Ohio State University researchers have taught computers how to recognize 21 different human emotions from distinct facial expressions. If you did not realize there are that many feelings in our emotional vocabulary, however, then you may have forgotten about seemingly contradictory emotions, such as "happily disgusted" or "sadly angry."
Experts have been working on decoding the mechanisms that allow our faces to express emotion since at least the time of Aristotle and the Physiognomonics treatise. Cognitive scientists today are interested in tracing the origins of distinct facial expressions back to the genes, chemicals and neural pathways that cause our brains to experience emotion.
However, these scientists have largely restricted their investigations to six primary emotions - happy, sad, fearful, angry, surprised and disgusted. This is because the facial expressions associated with these emotions were considered to be self evident.
But according to co-author of the new study, Aleix Martinez, cognitive scientist and associate professor of electrical and computer engineering at Ohio State, the problem of applying this analytical approach to an individual subject is that it is like painting a portrait with only primary colors. "It can provide an abstracted image of the person," Prof. Martinez claims, "but not a true-to-life one."
What are 'compound emotions?'
By mixing different combinations of emotions and coding their
associated facial expressions, Prof. Martinez has more than tripled the
"palette" of human emotions that can be understood by a computer and,
therefore, examined through "rigorous scientific study."
He refers to these combined emotions as "compound
emotions."
Prof. Martinez and colleagues photographed 130 female and 100
male volunteers who were asked to provide appropriate facial expressions to
verbal cues such as "you just got some great unexpected news" or
"you smell a bad odor."
The facial muscles used
prominently in each of the 5,000 resulting photographs were then tagged
according to the Facial Action Coding System (FACS) created by psychologist
Paul Ekman. Using the FACS data, the researchers were able to run searches
cross-referencing similarities and differences in expressions.
Near-universal
expression of different emotions
From these distinct expressions, Prof. Martinez and team
recorded a total of 21 emotions. Many of these were expressed by participants
in an almost universal way - 99% of the volunteers depicted happiness by
stretching their mouths into a smile, for example.
The "compound emotions" were also generally expressed
using the same facial expressions. About 93% of the participants made the same
face for "happily surprised," which involved opening their eyes wide
and raising their cheeks in a combination of a smile and a surprised face.
Even more contradictory compound emotions, such as "happily
disgusted" - defined by Prof. Martinez as "how you feel when you
watch one of those funny 'gross-out' movies and something happens that's really
disgusting, but you just have to laugh because it's so incredibly funny" -
had a universal expression. In this case: scrunched up eyes and nose, but with
a smile.
Prof. Martinez believes
this cognitive research could have some therapeutic applications. For instance,
he explains, the model may be used in treatment of conditions like autism or post-traumatic stress
disorder(PTSD):
"For example, if in PTSD people are more attuned to anger
and fear, can we speculate that they will be tuned to all the compound emotions
that involve anger or fear, and perhaps be super-tuned to something like
'angrily fearful'? What are the pathways, the chemicals in the brain that
activate those emotions? We can make more hypotheses now and test them. Then,
eventually, we can begin to understand these disorders much better and develop
therapies or medicine to alleviate them."
Recently, Medical News Today reported on research into a computer
model that could predict with 85% accuracy whether people
were lying or not based on their facial expressions.
Written by David McNamee
Source:
0 comments:
Post a Comment