Recognizing static and dynamic facial expressions of pain : Gaze-tracking and Bubbles experiments
Bibliographic record
Abstract
Facial expression is considered to be the most reliable source of information when judging on the pain intensity experienced by another (Poole & Craig 1992). Nonetheless, observers in this situation show a systematic under-estimation bias (Harrison, 1993; Kappesser & Williams, 2002). In the medical domain, this bias results in under-treatment, which leads to an insufficient pain relief for suffering patients. Despite the important impact of pain identification on patient well-being, the visual processes involved in the recognition of the facial expression of pain remain unknown. In this study, we used gaze-tracking and Bubbles (Gosselin & Schyns, 2001) to investigate the visual information used for the recognition of static and dynamic facial expressions of pain. Observers were required to categorize 80 dynamic or static facial expressions (the 6 basic emotions, pain and neutral) from the STOIC database (Roy et al., 2007). In the gaze-tracking experiment, twenty observers saw 6 times each of the 80 static and dynamic emotionnal faces. Gaze position was recorded while the stimuli were presented; heat maps were computed. The results for pain expressions will be discussed. In the Bubbles experiments, 5,000 sparse versions of these static and dynamic stimuli were created by sampling facial information at random spatial locations at five one-octave non-overlapping spatial frequency bands for the static stimuli, as well as in space-time for the dynamic stimuli (see Vinette, Gosselin & Schyns, 2004). Online calibration of sampling density ensured 75% overall accuracy. We performed mulitple linear regressions on sample space or space-time locations and on accuracy to reveal the information effectively used to recognize pain. Preliminary findings with static stimuli reveal that optimal information for the recognition of pain partly overlap with sadness and disgust ones. Preliminary results with dynamic stimuli indicate that motion contributes to the decoding of facial expressions of pain.
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 | 0.001 | 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".