Table 4.
Affective states and sensor signals frequently employed in wearable-based AR. Table 9 provides further detail on algorithms, location and performance. Abbreviations: 3-axes acceleration (ACC), blood pressure (BP), (EEG), electromyogram (EMG), electrooculography (EOG), heart rate (HR), magnetoencephalogram (MEG), pupil diameter (PD), photoplethysmogram (PPG), respiration (RESP), skin-temperature (TEMP), arterial oxygen level (SpO2), low arousal/low valence (LALV), low arousal/high valence (LAHV), high arousal/low valence (HALV), high arousal/high valence (HAHV)
| Author | Affective States | Sensor Signals | |
|---|---|---|---|
| <2005 | Picard et al. [12] | Neutral, anger, hate, grief, joy, platonic/romantic love, reverence | EDA, EMG, PPG, RESP |
| Haag et al. [77] | Low/medium/high arousal and positive/negative valence | ECG, EDA, EMG, TEMP, PPG, RESP | |
| Lisetti and Nasoz [70] | Sadness, anger, fear, surprise, frustration, amusement | ECG, EDA, TEMP | |
| 2005 | Liu et al. [78] | Anxiety, boredom, engagement, frustration, anger | ECG, EDA, EMG |
| Wagner et al. [79] | Joy, anger, pleasure, sadness | ECG, EDA, EMG, RESP | |
| Healey and Picard [63] | Three stress levels | ECG, EDA, EMG, RESP | |
| 07 | Leon et al. [80] | Neutral/positive/negative valence | EDA, HR, BP |
| 2008 | Zhai and Barreto [81] | Relaxed and stressed | EDA, PD, PPG, TEMP |
| Kim et al. [82] | Distinguish high/low stress group of individuals | PPG | |
| Kim and André [34] | Four quadrants in valence-arousal space | ECG, EDA, EMG, RESP | |
| Katsis et al. [83] | High/low stress, disappointment, euphoria | ECG, EDA, EMG, RESP | |
| 2009 | Calvo et al. [84] | Neutral, anger, hate, grief, joy, platonic/romantic love, reverence | ECG, EMG |
| Chanel et al. [85] | Positively/negatively excited, calm-neutral (in valence-arousal space) | BP, EEG, EDA, PPG, RESP | |
| Khalili and Moradi [86] | Positively/negatively excited, calm (valence-arousal space) | BP, EEG, EDA, RESP, TEMP | |
| 2010 | Healey et al. [87] | Points in valence arousal space. moods | ACC, EDA, HR, audio |
| 2011 | Plarre et al. [46] | Baseline, different types of stress (social, cognitive and physical), perceived stress | ACC, ECG, EDA, RESP, TEMP, ambient temperature |
| Hernandez et al. [88] | Detect stressful calls | EDA | |
| 2012 | Valenza et al. [89] | Five classes of arousal and five valence levels | ECG, EDA, RESP |
| Hamdi et al. [90] | Joy, sadness, disgust, anger, fear, surprise | ECG, EEG, EMG | |
| Agrafioti et al. [91] | Neutral, gore, fear, disgust, excitement, erotica, game elicited mental arousal | ECG | |
| Koelstra et al. [35] | Four quadrants in valence-arousal space | ECG, EDA, EEG, EMG, EOG, RESP, TEMP, facial video | |
| Soleymani et al. [31] | Neutral, anxiety, amusement, sadness, joy, disgust, anger, surprise, fear | ECG, EDA, EEG, RESP, TEMP | |
| 2013 | Sano and Picard [92] | Stress vs. neutral | ACC, EDA, phone usage |
| Martinez et al. [93] | Relaxation, anxiety, excitement, fun | EDA, PPG | |
| 2014 | Valenza et al. [36] | Four quadrants in valence-arousal space | ECG |
| Adams et al. [94] | Stress vs. neutral (aroused vs. non-aroused) | EDA, audio | |
| 2015 | Hovsepian et al. [95] | Stress vs. neutral | ECG, RESP |
| Abadi et al. [37] | High/Low valence, arousal and dominance | ECG, EOG, EMG, near-infrared face video, MEG | |
| 2016 | Rubin et al. [96] | Panic attack | ACC, ECG, RESP |
| Jaques et al. [97] | Stress, happiness, health values | EDA, TEMP, ACC, phone usage | |
| Rathod et al. [98] | Normal, happy, sad, fear, anger | EDA, PPG | |
| Zenonos et al. [29] | Excited, happy, calm, tired, bored, sad, stressed, angry | ACC, ECG, PPG, TEMP | |
| Zhu et al. [99] | Angle in valence arousal space | ACC, phone context | |
| Birjandtalab et al. [76] | Relaxation, different types of stress (physical, emotional, cognitive) | ACC, EDA, TEMP, HR, SpO2 | |
| 2017 | Gjoreski et al. [13] | Lab: no/low/high stress; Field: stress vs. neutral |
ACC, EDA, PPG, TEMP |
| Mozos et al. [45] | Stress vs. neutral | ACC, EDA, PPG, audio | |
| Taylor et al. [100] | Tomorrow’s mood, stress, health | ACC, EDA, context | |
| Girardi et al. [101] | High vs. low valence and arousal | EEG, EDA, EMG | |
| 2018 | Schmidt et al. [64] | Neutral, amusement, stress | Torso: ACC, ECG, EDA, EMG, RESP, TEMP; Wrist: ACC, EDA, PPG, TEMP |
| Zhao et al. [102] | LALV, LAHV, HALV, HAHV | EDA, PPG, TEMP | |
| Marín-Morales et al. [103] | LALV, LAHV, HALV, HAHV | ECG, EEG | |
| Santamaria- Granados et al. [104] | LALV, LAHV, HALV, HAHV | ECG, EDA | |
| 2019 | Heinisch et al. [67] | High positive pleasure high arousal, high negative pleasure high arousal and neutral | EMG, PPG, TEMP |
| Hassan et al. [105] | Happy, relaxed, disgust, sad and neutral | EDA, PPG, EMG (from DEAP) | |
| Kanjo et al. [75] | Five valence classes | ACC, EDA, HR, TEMP, environmental, GPS | |
| Di Lascio et al. [66] | Detect laughter episodes | ACC, EDA, PPG |