Attention-based convolutional neural network with multi-modal temporal information fusion for motor imagery EEG decoding.

Convolutional neural network (CNN) has been widely applied in motor imagery (MI)-based brain computer interface (BCI) to decode electroencephalography (EEG) signals. However, due to the limited perceptual field of convolutional kernel, CNN only extracts features from local region without considering long-term dependencies for EEG decoding. Apart from long-term dependencies, multi-modal temporal information is equally important for EEG decoding because it can offer a more comprehensive understanding of the temporal dynamics of neural processes. In this paper, we propose a novel deep learning network that combines CNN with self-attention mechanism to encapsulate multi-modal temporal information and global dependencies. The network first extracts multi-modal temporal information from two distinct perspectives: average and variance. A shared self-attention module is then designed to capture global dependencies along these two feature dimensions. We further design a convolutional encoder to explore the relationship between average-pooled and variance-pooled features and fuse them into more discriminative features. Moreover, a data augmentation method called signal segmentation and recombination is proposed to improve the generalization capability of the proposed network. The experimental results on the BCI Competition IV-2a (BCIC-IV-2a) and BCI Competition IV-2b (BCIC-IV-2b) datasets show that our proposed method outperforms the state-of-the-art methods and achieves 4-class average accuracy of 85.03% on the BCIC-IV-2a dataset. The proposed method implies the effectiveness of multi-modal temporal information fusion in attention-based deep learning networks and provides a new perspective for MI-EEG decoding. The code is available at https://github.com/Ma-Xinzhi/EEG-TransNet.

Multimodal brain-controlled system for rehabilitation training: Combining asynchronous online brain-computer interface and exoskeleton.

BACKGROUND: Traditional therapist-based rehabilitation training for patients with movement impairment is laborious and expensive. In order to reduce the cost and improve the treatment effect of rehabilitation, many methods based on human-computer interaction (HCI) technology have been proposed, such as robot-assisted therapy and functional electrical stimulation (FES). However, due to the lack of active participation of brain, these methods have limited effects on the promotion of damaged nerve remodeling. NEW METHOD: Based on the neurofeedback training provided by the combination of brain-computer interface (BCI) and exoskeleton, this paper proposes a multimodal brain-controlled active rehabilitation system to help improve limb function. The joint control mode of steady-state visual evoked potential (SSVEP) and motor imagery (MI) is adopted to achieve self-paced control and thus maximize the degree of brain involvement, and a requirement selection function based on SSVEP design is added to facilitate communication with aphasia patients. COMPARISON WITH EXISTING METHODS: In addition, the Transformer is introduced as the MI decoder in the asynchronous online BCI to improve the global perception of electroencephalogram (EEG) signals and maintain the sensitivity and efficiency of the system. RESULTS: In two multi-task online experiments for left hand, right hand, foot and idle states, subject achieves 91.25% and 92.50% best accuracy, respectively. CONCLUSION: Compared with previous studies, this paper aims to establish a high-performance and low-latency brain-controlled rehabilitation system, and provide an independent and autonomous control mode of the brain, so as to improve the effect of neural remodeling. The performance of the proposed method is evaluated through offline and online experiments.

Alignment-Based Adversarial Training (ABAT) for Improving the Robustness and Accuracy of EEG-Based BCIs.

Machine learning has achieved great success in electroencephalogram (EEG) based brain-computer interfaces (BCIs). Most existing BCI studies focused on improving the decoding accuracy, with only a few considering the adversarial security. Although many adversarial defense approaches have been proposed in other application domains such as computer vision, previous research showed that their direct extensions to BCIs degrade the classification accuracy on benign samples. This phenomenon greatly affects the applicability of adversarial defense approaches to EEG-based BCIs. To mitigate this problem, we propose alignment-based adversarial training (ABAT), which performs EEG data alignment before adversarial training. Data alignment aligns EEG trials from different domains to reduce their distribution discrepancies, and adversarial training further robustifies the classification boundary. The integration of data alignment and adversarial training can make the trained EEG classifiers simultaneously more accurate and more robust. Experiments on five EEG datasets from two different BCI paradigms (motor imagery classification, and event related potential recognition), three convolutional neural network classifiers (EEGNet, ShallowCNN and DeepCNN) and three different experimental settings (offline within-subject cross-block/-session classification, online cross-session classification, and pre-trained classifiers) demonstrated its effectiveness. It is very intriguing that adversarial attacks, which are usually used to damage BCI systems, can be used in ABAT to simultaneously improve the model accuracy and robustness.

