pith. sign in

arxiv: 2605.25933 · v1 · pith:A6W6Z5TFnew · submitted 2026-05-25 · 💻 cs.LG · cs.AI

Quantitative Evaluation of the Severity of Posttraumatic Stress Disorder through Transfer Learning from Specific Phobia Data

Nicolas Ricka , Gauthier Pellegrin , Denis A. Fompeyrine , Thomas Rohaly , Leah Enders , Heather Roy This is my paper

Pith reviewed 2026-06-29 22:25 UTC · model grok-4.3

classification 💻 cs.LG cs.AI
keywords PTSDtransfer learningphysiological signalsmachine learningfear responseheart rategalvanic skin response
0
0 comments X

The pith

Machine learning transfers arachnophobia fear responses to classify PTSD severity at 86 percent accuracy

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper seeks to create an objective measure of PTSD severity by applying machine learning to physiological signals. It trains a multivariate kernel density estimation model on data from individuals with arachnophobia to learn fear response patterns. This model is then applied to heart rate and galvanic skin response data from 21 participants undergoing a military simulation. The transferred model classifies participants as having or not having PTSD with 86 percent accuracy and estimates the PCL-M severity score with a mean absolute error of 5.6. Such an approach could reduce reliance on time-consuming and potentially biased subjective evaluations.

Core claim

Training a fear-response model on arachnophobia data allows extraction of PTSD predictive features from fear-response curves in military simulation data, yielding 86 percent accuracy for PTSD classification and a mean absolute error of 5.6 for severity estimation on the PCL-M scale.

What carries the argument

multivariate kernel density estimation (MKDE) fear-response model

Load-bearing premise

Fear responses measured in arachnophobia are structured similarly to those triggered by PTSD-related stimuli in the military simulation.

What would settle it

If applying the model to the military dataset resulted in classification accuracy close to 50 percent, this would show that the transfer did not capture relevant predictive features.

Figures

Figures reproduced from arXiv: 2605.25933 by Denis A. Fompeyrine, Gauthier Pellegrin, Heather Roy, Leah Enders, Nicolas Ricka, Thomas Rohaly.

Figure 1
Figure 1. Figure 1: Overview of the proposed algorithm, SPIDERP. The model reads sensor measurement time series, and outputs a probability curve for the different possible outcomes. The data flow is divided into 4 steps: physiology extraction, estimation of the fear response, computation of static fear response features, and the multivariate kernel density estimator (MKDE) providing the final prediction. Materials and methods… view at source ↗
Figure 2
Figure 2. Figure 2: The median fear reaction curves for the PTSD (red) and non-PTSD (blue) subjects. We observe that the fear reactions of the non-PTSD patients decrease over time, whereas the PTSD patients have low initial fear reactions that drastically increase during the experiment. The time is in arbitrary units between 0 and 1, and the fear reaction is the output of the FR model, which is between 0 and 1. PTSD model The… view at source ↗
Figure 3
Figure 3. Figure 3: Raw outputs of the PTSD models for the individual subjects. The black curve represents the probability distribution, the gray vertical line represents the ground-truth PCL-M, and the red vertical line represents the most likely prediction according to the model. Discussion Higher accuracies to distinguish between fear response and rest samples than ours (70%) were reported in the literature [36]. This disc… view at source ↗
Figure 4
Figure 4. Figure 4: Confusion matrix for the 2-class classification problem. The model reaches an accuracy of 86% in distinguishing between PTSD and non-PTSD subjects. increasing fear responses tend to present with higher PCL-M scores. These trends are consistent with previous findings on anxiety sensitivity and emotional reactivity in PTSD populations [40], and aligns with physiological models of PTSD that highlight dysregul… view at source ↗
Figure 5
Figure 5. Figure 5: Two-dimensional representation of our model’s inputs. The 2 quantitative indicators used in this study are represented as follows: the fear response slope is on the x-axis and the initial fear response is on the y-axis; the Boolean variable sex is shown via the marker shape (a circle for a male and a star for a female). The circled subjects correspond to the two subjects analyzed in the Discussion. examine… view at source ↗
read the original abstract

