Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This research is grounded in a robust scientific approach. Its conceptualization displays novelty and originality. Understanding the associations between levels of shyness and body fat with heart rate reactivity during mental challenges has significant implications in both psychology and sports training practice. The methodologies are well-articulated, and the results are presented clearly, with thorough discussions included. This paper holds substantial value for the knowledge economy of youth sports and can be replicated by future studies by incorporating modified or additional factors.

Author Response

The response is attached as a separate .pdf file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript addresses an interesting topic—the interaction of shyness and adiposity on HR reactivity in adolescent athletes—but suffers from methodological, statistical, and theoretical weaknesses that undermine the validity, generalizability, and interpretability of its conclusions.

The entire analysis is based on 20 participants drawn from a single canoe club (15 boys and 5 girls). Interaction effects (Shyness × Body Fat) require substantially larger samples because interaction terms dramatically inflate required power. No valid interaction inference can be made with N=20.

Lasso regression can be used when the number of predictors is large relative to the sample size because it can shrink some coefficients to zero. However, Lasso does NOT eliminate the need for sufficient sample size. For Lasso to work reliably, preferred minimum numbers of samples are 30 to 50 × predictors. In this study Lasso regression with five-fold cross-validation is methodologically inappropriate for N=20—each fold contains only 4 participants, making model instability almost guaranteed despite the authors’ claim of “AI-assisted stability.” Bootstrapping with such a small N does not remedy the fundamental power problem.

HR recovery is not a baseline, nor is it a standard comparator in psychophysiology. Nearly all HR-reactivity research uses HR(Task) – HR(Baseline) or HR(Task) relative change, not Task minus Recovery. The justification that the baseline is influenced by “anticipatory arousal” is insufficient; nearly all psychophysiology studies manage this by including anticipation as a covariate, using multiple baseline epochs, using HRV metrics to normalize, and/or statistical adjustment (ANCOVA, multilevel modeling).

The manuscript repeatedly emphasizes AI support by writing “AI-assisted modeling”, “AI-guided cross-validation”, and “AI-supported analysis facilitated detection of subtle patterns.” However no algorithm is named, no model parameters, hyperparameters, or software details are provided, and no justification of why AI is relevant or necessary for a simple Lasso regression.

The conclusion claims that a “dual-risk profile”, “psychophysiological vulnerability”, and “cardiovascular stress responses with performance implications.” But, performance was not impaired, and no performance metrics correlated with HR, no longitudinal outcomes are measured, and HR reactivity alone does not imply vulnerability without clinical or functional impairment.

Author Response

The response(s) is attached as a separate .pdf file.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This research manuscript explored whether discussing shyness and body fat percentage jointly predicted individuals' heart rate reactivity during a psychomotor response during a challenge task.. However, several methodological and reporting issues still need to be addressed.

In the introduction, the authors define the functional significance of heart rate reactivity (HR reactivity) in a slightly ambiguous manner. On the one hand, studies cited in the text suggest that higher HR reactivity may be associated with better cognitive performance and faster reaction times. On the other hand, it is emphasized that it may reflect greater stress sensitivity or increased health risks. These two perspectives may point in different theoretical directions, and the authors are recommended to specify the main perspective from which HR reactivity is understood in this study.

Much of the literature on shyness cited in the introduction is based on social or socially evaluative situations (e.g., speeches, peer pressure observations). In contrast, the stressful task used in this study was a non-social reaction-time task. The theoretical link between the two has not yet been fully articulated.

The Participants section currently provides only basic demographic and training background information, but lacks descriptions of inclusion/exclusion criteria and pre-existing behavioral and health status controls. The text does not explain whether individuals with cardiovascular or neuropsychiatric disorders were excluded, or whether those who were using relevant medications that could affect heart rate or mood were excluded. In addition, the Methods section does not indicate whether there were uniform requirements for vigorous exercise, caffeine or energy drinks, nicotine, and other intake behaviors before the experiment, or whether testing was conducted within a uniform or relatively fixed time window with respect to eating. This lack of information prevents the reader from adequately assessing the effects of potential medical and behavioral confounders on heart rate and body composition indices.

The section on body composition measurement states that bioelectrical impedance analysis was performed using the InBody 720, but the conditions and procedures for the measurement were not sufficiently detailed. The article does not specify whether the measurements were scheduled at a consistent time (e.g., a specific time in the morning), whether the subjects were fasting during the measurement, or if they were instructed to avoid strenuous exercise or eating a large meal for a certain period beforehand.

The description of the statistical analysis and the so-called AI-assisted methods is still incomplete and lacks details on several key techniques. Although the authors mention using paired t-tests, bootstrapped Spearman correlation, Lasso regression, repeated cross-validation, and AI-supported data analyses, the text does not specify the statistical software and related packages used. It also does not clarify whether the independent variables were standardized or centered in the Lasso modeling process or how the interaction terms were constructed.

Author Response

The response(s) is attached as a separate .pdf file.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed the comments, and the manuscript has been improved for its possible publication. Thank the authors for their efforts on this manuscript.