Altered grid-like coding in early blind people.

Cognitive maps in the hippocampal-entorhinal system are central for the representation of both spatial and non-spatial relationships. Although this system, especially in humans, heavily relies on vision, the role of visual experience in shaping the development of cognitive maps remains largely unknown. Here, we test sighted and early blind individuals in both imagined navigation in fMRI and real-world navigation. During imagined navigation, the Human Navigation Network, constituted by frontal, medial temporal, and parietal cortices, is reliably activated in both groups, showing resilience to visual deprivation. However, neural geometry analyses highlight crucial differences between groups. A 60° rotational symmetry, characteristic of a hexagonal grid-like coding, emerges in the entorhinal cortex of sighted but not blind people, who instead show a 90° (4-fold) symmetry, indicative of a square grid. Moreover, higher parietal cortex activity during navigation in blind people correlates with the magnitude of 4-fold symmetry. In sum, early blindness can alter the geometry of entorhinal cognitive maps, possibly as a consequence of higher reliance on parietal egocentric coding during navigation.

Fatigue in children using motor imagery and P300 brain-computer interfaces.

BACKGROUND: Brain-computer interface (BCI) technology offers children with quadriplegic cerebral palsy unique opportunities for communication, environmental exploration, learning, and game play. Research in adults demonstrates a negative impact of fatigue on BCI enjoyment, while effects on BCI performance are variable. To date, there have been no pediatric studies of BCI fatigue. The purpose of this study was to assess the effects of two different BCI paradigms, motor imagery and visual P300, on the development of self-reported fatigue and an electroencephalography (EEG) biomarker of fatigue in typically developing children. METHODS: Thirty-seven typically-developing school-aged children were recruited to a prospective, crossover study. Participants attended three sessions: (A) motor imagery-BCI, (B) visual P300-BCI, and (C) video viewing (control). The motor imagery task involved an imagined left- or right-hand squeeze. The P300 task involved attending to one square on a 3 × 3 grid during a random single flash sequence. Each paradigm had respective calibration periods and a similar visual counting game. Primary outcomes were self-reported fatigue and the power of the EEG alpha band both collected during resting-state periods pre- and post-task. Self-reported fatigue was measured using a 10-point visual analog scale. EEG alpha band power was calculated as the integrated power spectral density from 8 to 12 Hz of the EEG spectrum. RESULTS: Thirty-two children completed the protocol (age range 7-16, 63% female). Self-reported fatigue and EEG alpha band power increased across all sessions (F(1,155) = 33.9, p < 0.001; F = 5.0(1,149), p = 0.027 respectively). No differences in fatigue development were observed between session types. There was no correlation between self-reported fatigue and EEG alpha band power change. BCI performance varied between participants and paradigms as expected but was not associated with self-reported fatigue or EEG alpha band power. CONCLUSION: Short periods (30-mintues) of BCI use can increase self-reported fatigue and EEG alpha band power to a similar degree in children performing motor imagery and P300 BCI paradigms. Performance was not associated with our measures of fatigue; the impact of fatigue on useability and enjoyment is unclear. Our results reflect the variability of fatigue and the BCI experience more broadly in children and warrant further investigation.

Anodal transcranial direct current stimulation does not enhance the effects of motor imagery training of a sequential finger-tapping task in young adults.

