EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Uncovering Heart Rate Response Patterns to Threat Pictures Through Deep Latent Representation Learning with a Variational Autoencoder

Stephan Moratti iD, Sergio Felipe Calvo García iD

Psychophysiology · 2026-05 · 63 (5)

DOI: 10.1111/psyp.70314 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Psychophysiology, online erschienen am 10.05.2026. Verfasst von Stephan Moratti und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Experimental Psychology, Cognitive Processes and Speech Therapy, Complutense University of Madrid, Madrid, Spain. Hauptthemen laut MEDLINE: Autoencoder, Heart Rate, Fear, Deep Learning. Die Arbeit nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Ministerio de Ciencia, Innovación y Universidades.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Group‐level averaging of psychophysiological data often obscures meaningful individual differences, masking response patterns that may explain variability in behavior and central nervous system activity. Identifying such patterns is particularly relevant in heart rate (HR) responses to threat, where subtle variations may reflect distinct coping mechanisms such as orienting or defense. Machine learning techniques that learn latent representations, particularly variational autoencoders (VAEs), offer powerful tools for revealing such hidden structures. This methodological report introduces a simple VAE‐based approach for characterizing HR responses to threat pictures in 165 participants. To validate the method, simulations first demonstrated that the model accurately separated simulated HR waveforms. The VAE was then applied to empirical HR responses, mapping them into a three‐dimensional latent space for subsequent cluster analysis, which was compared to clustering based directly on raw HR waveforms. The VAE revealed three distinct response profiles: (1) strong decelerators (cardiac orienting response), (2) weak decelerators with late acceleration, and (3) immediate accelerators without a decelerative phase. In contrast, clustering raw HR waveforms identified only two groups. Clusters derived from the latent space were more coherent and exhibited greater within‐group consistency. Finally, applying the pre‐trained autoencoder to a small fear‐conditioning dataset enabled characterization of distinct HR response patterns despite limited sample size. These findings show that even a basic autoencoder enhances the categorization of psychophysiological response patterns, offering a framework for linking individual autonomic variability to broader models of affective and defensive behavior.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AutoencoderHeart RateFearDeep LearningHumansFemaleMaleAdultYoung AdultRepresentation Machine LearningElectrocardiography

Stichwörter der Autor:innen

deep learningdefenseheart rateorientingthreatvariational autoencoder

Autor:innen und Institutionen

  • Department of Experimental Psychology, Cognitive Processes and Speech Therapy, Complutense University of Madrid, Madrid, Spain.
  • Emotional Processing Laboratory (EPL), Center for Cognitive and Computational Neuroscience C3N, Complutense University of Madrid, Madrid, Spain.

ORCID-Profile: Stephan Moratti, Sergio Felipe Calvo García

Förderung

  • Ministerio de Ciencia, Innovación y Universidades

Zitieren

APA

Moratti, S., & García, S. F. C. (2026). Uncovering Heart Rate Response Patterns to Threat Pictures Through Deep Latent Representation Learning with a Variational Autoencoder. Psychophysiology, 63(5). https://doi.org/10.1111/psyp.70314

Vancouver

Moratti S, García SFC. Uncovering Heart Rate Response Patterns to Threat Pictures Through Deep Latent Representation Learning with a Variational Autoencoder. Psychophysiology. 2026;63(5). doi:10.1111/psyp.70314

BibTeX
@article{Moratti2026,
  author = {Moratti, Stephan and García, Sergio Felipe Calvo},
  title = {Uncovering Heart Rate Response Patterns to Threat Pictures Through Deep Latent Representation Learning with a Variational Autoencoder},
  journal = {Psychophysiology},
  year = {2026},
  volume = {63},
  number = {5},
  doi = {10.1111/psyp.70314}
}
Kostenlos starten →