Krista Kappeler (semester project)

MSc Semester Project


Extraction of valence and arousal information from EEG signals for emotion classification


Krista Kappeler


Prof. Dr. Touradj Ebrahimi

Assistant: Ashkan Yazdani


June 11, 2010




During the past years, the field of automatic recognition of users affective states has gained a great deal of attention. Automatic, implicit recognition of affective states has many applications, ranging from personalized content recommendation to automatic tutoring systems.
In this project, brain electrical activity will be analyzed and studied to detect different emotions. To this end, the valence-arousal model of emotion will be used. Valence, as used in psychology, especially in discussing emotions, means the intrinsic attractiveness (positive valence) or aversiveness (negative valence) of an event, object, or situation. Arousal is a physiological and psychological state of being awake or reactive to stimuli. It involves the activation of the reticular activating system in the brain stem, the autonomic nervous system and the endocrine system, leading to increased heart rate and blood pressure and a condition of sensory alertness, mobility and readiness to respond. To extract information about these components, electroencephalogram (EEG) signals will be analyzed to extract features that correspond to different levels of valence and arousal.
More specifically, emotional states while watching different still images will be analyzed. These images have been selected from a standard database which aims at eliciting different levels of emotions. The following tasks should be performed:

  • Reviewing the literature about emotion, emotion modeling, EEG signal processing, and the relevant studies on affective assessment based on EEG signals.
  • Studying different methods for preprocessing the EEG signals, feature extraction, regression and classification of biological signals and selecting the methodology.
  • Studying an EEG database, acquired while watching images for emotional assessment.
    • Studying the signal acquisition paradigm.
    • Importing the data into Matlab using EEGLAB toolbox and extraction of single trials.
  • Implementing and validating the selected preprocessing, feature extraction, and classification methods.
  • Assessing and comparing the results of different features extracted from the signal.