Table 2.
Summary of previous research.
| No | Author | Emotion | Signals | Features | Data Analysis | Subjects | HMD | VR Stimuli | Stimuli Comparison | Dataset Availability |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Jang et al. (2002) [104] | Arousal | HRV, EDA | HR, HRV frequency domain, SCL, ST | t-test | 11 | VFX3D | 3D flying and driving simulator | No | No |
| 2 | Meehan et al. (2005) [107] | Arousal | HRV, EDA | HR, SC, ST | t-test | 67 | Not reported | 3D training room vs. pit room | No | No |
| 3 | Wilhelm et al. (2005) [108] | Anxiety | HRV, EDA | HR, SC | ANOVA, correlations | 86 | Not reported | 3D height exposure | Partially (with a different real dataset) | No |
| 4 | Gorini et al. (2010) [109] | Anxiety | HRV, EDA | HR, SC | ANOVA | 30 (20 with food disorders) | Not reported | 3D photo and real food catering | VR vs. photo vs. real | No |
| 5 | Philipp et al. (2012) [110] | Valence | EMG | EMG | ANOVA | 49 | Virtual Research V8 | 3D room with IAPS pictures projected | No | No |
| 6 | Parsons et al. (2013) [111] | Arousal | HRV, EDA | HR, SC | ANOVA | 50 | eMagin Z800 | 3D high-mobility wheeled vehicle with Stroop task | No | No |
| 7 | Pallavicini et al. (2013) [112] | Stress | HRV, EMG, RSP | HR, SC, RR | ANOVA | 39 | Vuzix VR Bundle | 3D classroom | No | No |
| 8 | Peperkorn et al. (2014) [43] | Fear | HRV, EDA | HR, SC | ANOVA | 96 (48 spider-phobic) | eMagin Z800 | 3D virtual lab with time-varying threat (spiders and snakes) | No | No |
| 9 | Felnhofer et al. (2014) [113] | Anxiety | HRV | HR | ANOVA | 75 (30 high anxiety) | eMagin Z800 | 3D lecture hall | No | No |
| 10 | Hartanto et al. (2014) [114] | Stress | HRV | HR | MANOVA | 24 healthy subjects | eMagin Z800 | 3D stressful social environment | No | No |
| 11 | McCall et al. (2015) [115] | Arousal | HRV, EDA | HR, SC | Cross-correlations | 306 | NVIS nVisor SX60 | 3D room with time-varying threat (explosions, spiders, gunshots, etc.) | No | No |
| 12 | Felnhofer et al. (2015) [54] | Arousal | EDA | SCL | ANOVA | 120 | Sony HMZ-T1 3D | 3D park with 5 variations (joy, sadness, boredom, anger and anxiety) | No | No |
| 13 | Notzon et al. (2015) [116] | Anxiety | HRV, EDA | HR, SC | ANOVA | 83 (42 spider-phobic) | eMagin Z800 | 3D virtual lab with spiders | No | No |
| 14 | Hildebrandt et al. (2016) [117] | Arousal | HRV, EDA | RMSSD, SC | Regression | 300 | NVIS nVisor SX60 | 3D room with time-varying threats (explosions, spiders, gunshots, etc.) | No | No |
| 15 | Higuera-Trujillo et al. (2016) [118] | Stress | EDA | SCR | Kruskall–Wallis Test and correlations | 12 | Oculus Rift DK2 | 3D rooms (neutral, stress and calm) | No | No |
| 16 | Bian et al. (2016) [119] | Arousal | HRV, EMG, RSP | HR, LF, HF, LF/HF, RR, RS | Regression | 36 | Oculus Rift DK2 | 3D Flight simulator | No | No |
| 17 | Shiban et al. (2016) [120] | Stress | HRV, EDA | HR, SC | ANOVA | 45 | NVIS nVisor SX60 | 3D Trier Social Stress Test | No | No |
| 18 | Chirico et al. (2017) [121] | Awe | HRV, EDA, EMG | HF, VLF, SC | ANOVA | 42 | Samsung Gear VR | 360° neutral and awe videos | Immersive vs. non-immersive | No |
| 19 | Zou et al. (2017) [122] | Arousal | HRV, EDA | HRV time domain (AVNN, SDNN…) and frequency domain (LF, HF…), SC, SCL, SCR | t-test | 40 | Oculus Rift DK2 | 3D fire evacuation | No | No |
| 20 | Breuninger et al. (2017) [123] | Arousal | HRV, EDA | HR, HF, SC | t-test | 51 (23 agoraphobics) | TriVisio VR Vision | 3D car accident | No | No |
| 21 | van’t Wout et al. (2017) [124] | Stress | EDA | SCR | MANOVA | 44 veterans (19 with PTSD) | eMagin Z800 | 3D combat-related and classroom-related | No | No |
| 22 | Banaei et al. (2017) [125] | Arousal, Valence | EEG | PSD, ERSPs | MANOVA | 17 | Samsung Gear VR | 3D rooms | No | No |
| 23 | Anderson et al. (2017) [126] | Stress | HRV, EDA | LF, HF, LF/HF, SC | MANOVA | 18 | Oculus Rift DK2 | 360° indoor vs. natural panoramas | No | No |
| 24 | Chittaro et al. (2017) [127] | Arousal | HRV | HR, LF, HF, LF/HF | ANOVA | 108 | Sony HMZ-T1 3D | 3D cemetery and park | No | No |
