Valence
Valence is the pleasure-to-displeasure dimension in Russell's circumplex model of affect.
The circumplex model
James A. Russell proposed a circumplex model of affect in 1980. It organizes affect in a circle across two bipolar dimensions. Valence runs from pleasure to displeasure; arousal runs from arousal to sleepiness. The dimensions describe a framework for affect, not a rule for reading a face.
Valence and arousal answer different conceptual questions within that framework. Do not treat positive valence as a measure of intensity. Likewise, a stronger facial movement does not by itself establish greater pleasure. Specify whether a project discusses a theoretical dimension, a person's report, or a software estimate.
Software examples and scope
Py-Feat is an open-source Python toolbox for images and video. Its second version adds valence and arousal alongside gaze and a MediaPipe mesh. Those features describe Py-Feat, not an additional Mental Edge capability. A model-generated dimension should remain distinguishable from the person's own account of experience.
When evaluating a proposed score, ask what the reference labels represent. Keep the evaluation question narrow and preserve disagreement rather than relabeling it as certainty. Comparing a score with another model does not establish a person's felt experience. Use held-out clips as a review practice, not a promised performance result.
Limits of emotional interpretation
Barrett and colleagues explain that facial configurations vary across people, situations, and cultures. A configuration can express different emotions, or none. Jack and colleagues found differences between Western and East Asian representations of basic expressions. A reply argued that commonalities outweigh differences, so the debate remains active. Neither an Action Unit nor a category score proves feelings, intent, or truthfulness.
The Mental Edge review boundary
Mental Edge surfaces observable expression and movement signals from recorded video for contextual human review. Frame-by-frame scores are associated with seven expression categories, not verified inner states. Use a time-coded signal to revisit the footage and ask an open question. Do not use these outputs for hiring, promotion, diagnosis, deception detection, threat assessment, student scoring, or surveillance.
Review expression signals in context
Mental Edge provides time-coded expression signals from recorded video for human review, not proof of feelings or intent.
Start a free trial