Recognizing emotion from movement has no single established method
A systematic review compared human and machine approaches, finding varied methods and unresolved limits in accuracy and individual emotional expression.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
This abstract-based review found no universally accepted method for identifying emotions from body movement. The authors considered artificial intelligence useful but limited when used alone, and favored combining it with human observation and conventional motion analysis. The abstract does not provide numerical accuracy comparisons that would establish which approach performs best or in whom.
The abstract does not describe study participants, testing settings, or method-specific accuracy estimates. It also provides no quantitative comparison supporting the proposed combined approach, limiting judgments about reliability in a particular setting.
What the review catalogued
The review included 16 studies and identified 9 motion-analysis methods, including 4 that used artificial intelligence. Its focus was how emotions were recognized from movement, rather than whether any method improved health or wellbeing.
Why a universal method remains elusive
The methods differed in their technologies, data processing, and reported accuracy. The authors also identified challenges in algorithm development and differences between people's emotional expression, which complicate interpreting movement as a dependable signal of emotion.
The original publication
[Emotion detection by motion analysis: a comparison of human and machine-based processing].
Naor B, Egri D, Somogyi A et al.
Orv Hetil · 2025
- PubMed ID
- 41241882
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.