Effects of Mental Workload Manipulation on Electroencephalography Spectrum Oscillation and Microstates in Multitasking Environments

Abstract

Introduction: Multitasking during flights leads to a high mental workload, which is detrimental for maintaining task performance. Electroencephalography (EEG) power spectral analysis based on frequency-band oscillations and microstate analysis based on global brain network activation can be used to evaluate mental workload. This study explored the effects of a high mental workload during simulated flight multitasking on EEG frequency-band power and microstate parameters.

Methods: Thirty-six participants performed multitasking with low and high mental workloads after 4 consecutive days of training. Two levels of mental workload were set by varying the number of subtasks. EEG signals were acquired during the task. Power spectral and microstate analyses were performed on the EEG. The indices of four frequency bands (delta, theta, alpha, and beta) and four microstate classes (A-D) were calculated, changes in the frequency-band power and microstate parameters under different mental workloads were compared, and the relationships between the two types of EEG indices were analyzed.

Results: The theta-, alpha-, and beta-band powers were higher under the high than under the low mental workload condition. Compared with the low mental workload condition, the high mental workload condition had a lower global explained variance and time parameters of microstate B but higher time parameters of microstate D. Less frequent transitions between microstates A and B and more frequent transitions between microstates C and D were observed during high mental workload conditions. The time parameters of microstate B were positively correlated with the delta-, theta-, and beta-band powers, whereas the duration of microstate C was negatively correlated with the beta-band power.

Conclusion: EEG frequency-band power and microstate parameters can be used to detect a high mental workload. Power spectral analyses based on frequency-band oscillations and microstate analyses based on global brain network activation were not completely isolated during multitasking.

Keywords: EEG microstate; band power; mental workload; multitasking.

FIGURE 1

Experiment protocol. Tasks 1 and 2 represent two conditions of mental workload. NASA‐TLX, National Aeronautics and Space Administration‐task load index.

FIGURE 2

Task performance at different training times. (A) Average distance, (B) meter response time, (C) dot response time, and (D) number of numeral responses. The error bars represent the standard deviation. *p < 0.05, ***p < 0.001.

FIGURE 3

NASA‐TLX scores in different mental workloads. (A) Total scores and (B) scores of six dimensions. The lower and upper ends of the box represent the first quartile (Q1) and third quartile (Q3), respectively, and the difference between Q3 and Q1 is the interquartile range (IQR). The horizontal lines and white squares inside the box represent the median and mean values. The lower and upper ends of the error bars are Q1 − 1.5IQR and Q3 + 1.5IQR. *p < 0.05, **p < 0.01, ***p < 0.001. NASA‐TLX, National Aeronautics and Space Administration‐task load index.

FIGURE 4

Frequency‐band power in different mental workloads. (A) Topographies of frequency‐band power, (B) statistical graph of total average spectra power, and (C) distribution graph of statistical values. The lower and upper ends of the box represent the first quartile (Q1) and third quartile (Q3), respectively, and the difference between Q3 and Q1 is the interquartile range (IQR). The horizontal lines and white squares inside the box represent the median and mean values. The lower and upper ends of the error bars are Q1 − 1.5IQR and Q3 + 1.5IQR. **p < 0.01, ***p < 0.001.

FIGURE 5

Topographies of the four classes of microstates in different conditions.

FIGURE 6

Microstate temporal parameters results. (A) The global explained variance, (B) the mean duration, (C) the occurrence, and (D) the coverage. The lower and upper ends of the box represent the first quartile (Q1) and third quartile (Q3), respectively, and the difference between Q3 and Q1 is the interquartile range (IQR). The horizontal lines and white squares inside the box represent the median and mean values. The lower and upper ends of the error bars are Q1 − 1.5IQR and Q3 + 1.5IQR. *p < 0.05, **p < 0.01.

FIGURE 7

Microstate syntax results. (A) Transition probabilities in the low mental workload condition, (B) transition probabilities in the high mental workload condition, (C) distribution graph of statistical values, and (D) the significant transition probabilities between the two conditions. *p < 0.05.

FIGURE 8

Scatterplot with regression lines. Regression plots of the correlation between the (A) duration of microstate B and delta‐band power, (B) duration of microstate B and theta‐band power, (C) duration of microstate C and beta‐band power, and (D) occurrence of microstate B and beta‐band power.