Machine learning and its subfield deep learning techniques provide opportunities for the development of operator mental state monitoring, especially for cognitive workload recognition using electroencephalogram (EEG) signals. Although a variety of machine learning methods have been proposed for recognizing cognitive workload via EEG recently, there does not yet exist a review that covers in-depth the application of machine learning methods. To alleviate this gap, in this article, we survey cognitive workload and machine learning literature to identify the approaches and highlight the primary advances. To be specific, we first introduce the concepts of cognitive workload and machine learning. Then, we discuss the steps of classical machine learning for cognitive workload recognition from the following aspects, i.e., EEG data preprocessing, feature extraction and selection, classification method, and evaluation methods. Further, we review the commonly used deep learning methods for this domain. Finally, we expound on the open problem and future outlooks.

Depression is the most common psychiatric disorder observed in Parkinson’s disease (PD) patients, however the neural contribution to the high rate of depression in the PD group is still unclear. In this study, we used resting-state functional magnetic resonance imaging (fMRI) to investigate the underlying neural mechanisms of depression in PD patients. Twenty-one healthy individuals and thirty-three patients with idiopathic PD, seventeen of whom were diagnosed with major depressive disorder, were recruited. An analysis of amplitude of low-frequency fluctuations (ALFF) was performed on the whole brain of all subjects. Our results showed that depressed PD patients had significantly decreased ALFF in the dorsolateral prefrontal cortex (DLPFC), the ventromedial prefrontal cortex (vMPFC) and the rostral anterior cingulated cortex (rACC) compared with non-depressed PD patients. A significant positive correlation was found between Hamilton Depression Rating Scale (HDRS) and ALFF in the DLPFC. The findings of changed ALFF in these brain regions implied depression in PD patients may be associated with abnormal activities of prefrontal-limbic network.

We sought to use a regional homogeneity (ReHo) approach as an index in resting-state functional magnetic resonance imaging (fMRI) to investigate the features of spontaneous brain activity within the default mode network (DMN) in patients suffering from bipolar depression (BD). Twenty-six patients with BD and 26 gender-, age-, and education-matched healthy subjects participated in the resting-state fMRI scans. We compared the differences in ReHo between the two groups within the DMN and investigated the relationships between sex, age, years of education, disease duration, the Hamilton Rating Scale for Depression (HAMD) total score, and ReHo in regions with significant group differences. Our results revealed that bipolar depressed patients had increased ReHo in the left medial frontal gyrus and left inferior parietal lobe compared to healthy controls. No correlations were found between regional ReHo values and sex, age, and clinical features within the BD group. Our findings indicate that abnormal brain activity is mainly distributed within prefrontal-limbic circuits, which are believed to be involved in the pathophysiological mechanisms underlying bipolar depression.

Cognitive workload recognition is pivotal to maintain the operator’s health and prevent accidents in the human-robot interaction condition. So far, the focus of workload research is mostly restricted to a single task, yet cross-task cognitive workload recognition has remained a challenge. Furthermore, when extending to a new workload condition, the discrepancy of electroencephalogram (EEG) signals across various cognitive tasks limits the generalization of the existed model. To tackle this problem, we propose to construct the EEG-based cross-task cognitive workload recognition models using domain adaptation methods in a leave-one-task-out cross-validation setting, where we view any task of each subject as a domain. Specifically, we first design a fine-grained workload paradigm including working memory and mathematic addition tasks. Then, we explore four domain adaptation methods to bridge the discrepancy between the two different tasks. Finally, based on the supporting vector machine classifier, we conduct experiments to classify the low and high workload levels on a private EEG dataset. Experimental results demonstrate that our proposed task transfer framework outperforms the non-transfer classifier with improvements of 3% to 8% in terms of mean accuracy, and the transfer joint matching (TJM) consistently achieves the best performance.