Posttraumatic stress disorder (PTSD) is a prevalent and debilitating mental health condition with significant personal and societal impacts. Current clinical assessments of PTSD often rely on subjective evaluations, which can be time-consuming, costly, and prone to human bias. This study proposes a machine learning (ML) approach based on multivariate kernel density estimation (MKDE) technique for the objective evaluation of PTSD severity. We collected heart rate (HR) and galvanic skin response (GSR) signals as well as PTSD Checklist - Military Version (PCL-M) labels from 21 participants during an immersive simulation. A fear-response model was trained on a public arachnophobia dataset, and predictive features of PTSD were extracted from the fear-response curves estimated on the military dataset. The model achieved an accuracy of 86\% in classifying PTSD status, effectively distinguishing participants with and without PTSD (PCL-M threshold of 36). The average mean absolute error (MAE) of the models is 5.6, and it estimated a clinical PTSD severity scale with a mean absolute percentage error of 17\%. Our algorithm demonstrates promising potential for enhancing estimation of PTSD severity and followup by offering an objective and low-effort evaluation approach using physiology. These findings suggest clinical utility in both screening and follow-up settings.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 0 minor

Summary. The paper claims that a multivariate kernel density estimation (MKDE) fear-response model trained on a public arachnophobia dataset can be transferred to extract predictive features from heart rate (HR) and galvanic skin response (GSR) signals collected during an immersive military simulation with 21 participants. Using these features, the approach classifies PTSD status (PCL-M threshold of 36) at 86% accuracy, estimates severity with mean absolute error (MAE) of 5.6, and achieves mean absolute percentage error (MAPE) of 17%, offering an objective physiological alternative to subjective clinical assessments.

Significance. If the transfer learning result holds after proper validation, the work would demonstrate a low-effort, physiology-based method for PTSD screening and follow-up that leverages an independent public dataset, which is a strength for reproducibility. The reported performance numbers on a small clinical sample would suggest clinical utility if shown to be robust, but the current evidence does not yet establish this.

major comments (3)
  1. [Abstract] Abstract and methods description: the headline results (86% accuracy, MAE 5.6, MAPE 17% on n=21) are presented without any description of cross-validation procedure, statistical testing, or handling of the small sample size, so it is impossible to determine whether the numbers reflect genuine predictive power or overfitting.
  2. [Abstract] Abstract: the central transfer-learning claim requires that MKDE fear-response curves fitted to arachnophobia HR/GSR traces remain predictive when applied to military-simulation data, yet no comparison of response-shape statistics, timing, or amplitude distributions across the two domains is supplied; without this, the 86% accuracy cannot be attributed to transferable fear encoding rather than dataset-specific artifacts.
  3. [Methods description] Methods description: feature extraction from the MKDE model and final evaluation both occur on the identical 21-participant PTSD dataset with no reported train/test separation or independent hold-out, creating a data-dependent fitting risk that directly undermines the reported classification and regression performance.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments, which highlight important gaps in the reporting of validation and transfer-learning evidence. We will revise the manuscript to address each point.

read point-by-point responses

  1. Referee: [Abstract] Abstract and methods description: the headline results (86% accuracy, MAE 5.6, MAPE 17% on n=21) are presented without any description of cross-validation procedure, statistical testing, or handling of the small sample size, so it is impossible to determine whether the numbers reflect genuine predictive power or overfitting.

    Authors: We agree that the abstract and methods lack these details. In the revised manuscript we will expand both sections to describe the cross-validation procedure employed, any statistical tests performed on the performance metrics, and an explicit discussion of limitations arising from the n=21 cohort size. revision: yes

  2. Referee: [Abstract] Abstract: the central transfer-learning claim requires that MKDE fear-response curves fitted to arachnophobia HR/GSR traces remain predictive when applied to military-simulation data, yet no comparison of response-shape statistics, timing, or amplitude distributions across the two domains is supplied; without this, the 86% accuracy cannot be attributed to transferable fear encoding rather than dataset-specific artifacts.

