Face time: AI research reveals 16 facial expressions most common to emotional situations

Artificial intelligence: Expert discusses research on future crime

The landmark new research by the University of California confirms the universality of human emotional expression across geographic and cultural boundaries. Professor Dacher Keltner, a UC Berkeley psychologist and the study’s co-lead author, said: “This study reveals how remarkably similar people are in different corners of the world in how we express emotion in the face of the most meaningful contexts of our lives.”

Researchers at UC Berkeley and Google used machine-learning Artificial Intelligence (AI) tech known as a “deep neural network”.

This supports Darwin’s theory that expressing emotion in our faces is universal among humans

Professor Dacher Keltner

This enabled analysis of facial expressions in approximately six million video clips uploaded to YouTube from 144 countries.

Dr Alan Cowen, a researcher at both UC Berkeley and Google who helped develop the AI algorithm, said: “This is the first worldwide analysis of how facial expressions are used in everyday life, and it shows us that universal human emotional expressions are a lot richer and more complex than many scientists previously assumed.”

Dr Cowen created an interactive map to demonstrate how the algorithm kept track of variations of facial expressions commonly associated with 16 emotions.

READ MORE: AI-manipulated media will be ‘WEAPONISED’ to trick military

We will use your email address only for sending you newsletters. Please see our Privacy Notice for details of your data protection rights.

Applications of the cutting-edge AI tech include helping people who have trouble reading emotions, such as infants and those with autism, to recognise the faces humans commonly make to convey certain feelings.

And researchers also hope their work may also help promote cross-cultural empathy.

The typical human face has 43 different muscles which can be activated to create possibly thousands of different expressions.

Dr Cowen’s machine-learning algorithm was first employed to track facial expressions shown in six million video clips.

These ranged from firework spectators, people dancing or consoling an upset child.

They used the algorithm to track examples of 16 facial expressions commonly associated with the following emotions: amusement, anger, awe, concentration, confusion, contempt, contentment, desire, disappointment, doubt, elation, interest, pain, sadness, surprise and triumph.

They then cross-referenced the facial expressions with the contexts and scenarios in which they were made across different areas of the world.

This resulted in discovering incredible similarities in how people across geographic and cultural boundaries use facial expressions in different social contexts.

Dr Cowen said: ”We found that rich nuances in facial behaviour—including subtle expressions we associate with awe, pain, triumph, and 13 other feelings — are used in similar social situations around the world.”

The YouTube footage revealed people around the world tended to stare in awe during firework displays, display contentment at weddings and, furrow their brows in concentration when performing martial arts.

They also show doubt at protests, pain when working out, and triumph at attending gigs and football matches.

The results revealed people from different cultures share approximately 70 percent of the facial expressions used in response to different social and emotional situations.

Professor Keltner said: ”This supports Darwin’s theory that expressing emotion in our faces is universal among humans.

“The physical display of our emotions may define who we are as a species, enhancing our communication and cooperation skills and ensuring our survival.”

Source: Read Full Article