Farshad Rafiei - CIPCE, Control and Intelligent Processing Center of Excellence, School of Electrical and Computer, Engineering, College of Engineering, University of Tehran, Tehran, Iran
Gholam-Ali Hossein-Zadeh - CIPCE, Control and Intelligent Processing Center of Excellence, School of Electrical and Computer, Engineering, College of Engineering, University of Tehran, Tehran, Iran

In a brain decoding study, using the functional magnetic resonance imaging (fMRI) data we determined the facial expression of the visual stimulus that the subject perceived. fMRI data acquired from a healthy right-handed adult volunteer who participated in three separate sessions. Participant viewed blocks of emotionally expressive faces alternating with blocks of neutral faces and scrambled images. Multi-voxel pattern analyses are then used to decode different expressions using the activity pattern of most active parts of brain. We used multi-class support vector machine (SVM) to distinct five brain states corresponding to neutral, happy, sad, angry and surprised. Results show that these facial expressions can be classified from fMRI data with the average sensitivity of 90 percent.

Brain decoding, facial expressions, Multi-voxel pattern analysis, Support vector machines