive much deeper into the existing research to see if there are any more established and studied auditory triggers that I can use.
Besides exploring new soundscapes, I will now also start experimenting with making this and other soundscapes reactive, modulating them on the fly based on real-time biodata.
Finally, as already mentioned above, I will be releasing this soundscape, tuned specifically to a range of different resonance frequencies, next week on all major streaming platforms. To hear when that launches, sign up to my mailing list on the Yudemon website.
[Update: They are live now. Search for “Yudemon” on your favourite music platform, or have a look at this link for a few more details.]
I would really love to hear your feedback on this. If you know your resonance frequency, pick that one and practice with it. Otherwise start at 6 breaths per minute and adjust from there depending on what feels right.
Let me know how it feels, or if you have the means to measure, even share some of your data.