When applied over the primary motor cortex (M1), anodal transcranial direct current stimulation (a-tDCS) could enhance the effects of a single motor imagery training (MIt) session on the learning of a sequential finger-tapping task (SFTT). This study aimed to investigate the effect of a-tDCS on the learning of an SFTT during multiple MIt sessions. Two groups of 16 healthy young adults participated in three consecutive MIt sessions over 3 days, followed by a retention test 1 week later. They received active or sham a-tDCS during a MIt session in which they mentally rehearsed an eight-item complex finger sequence with their left hand. Before and after each session, and during the retention test, they physically repeated the sequence as quickly and accurately as possible. Both groups (i) improved their performance during the first two sessions, showing online learning; (ii) stabilised the level they reached during all training sessions, reflecting offline consolidation; and (iii) maintained their performance level one week later, showing retention. However, no significant difference was found between the groups, regardless of the MSL stage. These results emphasise the importance of performing several MIt sessions to maximise performance gains, but they do not support the additional effects of a-tDCS.

Multi-Task Heterogeneous Ensemble Learning-Based Cross-Subject EEG Classification Under Stroke Patients.

Robot-assisted motor training is applied for neurorehabilitation in stroke patients, using motor imagery (MI) as a representative paradigm of brain-computer interfaces to offer real-life assistance to individuals facing movement challenges. However, the effectiveness of training with MI may vary depending on the location of the stroke lesion, which should be considered. This paper introduces a multi-task electroencephalogram-based heterogeneous ensemble learning (MEEG-HEL) specifically designed for cross-subject training. In the proposed framework, common spatial patterns were used for feature extraction, and the features according to stroke lesions are shared and selected through sequential forward floating selection. The heterogeneous ensembles were used as classifiers. Nine patients with chronic ischemic stroke participated, engaging in MI and motor execution (ME) paradigms involving finger tapping. The classification criteria for the multi-task were established in two ways, taking into account the characteristics of stroke patients. In the cross-subject session, the first involved a direction recognition task for two-handed classification, achieving a performance of 0.7419 (±0.0811) in MI and 0.7061 (±0.1270) in ME. The second task focused on motor assessment for lesion location, resulting in a performance of 0.7457 (±0.1317) in MI and 0.6791 (±0.1253) in ME. Comparing the specific-subject session, except for ME on the motor assessment task, performance on both tasks was significantly higher than the cross-subject session. Furthermore, classification performance was similar to or statistically higher in cross-subject sessions compared to baseline models. The proposed MEEG-HEL holds promise in improving the practicality of neurorehabilitation in clinical settings and facilitating the detection of lesions.

Remembering what we imagine: the role of event schemas in shaping how imagined autobiographical events are recalled.

Much like recalling autobiographical memories, constructing imagined autobiographical events depends on episodic memory processes. The ability to imagine events contributes to several future-oriented behaviors (e.g., decision-making, problem solving), which relies, in part, on the ability to remember the imagined events. A factor affecting the memorability of such events is their adherence to event schemas-conceptualizations of how events generally unfold. In the current study, we examined how two aspects of event schemas-event expectancy and familiarity-affect the ability to recall imagined events. Participants first imagined and described in detail autobiographical events that either aligned with or deviated from an event, expected to occur in a context (e.g., a kitchen) that was either familiar or unfamiliar. This resulted in imaginations ranging from maximally schema-congruent (expected events in a familiar context) to maximally novel (unexpected events in an unfamiliar context). Twenty-four hours later, participants recalled these imagined events. Recollections were scored for the number of reinstated details from the imaginations and the number of newly added details. We found greater reinstatement of details for both the maximally congruent and maximally novel events, while maximally novel events were recalled more precisely than other events (i.e., fewer added details). Our results indicate a complementary benefit to remembering schematic and novel imagined events, which may guide equally important but distinct future-oriented behaviors.

[A review on electroencephalogram based channel selection].

The electroencephalogram (EEG) signal is the key signal carrier of the brain-computer interface (BCI) system. The EEG data collected by the whole-brain electrode arrangement is conducive to obtaining higher information representation. Personalized electrode layout, while ensuring the accuracy of EEG signal decoding, can also shorten the calibration time of BCI and has become an important research direction. This paper reviews the EEG signal channel selection methods in recent years, conducts a comparative analysis of the combined effects of different channel selection methods and different classification algorithms, obtains the commonly used channel combinations in motor imagery, P300 and other paradigms in BCI, and explains the application scenarios of the channel selection method in different paradigms are discussed, in order to provide stronger support for a more accurate and portable BCI system.