    Authors: The referee is correct that no such domain-comparison statistics are currently provided. We will add quantitative comparisons (e.g., Kolmogorov-Smirnov tests or summary statistics on peak amplitude, latency, and curve shape) between the arachnophobia and military-simulation fear-response curves to support the transferability claim. revision: yes

  3. Referee: [Methods description] Methods description: feature extraction from the MKDE model and final evaluation both occur on the identical 21-participant PTSD dataset with no reported train/test separation or independent hold-out, creating a data-dependent fitting risk that directly undermines the reported classification and regression performance.

    Authors: We acknowledge the risk of data-dependent fitting. The MKDE model was trained only on the arachnophobia data, yet the downstream classifiers/regressors were evaluated on features from all 21 participants without an independent hold-out. In the revision we will implement and report a proper cross-validation scheme (e.g., leave-one-subject-out) with clear separation between feature extraction and model evaluation. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The derivation trains an MKDE fear-response model exclusively on an independent public arachnophobia dataset, then applies the fixed model to estimate curves and extract features from the separate 21-participant military dataset. No equation or step defines a target quantity in terms of itself, renames a fitted parameter as a prediction, or relies on a self-citation chain for a uniqueness claim. The reported classification accuracy and error metrics are downstream applications of the transferred model rather than reductions to the input data by construction. The central claim therefore remains self-contained with external training data providing independent support.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claim rests on the transferability of fear physiology across disorders and on the representativeness of a 21-person sample; no free parameters are explicitly named beyond the PCL-M cutoff, and no new entities are postulated.

free parameters (1)
  • PCL-M threshold = 36
    Fixed cutoff of 36 used to define PTSD-positive status; chosen from clinical convention but directly affects reported accuracy.
axioms (2)
  • domain assumption Fear-response curves derived from arachnophobia data share transferable structure with PTSD physiological responses
    Invoked when the arachnophobia-trained MKDE model is applied to the military dataset
  • domain assumption HR and GSR signals during simulation are sufficient to capture PTSD severity variation
    Core premise of the feature extraction step

pith-pipeline@v0.9.1-grok · 5780 in / 1576 out tokens · 44062 ms · 2026-06-29T22:25:16.168875+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

44 extracted references · 36 canonical work pages

  1. [1]

    Diagnostic and statistical manual of mental disorders: DSM-5

    American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-5. 5th ed. Washington, DC: American Psychiatric Publishing, a division of American Psychiatric Association; 2013

    2013

  2. [2]

    Posttraumatic stress disorder in the World Mental Health Surveys

    Koenen KC, Ratanatharathorn A, Ng L, McLaughlin KA, Bromet EJ, Stein DJ, et al. Posttraumatic stress disorder in the World Mental Health Surveys. Psychological Medicine. 2017;47(13):2260–2274. doi:10.1017/S0033291717000708

    work page doi:10.1017/s0033291717000708 2017

  3. [3]

    The epidemiology of traumatic event expo- sure worldwide: results from the World Mental Health Survey Consortium

    Benjet C, Bromet E, Karam EG, Kessler RC, McLaughlin KA, Ruscio AM, et al. The epidemiology of traumatic event expo- sure worldwide: results from the World Mental Health Survey Consortium. Psychological Medicine. 2016;46(2):327–343. doi:10.1017/S0033291715001981

    work page doi:10.1017/s0033291715001981 2016

  4. [4]

    New WHO prevalence estimates of mental disorders in conflict settings: a systematic review and meta-analysis

    Charlson F, Ommeren M, Flaxman A, Cornett J, Whiteford H, Saxena S. New WHO prevalence estimates of mental disorders in conflict settings: a systematic review and meta-analysis. The Lancet. 2019;394. doi:10.1016/S0140- 6736(19)30934-1. 8/14

    work page doi:10.1016/s0140- 2019

  5. [5]

    Lifetime Events and Posttraumatic Stress Disorder in 4 Postconflict Settings

    de Jong JTVM, Komproe IH, Van Ommeren M, El Masri M, Araya M, Khaled N, et al. Lifetime Events and Posttraumatic Stress Disorder in 4 Postconflict Settings. JAMA. 2001;286(5):555–562. doi:10.1001/jama.286.5.555

    work page doi:10.1001/jama.286.5.555 2001

  6. [6]

