Validity of the Polar H10 for Continuous Measures of Heart Rate and Heart Rate Synchrony Analysis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents an evaluation of wearable Polar H10 ECG and heart rate sensor in typical psychology research scenario. Although tests of the device has already be mader and published (also by the same authors), in my opinion the manuscript is worth publishing mainly to broaden the range of tools for experimental psychology. 

I suggest several improvements to the manuscript below: 

1. in Abstract authors report 'strong agreement' and 'high correspondence' - please also provide correlation coefficients 
2. Content presented in Fig. 1 b-d appears much before it was referenced in main text, what hinders understanding. Please consider putting the content of Fig. 1.b after the video stimulus will be decribed and Figs. c, d after the comparison methods will be concerned. 
3. Replicating of prior validation is repeated at the end of Introduction, please rather highlight the novelty of the paper
4. In Section 2.3 please refer to gold standard affective picture stimuli such as IAPS or NAPS.
5. Page 5: ECG data points ? do you mean signal samples (which are equidistant) or R-peaks (the interwal of which is irregular)?
6. Please explain why the authors didn't use methods for direct comparison of non-uniform time series? they produce a direct correlation-like coefficient of likelihood
7. Section 2.4.2 Based on literature, PLV and WTC are obsolete, discontinued methods with narrow area of application. To make things worse I'm affaid they are hardly explainable to the psychologist who seeks for difference in emotional reactions among individuals. A distance of non-uniform time series would be much easier to understand.
8. A limitation of the reported studies compared to exercise tests may also be the much narrower range of heart rate values ​​(the authors do not provide the HR values ​​obtained in the studies). Information about possible skin sweating, which affects the quality of the signal recorded by the Polar H10, is also desirable. 
9. It's unclear whether the authors base their calculations on the HR value determined by the Polar H10 or the electrocardiogram recorded by the Polar H10. In the latter case, interpolation is possible, so the 130 Hz sampling rate is no longer a limitation. However, in this case, it's necessary to know what algorithm was used to detect QRS complexes in the ECG signal from the Polar H10 and what the time accuracy (jitter) of this algorithm is.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Although the main objective of the study was to validate the Polar H10 for use in 'natural' and real-world environments, validation was performed under highly controlled laboratory conditions.

1. Participants remained seated and immobile in a soundproof room. In real-life social interaction situations, which the study aims to facilitate, people gesticulate, walk around and talk. The study did not assess the effect of motion artefacts (common in wearable) on signal quality or synchrony measurement. The task involved watching a video, rather than active face-to-face interaction. Speech and breathing significantly alter heart rate variability (HRV) during conversation, which was not tested here.
2. Technical synchronization challenge (the 'clock' problem). This is arguably the most significant technical challenge for those seeking to replicate the study in practice. Dependence on reference for alignment: To validate synchronization, the authors aligned the Polar H10 data using event markers from the reference ECG system with wires. However, in a real-world scenario using only Polar straps, the researcher will not have access to the reference system for alignment. The Polar H10 uses Bluetooth and independent internal clocks that can drift. While the study showed that drift is minimal over 5 minutes, it does not guarantee that, over a longer period (e.g. 1 hour), the devices of two different people will remain perfectly synchronized without external synchronization hardware.
3. Sample issues and generalization. a) Although a sample size of 26 individuals is reasonable for intra-subject physiological measures, synchrony analysis depends on the pair. This means that the dyadic analysis was performed with only 13 pairs. This is a small sample size for making robust generalizations about social interactions. b) Participants were young (average age 25), healthy, educated and from a Western background. The study does not validate the device for use with the elderly, people with cardiac arrhythmias or people with a high body mass index (adipose tissue may interfere with electrical conduction to the chest strap).
4. Limitations of the experimental stimulus. a) The authors report that the 'positive' video increased subjective emotional valence, but did not generate significant changes in heart rate compared to the neutral video. This restricts the ability of the study to claim that the device can detect subtle changes in positive emotions, given that the stimulus itself failed to generate these fluctuations. b) The task involved watching a video (joint attention), rather than active face-to-face interaction. Speech and breathing significantly alter heart rate variability (HRV) during conversation, which was not tested here.
5. Temporal resolution, the sampling rate (130 Hz vs. 1000 Hz). The Polar H10 has a sampling rate of 130 Hz. For measuring average heart rate, this is excellent. However, to detect the exact peak of the R wave (which is necessary for accurately calculating variability and fine phase synchrony), this rate is considered low compared to the 1000 or 2000 Hz of medical equipment. While the authors argue that the impact was minimal, there is an inherent millisecond 'jitter' (tremor/temporal error) that can affect more complex or nonlinear synchrony analyses.

In summary, the study successfully demonstrates that the Polar H10's signal quality is comparable to that of a medical ECG. However, the practical applicability for synchronization is limited; the study did not solve the problem of precisely synchronizing multiple Polar devices without expensive laboratory equipment acting as a 'time judge'.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors, 

Thanks for careful consideration of my remarks, I recommend publishing the manuscript in the present form. 

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have successfully addressed the reviewer’s concerns by refining their claims. Rather than claiming to have 'solved' the problem of measuring real-world heart rate synchrony, they now state that they have validated a low-cost tool that will enable future research in this area.