Heart rate variability variance should highly be explained by respiration (up to 90 % depending on resp rate and tidal volume) during meditation. Especially, when there is low resp. rate (ca. 6/min)! If  this reveals to become equivalent to homeostasis - nobody knows. But anyhow it is an indication of cv & resp & autonomic centrally (medulla obl.) mediated coupling mechanisms in combination with peripheral reflex mechanisms (e.g. baroreflex).

Thank you M. Raschke for  your answer. I understand the general physiological mechanism, nevertheless the "strange" transition happened at the end of the exercise and the characteristics on this state remains somewhat unexplained.

As shown in the document, before entering the 4th phase the subject was in parasympathetic clear dominance with an average heart rate of 52 bpm evaluating (more or less) hectically. Respiration was spontaneous here, not controlled at 6cycles/min.   

So I'm wondering why suddenly the HR jumped from 52 bpm to a 57 bpm and becoming a sine wave type? Moreover, the ANS balance was near perfect and the subject perception was definitively very different than before.

It looks like the ANS loop would be able to settle itself, souddenly, to kind of "resonant" HR of (57 bpm here)?

I'm also wondering if having an integer ratio (10 here) between HR and Resp may have an explanation on this state and if the "trigger" of this transition migth be the one we explained in the document.   

I worked with HRV data during meditation available at internet. I have a method for estimating fractal dimension and curiously I found a "homeostatic " effect of meditation upon fractal dimension.

This is the abstract

Utilidad del índice ARDI (Auto Regressive Dimensional Index) para evaluar los cambios inducidos por la meditación yoga sobre la variabilidad de la frecuencia cardiaca

Eligio Kindelán Cira · Ana María Morales · José Luis Hernández Cáceres

[Hide abstract]

ABSTRACT: Changes induced by Yoga meditation upon the complexity of Heart Rate Variability were explored via the Auto-Regressive Dimensionality Index (ARDI). Data were downloaded from the "Physionet.org" website, and corresponded to HRV traces recorded before and during meditation with two Yoga techniques. Statistical support was found for the changes in ARDI being inversely dependent upon the basal ARDI levels before meditation (p

May I ask how you have isolated "sympatheic activity" and "parasympathetic activity" from heart rate? I am referring to the fact that LF is not anymore considered to be just sympathetic activity and LF/HF ratio does not seem to be a reliable measure of autonomic balance ( http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3576706/ )

Dear Anu, I fully agree with your point, we don't use LF/HF for ANS assessment.    

We have a real time, non linear mathematical method, to separate the sympathetic/parasympathetic activities of ANS from HRV.This opens great research capabilities on ANS/heart since we show the corresponding activities plots in real time.

We have incorporated this method in CODESNA_HRV software able to process recorded RR files (traditional LF/HF analysis are also supported).             

You might find useful to have look at these capabilities on the link below.

The attached picture shows an application of our method (real time ANS monitoring) during tilt test on elite athletes.  

http://www.codesna.com/en/products/codesna_hrv/

Thanks for the link. I tried to study how the program assesses "Orthosympatheticactivity(in msec2 and %)" and "Parasympatheticactivity(in msec2 and %)", but did not find the answer. I assume the program calculates inspiration and exhalation phases during the 2 minutes of controlled breathing, and compares that to the actual recording? (inspiration = orthosympathetic, exhalation = parasympathetic?). It's hard to assess the method when the exact formula of calculations is not revealed (and one does not know French to read the validation study!).

Anu, a clarification is needed here: CODESNA_HRV is a standalone program which takes on the input any recorded RR file, nothing else. It can be installed and used with any existing RR files from the link I sent previously. So that the ANS dynamics can be seen all along the measurement protocol, no matter how it was defined or its length in time. Regarding mathematical details, I can send you a link pointing our published patent is you're interested in. Publications based on CODESNA_HRV will go out shortly.  

Regarding the 2 min respiratory protocol you're pointing out, this is a different product and protocol and it is used to asses chronic ANS imbalances (like chronic stress). This is a different subject from the initial question on this post but of course I'm available to discuss about it if you're interested.