Evidence for grid-cell-like activity in the time domain.

The relation between the processing of space and time in the brain has been an enduring cross-disciplinary question. Grid cells have been recognized as a hallmark of the mammalian navigation system, with recent studies attesting to their involvement in the organization of conceptual knowledge in humans. To determine whether grid-cell-like representations support temporal processing, we asked subjects to mentally simulate changes in age and time-of-day, each constituting "trajectory" in an age-day space, while undergoing fMRI. We found that grid-cell-like representations supported trajecting across this age-day space. Furthermore, brain regions concurrently coding past-to-future orientation positively modulated the magnitude of grid-cell-like representation in the left entorhinal cortex. Finally, our findings suggest that temporal processing may be supported by spatially modulated systems, and that innate regularities of abstract domains may interface and alter grid-cell-like representations, similarly to spatial geometry.

The effect of episodic specificity inductions on cognitive tasks involving episodic retrieval: A quantitative review.

A growing body of research suggests that an episodic specificity induction (ESI), that is, training in recalled details of a (recent) past event, impacts performance on subsequent tasks that require episodic retrieval processes. The constructive episodic simulation hypothesis (Schacter and Addis, 2007) posits that various tasks which require, at least partially, episodic retrieval processes rely on a single, flexible episodic memory system. As such, a specificity induction activates that episodic memory system and improves subsequent performance on tasks that require use of that memory system. The present quantitative review analyzed the literature demonstrating that the Episodic Specificity Induction (ESI) improves performance on subsequence cognitive tasks that require (at least partial) episodic retrieval processes. Twenty-three studies met criteria for measuring the impact of ESI, compared to a non-specificity control induction(s), on subsequent tasks requiring edpisodic retrieval, including memory, imagination, problem solving, divergent thinking. The results of this review demonstrate a strong, positive effect of ESI on episodic memory, imagination, divergent thinking, and problem-solving tasks.

Slower but more accurate mental rotation performance in aphantasia linked to differences in cognitive strategies.

Mental rotation tasks are frequently used as standard measures of mental imagery. However, aphantasia research has brought such use into question. Here, we assessed a large group of individuals who lack visual imagery (aphantasia) on two mental rotation tasks: a three-dimensional block-shape, and a human manikin rotation task. In both tasks, those with aphantasia had slower, but more accurate responses than controls. Both groups demonstrated classic linear increases in response time and error-rate as functions of angular disparity. In the three-dimensional block-shape rotation task, a within-group speed-accuracy trade-off was found in controls, whereas faster individuals in the aphantasia group were also more accurate. Control participants generally favoured using object-based mental rotation strategies, whereas those with aphantasia favoured analytic strategies. These results suggest that visual imagery is not crucial for successful performance in classical mental rotation tasks, as alternative strategies can be effectively utilised in the absence of holistic mental representations.

Aphantasia and hyperphantasia: exploring imagery vividness extremes.

The vividness of imagery varies between individuals. However, the existence of people in whom conscious, wakeful imagery is markedly reduced, or absent entirely, was neglected by psychology until the recent coinage of 'aphantasia' to describe this phenomenon. 'Hyperphantasia' denotes the converse - imagery whose vividness rivals perceptual experience. Around 1% and 3% of the population experience extreme aphantasia and hyperphantasia, respectively. Aphantasia runs in families, often affects imagery across several sense modalities, and is variably associated with reduced autobiographical memory, face recognition difficulty, and autism. Visual dreaming is often preserved. Subtypes of extreme imagery appear to be likely but are not yet well defined. Initial results suggest that alterations in connectivity between the frontoparietal and visual networks may provide the neural substrate for visual imagery extremes.

Effects of finger pinch motor imagery on short-latency afferent inhibition and corticospinal excitability.

