Neurimaging Study Essay

485 Words1 Page

The neuroimaging study conducted by Karim S. Kassam et al involved 10 actors entering the state of nine emotions, anger, disgust, envy, fear, happiness, lust, pride, shame and sadness, in a randomised order whilst having their brain activity scanned using a fMRI. The experiment was comprised of a series of phases, the first being that the actors were told to enter each of the listed emotional states whilst attached to the fMRI. In the second phase of the experiment, the actors were presented with both neutral and disgusting images that they had not previously seen. Along with the use of a fMRI scanner, a computer model was also used in order to read the brain activity as shown from the fMRI scans. During the second phase of the study, the computer …show more content…

The final phase saw the researchers applying the machine learning analysis of neural activation patterns from each actor but one, in order to make a prediction of which emotions were to be experienced by the actor that wasn’t included. When the computer model was tested for its accuracy of predictions and results it achieved a rank accuracy of 0.91. Rank accuracy is the percentile rank of correct emotion in an ordered list of the guesses made by the computer model; guessing achieves a rank accuracy of 0.50. The results from the study found that from the nine emotions to select from, the computer model listed disgust as the most likely emotion 60% of the time and it was a part of the top two guesses 80% of the time. On average the computer model ranked the correct emotion as highest among all of its guesses; the computer model was most accurate at identifying happiness and least accurate at identifying envy. The computer model rarely confused positive and negative emotions which suggests that there are distinct neural signatures that make it difficult to confuse the

Open Document