Identifying human emotions based on brain activity (video)
For the first time, scientists at Carnegie Mellon University (CMU) have identified which emotion a person is experiencing based on brain activation.
Until now determining how a human being feels or detecting their emotion objectively has been a dream, but researchers from Dietrich College of Humanities of CMU combined functional magnetic resonance imaging (fMRI) and machine learning to measure brain signals to read people’s emotions.
The experiment represented human emotions as heat maps on a brain image as shown in the picture. Therefore, the purpose of the study is to introduce a new method to identify emotion through fMRI without the need to ask for self-report data given by the people.
In the studies 10 actors took part. Their task was to view words of nine different emotions: happiness, pride, sadness, shame, lust, envy, disgust, fear, anger.
While the actors where inside the fMRI scanner, the experimenters instructed them to enter into the emotional states various times. The study was made mainly with actors as they have experience going through different emotional states. According to the researchers of CMU their studies suggest that emotion is not just detected from the amygdala, but it is scattered throughout different regions of the brain.
In result of this tremendous discovery Carnegie Mellon University has recently launched a Brain, Mind and Learning initiative to continue working on problems such as this study.
If you would like to find more information about the study read their paper entitled “Identifying Emotions on the Basis of Neural Activation” published in PLOS One.