Advances in Reading Emotions

In January 2014 researchers at Glasgow University’s Institute of Neuroscience and Psychology questioned the long-held theory that human have six basic emotions.[1] Research scientists Rachael E. Jack, Oliver G. B. Garrod, and Philippe G. Schyns argued that children initially show four biologically-based emotions. With social experience they develop two more. The initial four instinctive emotional expressions are fear/surprise, sadness, happiness, and anger/disgust. Later surprise and disgust are signalled as separate emotions.

In this theory there are only four basic or innate emotions, not the usual six identified by psychologists.

In March 2014 other researchers developed an algorithm that allows computers to read human emotions, even complex ones formed by blending the primary facial expressions. Shichuan Du, Yong Tao, and Aleix M. Martinez at Ohio State University, went a long way to proving a theory originally developed by Austrian ethologist Konrad Lorenz, that facial expressions can be blended.[2]

The team had their computer code the facial expressions of 5,000 photographs of 230 people. The result was 15 compound emotions, including “sadly angry”, “fearfully surprised”, and “happily disgusted”. All are distinct blends of the basic six expressions. And all are formed by contractions of muscles in the face specialised for form the patterns we call emotional expressions.

These studies are the latest in a long series of exhaustive research stretching back to Charles Darwin’s ground breaking treatise, The Expression of the Emotions in Man and Animals, published in 1872.

Studies such as the two reviewed here bring the day closer when computers – personal computers – will be able to read the emotions of those using them. The new methods might also help diagnose and study such conditions as schizophrenia, autism and trauma.

However, computers are still not nearly as clever as children or even dogs, who are evolved to detect human emotions. By 18 months of age, toddlers can tell when an adult is faking a smile. Dogs are better judges of human emotions than are chimpanzees.

People are generally good at detecting emotions, though women are generally more adept than men. What we lack is a theory of social situations able to interpret our observations. Such a theory, to be reviewed in future posts, is needed to explain which emotions are aroused by which social situations.[3]

Frank Salter, 1st April 2014


[1] “Dynamic Facial Expressions of Emotion Transmit an Evolving Hierarchy of Signals over Time”, Current Biology., accessed 1 April 2014.

[2] “Compound facial expressions of emotion”, Proceedings of the National Academy of Sciences (USA).

[3] Salter, F. K. (2008/1995). Emotions in command: Biology, bureaucracy, and cultural evolution. New York, Transaction.


Comments are closed.