Although I just got back after a week of training, and do not want to pretend I have looked deeply into the details of this discussion, I can make a general comment.  Both SNS and PNS inputs control the activity of the SA node (and that of the cardiovascular system).  The only reason RSA is a marker for parasympathetic activity is that the SA node's properties can change virtually instantaneously as the changes in PNS activity also change the amount of acetylcholine bound to it.  This is true in the normal respiratory (HF) range.  The SNS nerve traffic is also changing with respiration, but it takes too long to affect the properties of the SA node for it to track respiration.  Once you get below, probably, 10 breaths/min, the whole situation changes as there is time between breaths for the SNS to change the properties of the SA node on the next breath.  Thus, once respiratory rates go below this, as in meditation, there is a huge increase in  RSA that is not only parasympathetic in origin.  Further complicating the picture is the distinction between the ventral and dorsal components of the PNS.  The ventral component acts quickly and is responsible for both immediate activation AND deactivation.  SNS activity comes on more slowly.  But the dorsal vagus is responsible for the extreme threat situation and acts to bring the body into freeze.  Even though you do not expect the meditators to go into this state, it can happen, because some innocuous situation is overcoupled with an involuntary threat response from a past trauma and might show up as a result of mindfulness meditation.  In fact, in some of the meditation trainings now, the leaders are being trained in dealing with trauma because they really do not want to be taking their students to the emergency room.. 

As for the specific non-linear metric, I do not know what evidence there is for it being a measure of SNS:PNS interactions but it certainly would be affected by the stability of breathing which might increase with practice.  I am a fan of non-linear measures in population studies because they have a strong relationship to outcomes than do the conventional ones, BUT they are also much more sensitive to scanning issues.

Dear Phyllis, thank you for the detailed response. Although emotional and physiological mechanisms of ANS polyvagal theory seem to be well understood, I have some difficulties to interpret it on this particular meditation exercise and results.

Kindly have a deep look at the document I attached at the beginning of the discussion thread. The first 3 phases of ANS are expected (1-Sympathetic dominance, 2-little hectically balanced, 3-Parasympathetic dominance), the 4th stage is curious.You mentioned that this 4th state might be similar to a response to an extreme threat situation, and here I'm a bit surprised. I understand that this primitive reaction can occur during an extreme threat and the dorsal vagus will bring the body to freeze. I guess that this state is not really comfortable and it is not intended to take long time, right? It's supposed to remain "frozen" until the extreme danger is over. But in the 4th stage of this meditation, the perception of the subject was really far away from an extreme fear reaction. He reported a very comfortable wellness feeling going along with measured sinusal balance of the ANS as seen on the graphs. Pleasant awareness was also reported.     

Another important point: during the first 25 minutes of the exercise, the subject was breathing (controlled) at 6cycles/min but ANS was sympathetically dominant. The same pace of 6cycles/min was observed during the 4th phase but here the breath was spontaneous and ANS reaction totally different (see the graphs).

You also suggest that such reaction (4th phase) during meditation could be linked to a past trauma; does it imply that this can occur only for people having already experienced such traumatic "body freeze" situation?  

I am relatively new to this field, and was recently doing a self-experiment using Polar H10 and Kubios recording a 1h mediation session and observe simiar thing. I was using in parallel brainwave stimulation. I also had a surprising low bpm compared to my typical state - which "jumped" at one point - actually it was doing a siteration of repeated steps some 15 min separated from each other ....  which might correlate with the stimulated brain waves - I will test that more over the next weeks. So a subjective experience miight be a first explanation to your observation: The practitioner jumped to another "vibrational level" - eventually it could make sense to record also EEG - and observe a shift from alpha to gamma and/or theta band ?

I don't have the means to do EEG measurements in parallel, but maybe that might be interesting to do combined with HRV.

I would like to echo Anu's comments that it is a bit of a mystery how you quantify sympathovagal balance from HRV given that LF is no longer considered a marker of sympathetic activation.