    Exposure to self-reported traumatic events and probable PTSD in a national sample of Poles: Why does Poland’s PTSD prevalence differ from other national estimates? PLOS ONE

    Rzeszutek M, Dragan M, Lis-Turlejska M, Schier K, Holas P, Drabarek K, et al. Exposure to self-reported traumatic events and probable PTSD in a national sample of Poles: Why does Poland’s PTSD prevalence differ from other national estimates? PLOS ONE. 2023;18. doi:10.1371/journal.pone.0287854

    work page doi:10.1371/journal.pone.0287854 2023

  7. [7]

    How Common Is PTSD in Adults? https://www

    National Center for PTSD, US Department of Veterans Affairs. How Common Is PTSD in Adults? https://www. ptsd.va.gov/understand/common/common_adults.asp

  8. [8]

    Psychometric properties of the PTSD Checklist for DSM-5 in a sam- ple of trauma exposed mental health service users

    Roberts N, Kitchiner N, Lewis C, Downes A, Bisson J. Psychometric properties of the PTSD Checklist for DSM-5 in a sam- ple of trauma exposed mental health service users. Eur J Psychotraumatol. 2022;12. doi:10.1080/20008198.2020.1863578

    work page doi:10.1080/20008198.2020.1863578 2022

  9. [9]

    Psychometric Properties of the PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition (PCL-5) in Veterans

    Bovin M, Marx B, Weathers F, Gallagher M, Rodriguez P, Schnurr P, et al. Psychometric Properties of the PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition (PCL-5) in Veterans. Psychological assessment. 2015;28. doi:10.1037/pas0000254

    work page doi:10.1037/pas0000254 2015

  10. [10]

    A comparison of the CAPS-5 and PCL-5 to assess PTSD in military and veteran treatment-seeking samples

    Resick, al PA. A comparison of the CAPS-5 and PCL-5 to assess PTSD in military and veteran treatment-seeking samples. European journal of psychotraumatology. 2023;14. doi:10.1080/20008066.2023.2222608

    work page doi:10.1080/20008066.2023.2222608 2023

  11. [11]

    Prevalence of post-traumatic stress disorder in the United States: a systematic literature review

    Schein J, Houle C, Urganus A, Cloutier M, Patterson-Lomba O, Wang Y , et al. Prevalence of post-traumatic stress disorder in the United States: a systematic literature review. Current Medical Research and Opinion. 2021;37:1–1. doi:10.1080/03007995.2021.1978417

    work page doi:10.1080/03007995.2021.1978417 2021

  12. [12]

    Prevalence, Detection and Correlates of PTSD in the Primary Care Setting: A Systematic Review

    Greene T, Neria Y , Gross R. Prevalence, Detection and Correlates of PTSD in the Primary Care Setting: A Systematic Review. Journal of clinical psychology in medical settings. 2016;23. doi:10.1007/s10880-016-9449-8

    work page doi:10.1007/s10880-016-9449-8 2016

  13. [13]

    Systematic review of machine learning in PTSD studies for automated diagnosis evaluation

    Wu Y , Mao K, Dennett L, Chen J. Systematic review of machine learning in PTSD studies for automated diagnosis evaluation. npj Mental Health Research. 2023;2. doi:10.1038/s44184-023-00035-w

    work page doi:10.1038/s44184-023-00035-w 2023

  14. [14]

    Psychophysiology of the posttraumatic stress disorder: from sulfur fumes to behavioral genetics

    Shalev AY , Rogel-Fuchs Y . Psychophysiology of the posttraumatic stress disorder: from sulfur fumes to behavioral genetics. Psychosomatic medicine. 1993;55:413–423. doi:10.1097/00006842-199309000-00003

    work page doi:10.1097/00006842-199309000-00003 1993

  15. [15]

    Neurobiological aspects of PTSD: Review of clinical and preclinical studies

    Krystal JH, Kosten TR, Southwick S, Mason JW, Perry BD, Giller EL. Neurobiological aspects of PTSD: Review of clinical and preclinical studies. Behavior Therapy. 1989;20(2):177–198. doi:https://doi.org/10.1016/S0005-7894(89)80068-1

    work page doi:10.1016/s0005-7894(89)80068-1 1989

  16. [16]

