Evaluation of EEG Signals by Spectral Peak Methods and Statistical Correlation for Mental State Discrimination Induced by Arithmetic Tasks

Abstract

Bringing out brain activity through the interpretation of EEG signals is a challenging problem that involves combined methods of signal analysis. The issue of classifying mental states induced by arithmetic tasks can be solved through various classification methods, using diverse characteristic parameters of EEG signals in the time, frequency, and statistical domains. This paper explores the results of an experiment that aimed to highlight arithmetic mental tasks contained in the PhysioNet database, performed on a group of 36 subjects. The majority of publications on this topic deal with machine learning (ML)-based classification methods with supervised learning support vector machine (SVM) algorithms, K-Nearest Neighbor (KNN), Linear Discriminant Analysis (LDA), and Decision Trees (DTs). Also, there are frequent approaches based on the analysis of EEG data as time series and their classification with Recurrent Neural Networks (RNNs), as well as with improved algorithms such as Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (BLSTM), and Gated Recurrent Units (GRUs). In the present work, we evaluate the classification method based on the comparison of domain limits for two specific characteristics of EEG signals: the statistical correlation of pairs of signals and the size of the spectral peak detected in theta, alpha, and beta bands. This study provides some interpretations regarding the electrical activity of the brain, consolidating and complementing the results of similar research. The classification method used is simple and easy to apply and interpret. The analysis of EEG data showed that the theta and beta frequency bands were the only discriminators between the relaxation and arithmetic calculation states. Notably, the F7 signal, which used the spectral peak criterion, achieved the best classification accuracy (100%) in both theta and beta bands for the subjects with the best results in performing calculations. Also, our study found the Fz signal to be a good sensor in the theta band for mental task discrimination for all subjects in the group with 90% accuracy.

Keywords: EEG signal processing; mental states discrimination; power spectral density; statistical correlation.

Figure 1

Correspondence of brain areas of interest with EEG measurement points. (a) Schematic circuit of neurocognitive nodes in arithmetic process, adapted from [1]. (b) Electrode placement in 10–20 system, colored according to the cognitive nodes involved in the arithmetic process, adapted from [25].

Figure 2

Spectral density in three bands and recorded spectral peaks.

Figure 3

Algorithms implemented for state discrimination: (a) scheme of the evaluation principle based on correlation; (b) scheme with the principle of evaluation by spectral maxima.

Figure 4

Matrices with the number of subjects fulfilling the correlation threshold condition for group G in (a,c,e) theta, alpha, and beta bands with c_corr R > 0.5 and in (b,d,f) theta, alpha, and beta bands with c_corr T < 0.5, where green and blue mark the extreme values in the matrices.

Figure 5

Matrices with the number of subjects fulfilling the correlation threshold condition for group B in (a,c,e) theta, alpha, and beta bands with c_corr R > 0.5; (b,d,f) theta, alpha, and beta bands with c_corr T < 0.5, where green and blue mark the extreme values in the matrices.

Figure 6

Spectral peaks for the F7 signal in the theta band of subjects from group G.

Figure 7

Spectral peaks for the F7 signal in the beta band of subjects from group G.

Figure 8

Spectral peaks for the Fz signal in the theta band of subjects from group G.

Figure 9

Ranges of variation in the size of the spectral peak theta for subjects in group “G*” with score > 26 on the mental test. Circles indicate the median of the ranges, and the black dashed lines the non-overlapping ranges.

Figure 10

Ranges of variation in the size of the spectral peak alpha band for subjects in group “G*” with scores > 26 on the mental test, where circles indicate the median of the ranges.

Figure 11

Ranges of variation in the size of the spectral peak in beta band for subjects in group “G*” with scores > 26 on the mental test. Circles indicate the median of the ranges, and the black dashed lines the non-overlapping ranges.

Figure 12

Ranges of variation in the size of the spectral peak in the theta band for subjects of group “B”, where circles indicate the median of the ranges, and the black dashed lines the non-overlapping ranges.

Figure 13

Ranges of variation in the size of the spectral peak in the beta band for subjects of group “B”, where circles indicate the median of the ranges, and the black dashed lines the non-overlapping ranges.