Neurofeedback training protocols based on spectral EEG feature subset and channel selection for performance enhancement of novice visual artists. (May 2018)

• Record Type: Journal Article
• Authors: Shourie, Nasrin; Firoozabadi, Mohammad; Badie, Kambiz
• Is Part Of: Biomedical signal processing and control. Volume 43(2018)
• Page Start: 117
• Page End: 129
• Publication Date: 2018-05
• Subjects: Neurofeedback -- Artist -- Genetic algorithm -- Visual perception -- Mental imagery; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2018.02.017
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Neurofeedback training protocols are proposed for performance improvement of visual artists. The effective discriminative EEG features were selected using genetic algorithm. Abstract: No extensive research has been designed to study neurofeedback training protocol to improve performance of novice visual artists although many neurofeedback protocols have been designed to enhance the performance of novices in the other domains. In this article, differences between multichannel EEG signals of visual artists and non-artists are investigated during visual perception and mental imagery of some paintings using PSD (power spectral density) analysis to propose some neurofeedback training protocols for performance enhancement of novice visual artists. In this approach, the relative power values of the EEG signals are calculated in the traditional frequency subbands (delta, theta, alpha, beta and gamma). Genetic algorithm is used to select effective discriminative features and their related appropriate channels. Fitness function of the genetic algorithm is defined based on Davies-Bouldin index and statistical Mann-Whitney U test. Therefore, the relative low beta enhancement in Fp1 and F7 is proposed as the neurofeedback training protocol for visual perception performance improvement of novice artists. The relative low beta increase in Fp1 is also proposed as the neurofeedback training protocol for enhancing mental imagery performance of the novices. Since, the brain frontal region is closer to the sources of ocular artifacts as compared to the other brain regions, two neurofeedback training protocols are also proposed in the other brain regions (parietal, central, temporal and occipital). The alpha/theta decrease in T6 and the relative low beta enhancement in T5 are also proposed for visual perception and mental imagery performance improvements of the novices, respectively. It is also recommended that the proposed neurofedback training protocol for mental imagery performance enhancement is verified to the relative low beta increase and the alpha decrease in T5 to ensure the proposed protocol provides the necessary spectral changes in the novices.