From subjective to objective facial data
Facial Action Coding System (FACS)
If you want to understand emotion, engagement, or subtle social signals, you need a method you can trust. FACS gives you a proven, scientific way to describe facial behavior objectively and consistently.
- Turn facial expressions into objective data
- Don’t miss critical micro-expressions
- Make your facial expression data publishable and comparable
What is FACS?
The Facial Action Coding System (FACS) is a scientific framework for describing facial movements. It was developed by Paul Ekman and Wallace Friesen, whose work laid the foundation for modern facial expression research.
Instead of categorizing emotional expressions, FACS breaks facial expressions down into Action Units (AUs), which are specific, observable muscle movements such as raising the eyebrows or tightening the lips. The level of activation of each Action Unit is specified on a 5-level intensity scale. Each expression can be described by a unique combination of these Action Units. That makes facial behavior measurable, comparable, and independent of interpretation.
Facial Action Coding Scheme / 20 Action Units
Below you can see the 20 Action Units from the Facial Action Coding Scheme (FACS) offered in FaceReader as well as some frequently occurring or difficult Action Unit combinations.
Some images have been zoomed in on the area of interest to explicitly show what muscle movement corresponds to the specific Action Unit.
AU 1. Inner Brow Raiser
Contributes to sadness, surprise, and fear. Muscular basis: frontalis (pars medialis).
AU 2. Outer Brow Raiser
Contributes to surprise and fear. Muscular basis: frontalis (pars lateralis).
AU 4. Brow Lowerer
Contributes to sadness, fear, and anger. Muscular basis: depressor glabellae, depressor supercilii, corrugator supercilii.
AU 5. Upper Lid Raiser
Contributes to surprise, fear, and anger. Muscular basis: levator palpebrae superioris, superior tarsal muscle.
AU 6. Cheek Raiser
Contributes to happiness. Muscular basis: orbicularis oculi (pars orbitalis).
AU 7. Lid Tightener
Contributes to fear and anger. Muscular basis: orbicularis oculi (pars palpebralis).
AU 9. Nose Wrinkler
Contributes to disgust. Muscular basis: levator labii superioris alaeque nasi.
AU 10. Upper Lip Raiser
Muscular basis: levator labii superioris, caput infraorbitalis.
AU 12. Lip Corner Puller
Contributes to happiness and contempt. Muscular basis: zygomaticus major.
AU 14. Dimpler
Contributes to contempt and boredom. Muscular basis: buccinator.
AU 15. Lip Corner Depressor
Contributes to sadness and disgust. Muscular basis: depressor anguli oris.
AU 17. Chin Raiser
This Action Unit contributes to the affective attitudes interest and confusion. The underlying facial muscle is mentalis.
AU 18. Lip Pucker
The underlying facial muscles are incisivii labii superioris and incisivii labii inferioris.
AU 20. Lip Stretcher
Contributes to the emotion fear. The underlying facial muscle is risorius w/ platysma.
AU 23. Lip Tightener
Contributes to the emotion anger, and to the affective attitudes confusion and boredom. Muscular basis: orbicularis oris.
AU 24. Lip Pressor
This Action Unit contributes to the affective attitude boredom. The underlying facial muscle is orbicularis oris.
AU 25. Lips Part
The muscular basis consists of depressor labii inferioris, or relaxation of mentalis or orbicularis oris.
AU 26. Jaw Drop
Contributes to the emotions surprise and fear. Muscular basis: masseter; relaxed temporalis and internal pterygoid.
AU 27. Mouth Stretch
The underlying facial muscles are pterygoids and digastric.
AU 43. Eyes Closed
Contributes to the affective attitude boredom. The muscular basis consists of relaxation of Levator palpebrae superioris.
Combinations of Action Units
AU 1 - 2 - 4
Contributes to the emotions fear and can be recognized by the wavy pattern of the wrinkles across the forehead.
AU 1 - 2
Contributes to the emotion surprise and can be recognized by a smooth line formed by the wrinkles across the forehead.
AU 1 - 4
Contributes to sadness. Recognizable by a wavy pattern of the wrinkles in the center of the forehead. Eye-brows come together and up.
AU 4 - 5
Contributes to the emotion anger.
AU 6 - 12
Contributes to happiness. Notice the wrinkles around the eyes caused by cheek raising, also known as the "Duchenne Marker".
AU 10 - 25
Contributes to the emotion disgust. When AU 10 is activated intensely, it causes the lips to part as the upper lip raises.
AU 18 - 23
Often confused as solely AU 18. Notice the lips almost appear to be pulled by a single string outward (AU 18) and then tightened (AU 23).
AU 23 - 24
The AUs marking lip movements are often the hardest to code. The lips are being pushed together (AU 24) and tightened (AU 23).
How does FaceReader help you work with FACS?
It takes humans several hundred hours to become an expert in recognizing these movements. Facial Action Coding System software, like FaceReader, automatically detects and quantifies Action Units from video, frame by frame.
It brings FACS into practice, removes practical barriers, and allows you to define custom expressions by combining Action Units in ways that match your research goals.
That means you can:
- Apply FACS-based facial coding automatically and consistently
- Quantify Action Units at scale, without introducing human bias
- Produce objective facial expression data suitable for scientific reporting
See how other researchers apply FACS and FaceReader in practice
The publications and blog posts below include references on FACS, the ADFES dataset, and examples from FaceReader users, offering context for how facial expression analysis is used in research.
For brands seeking to resonate with customers, being seen as authentic is vital. Using custom expressions in FaceReader, we explore how perception of brand authenticity relates to ad performance.
The study described in this guest blog post focuses on the facial expressions of emotions induced by affective stimuli in children aged between 7 and 14.
Using custom expressions in facial expression analysis can unlock a deeper understanding of human behavior. How can you use custom expressions in different fields of research?
Of all human expressions, a smile is the most universal. But can you tell which smile is real and which is false?
The concept engagement is gaining more and more attention. Many companies are looking for ways to increase consumer engagement. But, how do you know a consumer is feeling engaged?