    Posttraumatic Stress Disorder and Motor Vehicle Accidents: A Multidisciplinary Overview

    Kuch K, Cox B, Evans R. Posttraumatic Stress Disorder and Motor Vehicle Accidents: A Multidisciplinary Overview. Canadian journal of psychiatry Revue canadienne de psychiatrie. 1996;41:429–34. doi:10.1177/070674379604100707

    work page doi:10.1177/070674379604100707 1996

  17. [17]

    Associations Between Childhood Trauma and Emotion- Modulated Psychophysiological Responses to Startling Sounds: A Study of Police Cadets

    Pole N, Neylan T, Otte C, Metzler T, Best S, Henn-Haase C, et al. Associations Between Childhood Trauma and Emotion- Modulated Psychophysiological Responses to Startling Sounds: A Study of Police Cadets. Journal of abnormal psychology. 2007;116:352–61. doi:10.1037/0021-843X.116.2.352

    work page doi:10.1037/0021-843x.116.2.352 2007

  18. [18]

    De Novo Conditioning in Trauma-Exposed Individuals With and Without Posttraumatic Stress Disorder

    Orr S, Metzger L, Lasko N, Macklin M, Peri T, Roger P. De Novo Conditioning in Trauma-Exposed Individuals With and Without Posttraumatic Stress Disorder. Journal of abnormal psychology. 2000;109:290–8. doi:10.1037/0021- 843X.109.2.290

    work page doi:10.1037/0021- 2000

  19. [19]

    Assessment of PTSD in military personnel via machine learning based on physiological habituation in a virtual immersive environment

    Pellegrin G, Ricka N, Fompeyrine DA, Rohaly T, Enders L, Roy H. Assessment of PTSD in military personnel via machine learning based on physiological habituation in a virtual immersive environment. Scientific Reports. 2025;15:7562. doi:10.1038/s41598-025-91916-x

    work page doi:10.1038/s41598-025-91916-x 2025

  20. [20]

    In: Posttraumatic stress disorder: from neurobiology to clinical presentation: Bremner/Posttraumatic Stress Disorder

    Shalev A, Bremner J. In: Posttraumatic stress disorder: from neurobiology to clinical presentation: Bremner/Posttraumatic Stress Disorder. John Wiley & Sons; 2016. p. 1–26

    2016

  21. [21]

    Assessing the current utilization status of wearable devices in clinical research

    Miyakoshi T, Ito Y . Assessing the current utilization status of wearable devices in clinical research. Clinical trials (London, England). 2024; p. 17407745241230287. doi:10.1177/17407745241230287

    work page doi:10.1177/17407745241230287 2024

  22. [22]

    Using Fitness Trackers and Smartwatches to Measure Physical Activity in Research: Analysis of Consumer Wrist-Worn Wearables

    Henriksen A, Mikalsen M, Woldaregay A, Muzny M, Hartvigsen G, Hopstock L, et al. Using Fitness Trackers and Smartwatches to Measure Physical Activity in Research: Analysis of Consumer Wrist-Worn Wearables. Journal of Medical Internet Research. 2018;20:e110. doi:10.2196/jmir.9157

    work page doi:10.2196/jmir.9157 2018

  23. [23]

    Functional Neuroimaging of Anxiety: A Meta-Analysis of Emotional Processing in PTSD, Social Anxiety Disorder, and Specific Phobia

    Etkin A, Wager T. Functional Neuroimaging of Anxiety: A Meta-Analysis of Emotional Processing in PTSD, Social Anxiety Disorder, and Specific Phobia. The American journal of psychiatry. 2007;164:1476–88. doi:10.1176/appi.ajp.2007.07030504

    work page doi:10.1176/appi.ajp.2007.07030504 2007

  24. [24]

    Reduced Autobiographical Memory Specificity Predicts Depression and Posttraumatic Stress Disorder After Recent Trauma

    Kleim B, Ehlers A. Reduced Autobiographical Memory Specificity Predicts Depression and Posttraumatic Stress Disorder After Recent Trauma. Journal of consulting and clinical psychology. 2008;76:231–42. doi:10.1037/0022-006X.76.2.231

    work page doi:10.1037/0022-006x.76.2.231 2008

  25. [25]

