International Journal of
Medicine and Medical Sciences

  • Abbreviation: Int. J. Med. Med. Sci.
  • Language: English
  • ISSN: 2006-9723
  • DOI: 10.5897/IJMMS
  • Start Year: 2009
  • Published Articles: 531

Full Length Research Paper

Emotion recognition based on the asymmetric left and right activation

Kwang Shin Park
  • Kwang Shin Park
  • National Rehabilitation Center Research Institute, 142-884, Seoul, Korea.
  • Google Scholar
Hyun Choi
  • Hyun Choi
  • National Rehabilitation Center Research Institute, 142-884, Seoul, Korea.
  • Google Scholar
Kuem Ju Lee
  • Kuem Ju Lee
  • National Rehabilitation Center Research Institute, 142-884, Seoul, Korea.
  • Google Scholar
Jae Yun Lee
  • Jae Yun Lee
  • National Rehabilitation Center Research Institute, 142-884, Seoul, Korea.
  • Google Scholar
Kwang Ok An
  • Kwang Ok An
  • National Rehabilitation Center Research Institute, 142-884, Seoul, Korea.
  • Google Scholar
Eun Ju Kim
  • Eun Ju Kim
  • National Rehabilitation Hospital, Seoul, 142-884, South Korea.
  • Google Scholar


  •  Accepted: 02 June 2011
  •  Published: 30 June 2011

Abstract

 

We obtained EEG (eclectrocephalogram) data from 34 healthy subjects while they were watching emotion-inducing videos and we also developed a real-time emotion monitoring system based on the resulting data. When analyzed from the left and right asymmetric EEG data in each emotion state as compared to the rest state (p<0.1), the alpha wave significantly decreased only at the left temporal lobe for the negative emotion. In particular, the increase of the beta wave was observed only at the left temporal lobe in the fear emotion (p<0.0025), which means inactivation of the left brain. On the other hand, the alpha wave decreased at the C4 in the happy emotion; and in the peaceful state, the gamma wave increased in T5 and the alpha wave decreased in CP5. However, when considering that the number of sample groups was less than seventeen, this result satisfied the independence between the EEG of the frequency band of a certain electrode position of a certain emotion and that of other emotion which is the condition of the emotion recognition algorithm, but it almost does not satisfied the overlap among emotions. However, as this was possible by the additional test, the emotion recognition was considered to be possible. The purpose of the present study; area: systems > emotion was to extract emotion indicators/indices from the EEG signals detected in the human scalp and develop a real-time emotion monitor showing emotional states of people, so that they can express their thoughts and feelings, even if they are passive.

 

Key words: Electroencephalogram, emotion, index, activation, asymmetry.