Paperdoi:10.1088/1361-6579/aec1b3

A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality

Venue listedDated Nov 12, 2026Added Nov 13, 2026

  • MM
  • EF
  • EP
  • DF
  • DS
  • AP
  • RP

Authors: Maria Luisa Mura, Elisa Facchini, Edoardo Maria Polo, Davide Ferraris, Davide Simeone, Alessia Paglialonga, Riccardo Barbieri (Politecnico di Milano DEIB; LILT Milan; ethics approval PoliMi 39/2023)

Physiological Measurement (accepted manuscript, 2026)

First page or project page of “A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality” (opens in a new tab)

Maria Mura et al., "A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality", doi:10.1088/1361-6579/aec1b3, CC BY 4.0

TL;DR

Unsupervised look at how physiology organizes seven VR-induced emotions in 38 people using EEG, ECG, BVP, GSR and respiration. Four PCA dimensions explain 40.1% of variance. One tracks self-reported arousal, but none maps onto valence: emotions of opposite valence overlap in physiological space, and individuals respond very differently.

  • Form factor: lab (ProComp Infiniti), immersive VR

Abstract

Objective. Current affective computing relies on supervised models constrained by predefined frameworks such as the valence–arousal circumplex, precluding discovery of intrinsic physiological structures and obscuring inter-individual variability in emotional response. This study investigates the physiological organization of emotional states through an unsupervised, data-driven framework applied to multimodal physiological recordings acquired in virtual reality (VR). Approach. Thirty-eight healthy participants underwent a 35-minute VR protocol inducing seven emotions —Relaxation, Happiness, Sadness, Fear, Disgust, Empathy, and Frustration — via custom immersive scenarios. Simultaneous recordings of electroencephalographic, electrocardiographic, blood volume pulse, galvanic skin response, and respiration signals were acquired; 74 features were extracted and reduced to 36 via correlation-based selection. Principal Component Analysis (PCA) identified latent physiological dimensions, and K-means clustering was applied within each emotion to characterise inter-individual variability. Main Results. PCA revealed four physiologically interpretable dimensions, collectively explaining 40.1% of total variance, capturing distinct autonomic, cardiovascular, and neural signatures of emotional responding. Subjective arousal was broadly reflected in one component, but no dimension mapped onto subjective valence: emotions with opposing valences clustered indistinguishably in physiological space. Exploratory within-emotion clustering revealed marked inter-individual variability in the physiological response to the same emotion. A three-feature multimodal triplet (posterior alpha power, electrodermal phasic reactivity, parasympathetic cardiac tone) achieved complete separation across all emotions. Significance. Unsupervised analysis reveals organizational structures extending beyond standard affective frameworks. Individual physiological response profiles challenge the assumption of homogeneous emotional responses, providing a foundation for label-free, personalised affective computing in clinical and human-computer interaction contexts.

Verbatim from the publisher's DOI record (opens in a new tab) · Publisher licence: https://creativecommons.org/licenses/by/4.0/

Results on ChatPPG leaderboards

All leaderboards

No results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.

Tasks & signals

Tasks

StressPulse arrival time

Keywords

Affective ComputingEmotionValence ArousalUnsupervisedPCAK MeansInter Individual VariabilityVirtual Reality

How it works

How Method works: Setup → Method → TargetSetup: Signals (eeg, ecg, bvp, eda, resp); Form factor (lab (ProComp Infiniti), immersive VR). Method: Method. Target: Stress; Pulse arrival time.SetupSignals: eeg, ecg, bvp, eda, respSignalseeg, ecg, bvp, eda, respForm factor: lab (ProComp Infiniti), immersive VRForm factorlab (ProComp Infiniti), immersive VRMethodMethodTargetStressStressPulse arrival timePulse arrival time

Cite this paper

Cite this paper with the fields recorded here. Check it against the publisher before submitting.

Mura, M. L., Facchini, E., Polo, E. M., Ferraris, D., Simeone, D., Paglialonga, A., & Barbieri, R. (2026). A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality. Physiological Measurement. https://doi.org/10.1088/1361-6579/aec1b3
Mura, Maria Luisa, Elisa Facchini, Edoardo Maria Polo, Davide Ferraris, Davide Simeone, Alessia Paglialonga, and Riccardo Barbieri. 2026. "A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality." Physiological Measurement. https://doi.org/10.1088/1361-6579/aec1b3.
Mura, M. L., Facchini, E., Polo, E. M., Ferraris, D., Simeone, D., Paglialonga, A. and Barbieri, R. (2026) 'A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality', Physiological Measurement. doi: 10.1088/1361-6579/aec1b3.
Mura, Maria Luisa, et al. "A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality." Physiological Measurement, 2026. https://doi.org/10.1088/1361-6579/aec1b3.
M. L. Mura et al., "A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality," Physiological Measurement, 2026, doi: 10.1088/1361-6579/aec1b3.
@article{mura2026datadriven,
  title         = {{A Data-Driven Multidimensional Framework for Physiological Emotion Characterisation in Immersive Virtual Reality}},
  author        = {Mura, Maria Luisa and Facchini, Elisa and Polo, Edoardo Maria and Ferraris, Davide and Simeone, Davide and Paglialonga, Alessia and Barbieri, Riccardo},
  year          = {2026},
  journal       = {Physiological Measurement},
  doi           = {10.1088/1361-6579/aec1b3}
}

Added to the ChatPPG research hub