    Reexperiencing Symptoms, Dissociation, and Avoidance Behaviors in Daily Life of Patients With PTSD and Patients With Panic Disorder With Agoraphobia

    Pfaltz M, Michael T, Meyer A, Wilhelm F. Reexperiencing Symptoms, Dissociation, and Avoidance Behaviors in Daily Life of Patients With PTSD and Patients With Panic Disorder With Agoraphobia. Journal of traumatic stress. 2013;26. doi:10.1002/jts.21822

    work page doi:10.1002/jts.21822 2013

  26. [26]

    Emerging Wearable Medical Devices towards Personalized Healthcare

    Zheng J, Shen Y , Zhang Z, Wu T, Zhang G, Lu H. Emerging Wearable Medical Devices towards Personalized Healthcare. In: Conference on Body Area Networks; 2013. p. 427–431

    2013

  27. [27]

    The application of machine learning techniques in posttraumatic stress disorder: a systematic review and meta-analysis

    Wang J, Ouyang H, Jiao R, Cheng S, Zhang H, Shang Z, et al. The application of machine learning techniques in posttraumatic stress disorder: a systematic review and meta-analysis. npj Digital Medicine. 2024;7. doi:10.1038/s41746- 9/14 024-01117-5

    work page doi:10.1038/s41746- 2024

  28. [28]

    Differentiation chronic post traumatic stress disorder patients from healthy subjects using objective and subjective sleep-related parameters

    Tahmasian M, Jamalabadi H, Abedini M, Ghadami M, Sepehry A, Knight D, et al. Differentiation chronic post traumatic stress disorder patients from healthy subjects using objective and subjective sleep-related parameters. Neuroscience Letters. 2017;650. doi:10.1016/j.neulet.2017.04.042

    work page doi:10.1016/j.neulet.2017.04.042 2017

  29. [30]

    Post-Traumatic Stress Disorder: The Relationship Between the Fear Response and Chronic Stress

    Maeng L, Milad M. Post-Traumatic Stress Disorder: The Relationship Between the Fear Response and Chronic Stress. Chronic Stress. 2017;1:247054701771329. doi:10.1177/2470547017713297

    work page doi:10.1177/2470547017713297 2017

  30. [31]

    Prefrontal cortex, amygdala, and threat processing: implications for PTSD

    Kredlow M, Fenster R, Laurent E, Ressler K, Phelps E. Prefrontal cortex, amygdala, and threat processing: implications for PTSD. Neuropsychopharmacology. 2021;47:1–13. doi:10.1038/s41386-021-01155-7

    work page doi:10.1038/s41386-021-01155-7 2021

  31. [32]

    Using the PTSD Checklist for DSM-IV (PCL) https://www.ptsd.va.gov/professional/assessment/documents/PCL_handoutDSM4.pdf (accessed May 2026)

    National Center for PTSD. Using the PTSD Checklist for DSM-IV (PCL) https://www.ptsd.va.gov/professional/assessment/documents/PCL_handoutDSM4.pdf (accessed May 2026)

    2026

  32. [33]

    Soldier Performance in Sociotechnical Environments (Summary Technical Report)

    Roy H, Enders L, Cohen-Hoffing R, Touryan J, Rohaly T, Dalangin B, et al. Soldier Performance in Sociotechnical Environments (Summary Technical Report). ARL-TR-9997 APG, MD. 2024; p. 315–336

    2024

  33. [34]

    The PTSD Checklist (PCL): Reliability, validity, and diagnostic utility

    Weathers F, Litz B, Herman D, Huska JA, Keane T. The PTSD Checklist (PCL): Reliability, validity, and diagnostic utility. Paper Presented at the Annual Convention of the International Society for Traumatic Stress Studies. 1993

    1993

  34. [35]

    Electrocardiogram, skin conductance and respiration from spider-fearful individuals watching spider video clips (version 1.0.0)

    Ihmig FR, Gogeascoechea A, Schäfer S, Lass-Hennemann J, Michael T. Electrocardiogram, skin conductance and respiration from spider-fearful individuals watching spider video clips (version 1.0.0). PhysioNet. 2020. doi: 10.13026/sq6q- zg04

    work page doi:10.13026/sq6q- 2020

  35. [36]

