Towards PPG-based anger detection for emotion regulation
Bibliographic record
Abstract
BACKGROUND: Anger dyscontrol is a common issue after traumatic brain injury (TBI). With the growth of wearable physiological sensors, there is new potential to facilitate the rehabilitation of such anger in the context of daily life. This potential, however, depends on how well physiological markers can distinguish changing emotional states and for such markers to generalize to real-world settings. Our study explores how wearable photoplethysmography (PPG), one of the most widely available physiological sensors, could be used detect anger within a heterogeneous population. METHODS: This study collected the TRIEP (Toronto Rehabilitation Institute Emotion-Physiology) dataset, which comprised of 32 individuals (10 TBI), exposed to a variety of elicitation material (film, pictures, self-statements, personal recall), over two day sessions. This complex dataset allowed for exploration into how the emotion-PPG relationship varied over changes in individuals, endogenous/exogenous drivers of emotion, and day-to-day differences. A multi-stage analysis was conducted looking at: (1) times-series visual clustering, (2) discriminative time-interval features of anger, and (3) out-of-sample anger classification. RESULTS: Characteristics of PPG are largely dominated by inter-subject (between individuals) differences first, then intra-subject (day-to-day) changes, before differentiation into emotion. Both TBI and non-TBI individuals showed evidence of linear separable features that could differentiate anger from non-anger classes within time-interval analysis. However, what is more challenging is that these separable features for anger have various degrees of stability across individuals and days. CONCLUSION: This work highlights how there are contextual, non-stationary challenges to the emotion-physiology relationship that must be accounted for before emotion regulation technology can perform in real-world scenarios. It also affirms the need for a larger breadth of emotional sampling when building classification models.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".