Selection of Effective EEG Band for Estimation of Cognitive State while Listening Music

Monira Islam; Md. Salah Uddin Yusuf; Mohiuddin Ahmad

A new approach is represented in this paper in order to estimate the cognitive states depending on the variations of different effective frequency bands of EEG signal while listening music. In this work, EEG signals are collected for different types of songs to analyze the musical activity on human brain. Different types of songs have specific frequency which can either resonate or anti- resonate with human brain. Spatial and temporal analysis are applied to the brain signals to extract the salient features for estimation of human cognitive state. The extracted features are trained with SVM classifier to show the effectiveness of different EEG bands with different songs. The classification accuracy of different frequency bands indicates the larger alpha or beta activity on human brain for the selected songs. The highest classification accuracy for soft song was found for θ band which indicates the state of catnap. For medium and fast song the effectiveness of α band and β band are noteworthy, which indicate the state of relax and stress respectively. Therefore, human cognitive states can be estimated with our proposed approach which is innovative and efficient to show the impact of music on human brain activity.

Select Volume / Issues:

• Volume 6,Issue 6, Nov. 2015

Year:

2015

Type of Publication:

Article

Keywords:

Brain Activity; Cognitive State; EEG; Frequency Band; Music Signal

Journal:

IJECCE

Volume:

6

Number:

6

Pages:

719-726

Month:

November