    On-line anxiety level detection from biosignals: Machine learning based on a randomized controlled trial with spider-fearful individuals

    Ihmig F, Gogeascoechea A, Neurohr-Parakenings F, Schäfer S, Lass-Hennemann J, Michael T. On-line anxiety level detection from biosignals: Machine learning based on a randomized controlled trial with spider-fearful individuals. PLOS ONE. 2020;15:e0231517. doi:10.1371/journal.pone.0231517

    work page doi:10.1371/journal.pone.0231517 2020

  36. [37]

    Respiratory Markers Significantly Enhance Anxiety Detection Using Multimodal Physiological Sensing

    Gazi A, Lis P, Mohseni A, Ompi C, Giuste F, Shi W, et al. Respiratory Markers Significantly Enhance Anxiety Detection Using Multimodal Physiological Sensing. In: IEEE EMBS International Conference on Biomedical and Health Informatics

  37. [38]

    p. 1–4. doi: 10.1109/BHI50953.2021.9508589

    work page doi:10.1109/bhi50953.2021.9508589 2021

  38. [39]

    Spontaneous Breathing Rate Variations Linked to Social Exclusion and Emotion Self-assessment

    Hidalgo-Muñoz A, Cuadrado E, Castillo-Mayén R, Luque B, Tabernero C. Spontaneous Breathing Rate Variations Linked to Social Exclusion and Emotion Self-assessment. Applied Psychophysiology and Biofeedback. 2022;47:1–7. doi:10.1007/s10484-022-09551-5

    work page doi:10.1007/s10484-022-09551-5 2022

  39. [40]

    Toward machine emotional intelligence: analysis of affective physiological state

    Picard RW, Vyzas E, Healey J. Toward machine emotional intelligence: analysis of affective physiological state. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2001;23(10):1175–1191. doi:10.1109/34.954607

    work page doi:10.1109/34.954607 2001

  40. [41]

    Further Investigation of the Association between Anxiety Sensitivity and Posttraumatic Stress Disorder: Examining the Influence of Emotional Avoidance

    Bardeen J, Tull M, Stevens E, Gratz K. Further Investigation of the Association between Anxiety Sensitivity and Posttraumatic Stress Disorder: Examining the Influence of Emotional Avoidance. Journal of Contextual Behavioral Science. 2015;4. doi:10.1016/j.jcbs.2015.05.002

    work page doi:10.1016/j.jcbs.2015.05.002 2015

  41. [42]

    Psychometric properties of the PTSD Checklist with older primary care patients

    Cook J, Elhai J, Areán P. Psychometric properties of the PTSD Checklist with older primary care patients. Journal of traumatic stress. 2005;18:371–6. doi:10.1002/jts.20038

    work page doi:10.1002/jts.20038 2005

  42. [43]

    Screening for Posttraumatic Stress Disorder in V A Primary Care Patients with Depression Symptoms

    Gerrity M, Corson K, Dobscha S. Screening for Posttraumatic Stress Disorder in V A Primary Care Patients with Depression Symptoms. Journal of general internal medicine. 2007;22:1321–4. doi:10.1007/s11606-007-0290-5

    work page doi:10.1007/s11606-007-0290-5 2007

  43. [44]

    Bauer E. P. Sex differences in fear responses: Neural circuits. Neuropharmacology. 2023;222:109298. doi:10.1016/j.neuropharm.2022.109298

    work page doi:10.1016/j.neuropharm.2022.109298 2023

  44. [45]

    & Grunwald, G

    Hyndman, R., Bashtannyk, D. & Grunwald, G. Estimating and visualizing conditional densities.J. Comput. Graph. Stat.5, 315–336 (1996). doi: 10.1080/10618600.1996.10474715. 45.Nadaraya, E. A. On estimating regression.Theory Probab. & Its Appl.9, 141–142, 1964. doi:10.1137/1109020. 46.Watson, G. Smooth regression analysis.Sankhya: The Indian J. Stat.A, 359–3...

    work page doi:10.1080/10618600.1996.10474715 1996