| 25 | Higuera-Trujillo et al. (2017) [64] | Pleasantness | HRV, EDA | HF, SCR | Mann–Whitney U tests and correlations | 100 | Samsung Gear VR | 3D, 360° and real retail store | real vs. 3D VR vs. 360° VR | No |
| 26 | Biedermann et al. (2017) [128] | Anxiety | HRV, EDA, RSP | HR, SC, RR | ANOVA | 100 | HTC Vive | Mixed reality (3D VR with real-world elements) | No | Yes |
| 27 | Tsai et al. (2018) [129] | Anxiety | HRV | HRV time domain (HR, RMSSD…) and frequency domain (HF, LF…) | ANOVA | 30 | eMagin Z800 | 3D VR claustrophobic environments | Augmented reality vs. VR | Upon request |
| 28 | Marín-Morales et al. (2018) [105] | Arousal, Valence | EEG, HRV | PSD and functional connectivity, HRV Time (HR, RMSSD…), frequency (HF, LF…) and non-linear (SD1, SD2, Entropy…) domain | SVM | 60 | Samsung Gear VR | 360° virtual rooms | No | Upon request |
| 29 | Kisker et al. (2019) [130] | Arousal | HRV | HR | t-test, correlations and regressions | 30 | HTC Vive | 3D exposure to a high height | No | No |
| 30 | Gromer et al. (2019) [131] | Fear | HRV, EDA | HR, SC | ANOVA | 49 (height-fearful) | HTC Vive | 3D forest | No | Yes |
| 31 | Zimmer et al. (2019) [132] | Stress | HRV, salivary | HR, salivary cortisol responses, salivary alpha amylase | ANOVA | 50 | Oculus Rift DK2 | 3D Trier Social Stress Test | Replication of a real study | No |
| 32 | Lin et al. (2019) [133] | Stress | EDA, Navigation | SC, travel distance, travel time | Mann–Whitney U | 60 | HTC Vive | 3D, building on fire | No | No |
| 33 | Schweizer et al. (2019) [134] | Stress | HRV, EDA | HR, SC | t-test and correlations | 80 | TriVisio VR Vision | 3D neutral and trauma-related scene | No | No |
| 34 | Kim et al. (2019) [135] | Calm, sadness and joy | Gait Patterns | Step count, gait speed, foot plantar pressure | ANOVA | 12 | HTC Vive | 360° emotion-related videos | No | No |
| 35 | Uhm et at. (2019) [136] | Arousal | EEG | PSD | MANOVA | 28 | Samsung Gear VR | 360° sport videos | No | No |
| 36 | Takac et al. (2019) [137] | Anxiety | HRV | HR | ANOVA | 19 | Oculus Rift | 3D rooms with public audience | No | No |
| 37 | Marín-Morales et al. (2019) [65] | Arousal, Valence | HRV, EEG | PSD and functional connectivity, HRV Time (HR, RMSSD…), frequency (HF, LF…) and non-linear (SD1, SD2, Entropy…) domain | SVM | 60 | HTC Vive | 3D art museum | Real museum vs. 3D museum | Upon request |
| 38 | Stolz et al. (2019) [138] | Fear | EEG | ERPs | ANOVA | 29 | Oculus Rift | 3D room with angry avatars | No | No |
| 39 | Granato et al. (2020) [139] | Arousal, Valence | HRV, EDA, EMG, RSP | HR, SC, SCL, SCR, EMG, RR | SVM, RF, Gradient Boosting, Gaussian Process Regression | 33 | Oculus Rift DK2 | 3D video games | No | Yes |
| 40 | Bălan et al. (2020) [140] | Fear | HRV, EDA, EEG | HR, SC, PSD | kNN, SVM, RF, LDA, NN | 8 | HTC Vive | 3D acrophobia game | No | No |
| 41 | Reichenberger et al. (2020) [141] | Fear | Eye-tracking | Fixation counts, TTFF | ANOVA, t-test | 53 (26 socially anxious) | HTC Vive | 3D room with angry avatars | No | Upon request |
| 42 | Huang et al. (2020) [142] | Stress | EDA | SCL | MANOVA | 89 | Oculus Rift DK2 | 360° built vs. natural environments | No | Yes |
Signals: electroencephalograph (EEG), heart rate variability (HRV), electrodermalactivity (EDA), respiration (RSP) and electromyography (EMG). Features: heart rate (HR), high frequency (HF), low frequency (LF), LF/HF (low/high frequency ratio), very low frequency (VLF), total skin conductance (SC), skin conductance tonic level (SCL), fast varying phasic activity (SCR), skin temperature (ST), respiratory rate (RR), respiratory depth (RS), power spectral density (PSD), event-related spectral perturbations (ERSPs), event-related potencials (ERPs) and time to first fixation (TTFF). Data analysis: support vector machines (SVM), k-nearest neighbors algorithm (kNN), random forest (RF), linear discriminant analysis (LDA) and neural networks (NN).