Objectives: The temporal and spatial information of electroencephalogram (EEG) signals is crucial for recognizing features in emotion classification models, but it excessively relies on manual feature extraction. The transformer model has the capability of performing automatic feature extraction; however, its potential has not been fully explored in the classification of emotion-related EEG signals. To address these challenges, the present study proposes a novel model based on transformer and convolutional neural networks (TCNN) for EEG spatial–temporal (EEG ST) feature learning to automatic emotion classification. Methods: The proposed EEG ST-TCNN model utilizes position encoding (PE) and multi-head attention to perceive channel positions and timing information in EEG signals. Two parallel transformer encoders in the model are used to extract spatial and temporal features from emotion-related EEG signals, and a CNN is used to aggregate the EEG’s spatial and temporal features, which are subsequently classified using Softmax. Results: The proposed EEG ST-TCNN model achieved an accuracy of 96.67% on the SEED dataset and accuracies of 95.73%, 96.95%, and 96.34% for the arousal–valence, arousal, and valence dimensions, respectively, for the DEAP dataset. Conclusions: The results demonstrate the effectiveness of the proposed ST-TCNN model, with superior performance in emotion classification compared to recent relevant studies. Significance: The proposed EEG ST-TCNN model has the potential to be used for EEG-based automatic emotion recognition.

Brain–computer interfaces (BCIs) are a new technology that subverts traditional human–computer interaction, where the control signal source comes directly from the user’s brain. When a general BCI is used for practical applications, it is difficult for it to meet the needs of different individuals because of the differences among individual users in physiological and mental states, sensations, perceptions, imageries, cognitive thinking activities, and brain structures and functions. For this reason, it is necessary to customize personalized BCIs for specific users. So far, few studies have elaborated on the key scientific and technical issues involved in personalized BCIs. In this study, we will focus on personalized BCIs, give the definition of personalized BCIs, and detail their design, development, evaluation methods and applications. Finally, the challenges and future directions of personalized BCIs are discussed. It is expected that this study will provide some useful ideas for innovative studies and practical applications of personalized BCIs.

Stroke is a debilitating neurological condition which usually results in the abnormal electrical brain activity and the impairment of sensation, motor, or cognition functions. In this context, neurofeedback training, i.e., a non-invasive and relatively low cost technique that contributes to neuroplasticity and behavioral performance, might be promising for stroke rehabilitation. We intended to explore neurofeedback training on a 63-year-old male patient and a 77-year-old female patient with chronic stroke. Both of them had suffered from an ischemic stroke for rather long period (more than 3 years) and could not gain further improvement by traditional therapy. The neurofeedback training was designed to enhance alpha activity by 15 sessions distributed over 2 months, for the purpose of overall cognitive improvement and hopefully also motor function improvement for the female patient. We found that the two patients showed alpha enhancement during NFT compared to eyes open baseline within most sessions. Furthermore, both patients reduced their anxiety and depression level. The male patient showed an evolution in speech pattern in terms of naming, sentences completion and verbal fluency, while the female patient improved functionality of the march. These results suggested that alpha neurofeedback training could provide a spectrum of improvements, providing new hope for chronic stroke patients who could not gain further improvements through traditional therapies.

Human alpha oscillation (7–13 Hz) has been extensively studied over the years for its connection with cognition. The individual alpha frequency (IAF), defined as the frequency that provides the highest power in the alpha band, shows a positive correlation with cognitive processes. The modulation of alpha activities has been accomplished through various approaches aimed at improving cognitive performance. However, very few studies focused on the direct modulation of IAF by shifting the peak frequency, and the understanding of IAF modulation remains highly limited. In this study, IAFs of healthy young adults were up-regulated through short-term neurofeedback training using haptic feedback. The results suggest that IAFs have good trainability and are up-regulated, also that IAFs are correlated with the enhanced cognitive performance in mental rotation and n-back tests compared to sham-neurofeedback control. This study demonstrates the feasibility of self-regulating IAF for cognition enhancement and provides potential therapeutic benefits for cognitive-impaired patients.

Neurofeedback (NF) training is a closed-loop brain training in which participants learn to regulate their neural activation. NF training of alpha (8–12 Hz) activity has been reported to enhance working memory capacity, but whether it affects the precision in working memory has not yet been explored. Moreover, whether NF training distinctively influences performance in different types of working memory tasks remains unclear. Therefore, the present study conducted a randomized, single-blind, sham-controlled experiment to investigate how alpha NF training affected the capacity and precision of working memory, as well as the related neural change. Forty participants were randomly and equally assigned to the NF group and the sham control group. Both groups received NF training (about 30 min daily) for five consecutive days. The NF group received alpha (8–12 Hz) training, while the sham control group received sham NF training. We found a significant alpha increase within sessions but no significant difference across sessions. However, the behavioral performance and neural activity in the modified Sternberg task did not show significant change after alpha NF training. On the contrary, the alpha NF training group significantly increased visual working memory capacity measured by the Corsi-block tapping task and improved visual working memory precision in the interference condition in a color-recall task. These results suggest that alpha NF training influences performance in working memory tasks involved in the visuospatial sketchpad. Notably, we demonstrated that alpha NF training improves the quantity and quality of visual working memory.