Last updated March 28, 2018 at 1:30 pm
Lady Gaga bragged of hers, but even your best poker face might unconsciously betray your feelings
Keeping a “poker face” may not make you as inscrutable as you think and hope. Even if you don’t move a muscle, your emotions may be clearly on show.
Research at Ohio State University has found we can correctly identify another person’s feelings up to 75 per cent of the time based solely on subtle shifts in blood flow colour around the nose, eyebrows, cheeks or chin.
The reason is a never-before-documented connection between the central nervous system and emotional expression in the face.
“We identified patterns of facial colouring that are unique to every emotion we studied,” said cognitive scientist Professor Aleix Martinez.
“We believe these colour patterns are due to subtle changes in blood flow or blood composition triggered by the central nervous system. Not only do we perceive these changes in facial colour, but we use them to correctly identify how other people are feeling, whether we do it consciously or not.”
For their study, Martinez and colleagues separated hundreds of pictures of facial expressions into different colour channels that correspond to how human eyes see colour: a red-green channel and a blue-yellow channel.
Through computer analysis, they found that emotions such as “happy” or “sad” formed unique colour patterns. Everybody displayed similar patterns when expressing the same emotion, regardless of gender, ethnicity or overall skin tone.
Facial colour reveals emotion
To test whether colour alone could convey emotions the researchers then superimposed different emotional colour patterns on pictures of faces with neutral expressions. Emotions included basic ones like “happy” and “sad” as well as more complex ones such as “sadly angry” or “happily surprised.”
“Admittedly, these images look weird but we told people to go ahead and guess from the list of emotions what emotion they thought those faces were conveying,” Martinez said. “And they guessed right most of the time.”
Next, participants were shown facial expressions of happiness, sadness and other emotions, but with the colours mixed up on some of the images. For example, they put angry colours on a happy face, or vice versa. In these cases, participants noticed that something was wrong, even if they weren’t sure what it was.
From there, the researchers used what they had learned to develop computer algorithms that could detect emotions via face colour. Given photographs of people expressing emotion, the computer could match colour to feeling better than the human study participants could.
Happiness was the easiest emotion for the computer to recognise by colour alone, and it detected the emotion with 90 per cent accuracy. Anger was detectable 80 per cent of the time, sadness 75 per cent and fear 70 per cent. Least recognisable was “fearfully disgusted,” at 65 per cent.
The researchers are patenting the algorithms and hope they will enable future forms of artificial intelligence to recognise and emulate human emotions.
The paper published in the Proceedings of the National Academy of Sciences.