Motor imagery is a cognitive process involving the simulation of motor actions without actual movements. Despite the reported positive effects of motor imagery training on motor function, the underlying neurophysiological mechanisms have yet to be fully elucidated. Therefore, the purpose of the present study was to investigate how sustained tonic finger-pinching motor imagery modulates sensorimotor integration and corticospinal excitability using short-latency afferent inhibition (SAI) and single-pulse transcranial magnetic stimulation (TMS) assessments, respectively. Able-bodied individuals participated in the study and assessments were conducted under two experimental conditions in a randomized order between participants: (1) participants performed motor imagery of a pinch task while observing a visual image displayed on a monitor (Motor Imagery), and (2) participants remained at rest with their eyes fixed on the monitor displaying a cross mark (Control). For each condition, sensorimotor integration and corticospinal excitability were evaluated during sustained tonic motor imagery in separate sessions. Sensorimotor integration was assessed by SAI responses, representing inhibition of motor-evoked potentials (MEPs) in the first dorsal interosseous muscle elicited by TMS following median nerve stimulation. Corticospinal excitability was assessed by MEP responses elicited by single-pulse TMS. There was no significant difference in the magnitude of SAI responses between motor imagery and Control conditions, while MEP responses were significantly facilitated during the Motor Imagery condition compared to the Control condition. These findings suggest that motor imagery facilitates corticospinal excitability, without altering sensorimotor integration, possibly due to insufficient activation of the somatosensory circuits or lack of afferent feedback during sustained tonic motor imagery.

EEG rhythm separation and time-frequency analysis of fast multivariate empirical mode decomposition for motor imagery BCI.

Motor imagery electroencephalogram (EEG) is widely employed in brain-computer interface (BCI) systems. As a time-frequency analysis method for nonlinear and non-stationary signals, multivariate empirical mode decomposition (MEMD) and its noise-assisted version (NA-MEMD) has been widely used in the preprocessing step of BCI systems for separating EEG rhythms corresponding to specific brain activities. However, when applied to multichannel EEG signals, MEMD or NA-MEMD often demonstrate low robustness to noise and high computational complexity. To address these issues, we have explored the advantages of our recently proposed fast multivariate empirical mode decomposition (FMEMD) and its noise-assisted version (NA-FMEMD) for analyzing motor imagery data. We emphasize that FMEMD enables a more accurate estimation of EEG frequency information and exhibits a more noise-robust decomposition performance with improved computational efficiency. Comparative analysis with MEMD on simulation data and real-world EEG validates the above assertions. The joint average frequency measure is employed to automatically select intrinsic mode functions that correspond to specific frequency bands. Thus, FMEMD-based classification architecture is proposed. Using FMEMD as a preprocessing algorithm instead of MEMD can improve the classification accuracy by 2.3% on the BCI Competition IV dataset. On the Physiobank Motor/Mental Imagery dataset and BCI Competition IV Dataset 2a, FMEMD-based architecture also attained a comparable performance to complex algorithms. The results indicate that FMEMD proficiently extracts feature information from small benchmark datasets while mitigating dimensionality constraints resulting from computational complexity. Hence, FMEMD or NA-FMEMD can be a powerful time-frequency preprocessing method for BCI.

Effects of the perceived temporal distance of events on mental time travel and on its underlying brain circuits.

Mental Time Travel (MTT) allows us to remember past events and imagine future ones. According to previous literature, the Temporal Distance of events affects MTT: our ability to order events worsens for close, compared to far, events. However, those studies established distances a-priori, albeit the way we perceive events' temporal distance may subjectively differ from their objective distance. Thus, in the current study, we aimed to investigate the effects of Perceived Temporal Distance (PTD) on the MTT ability and the brain areas mediating this process. Thirty-three healthy volunteers took part in an fMRI MTT task. Participants were asked to project themselves into the past, present, or future, and to judge a series of events as relative-past or relative-future, in relation to the adopted time location. Outside the scanner, participants provided PTD estimates for each stimulus of the MTT task. Participants' performance and functional activity were analyzed as a function of these estimations. At the behavioural level, PTD predicts the modulation of the performance for relative-past and relative-future. Bilateral angular gyrus, retrosplenial cortex, temporo-parietal region and medial, middle and superior frontal gyri mediate the PTD effect. In addition to these areas, the closer the relative-future events are perceived, the higher the involvement of left parahippocampal and lingual gyri and right cerebellum. Thus, perceived proximity of events activates frontal and posterior parietal areas, which therefore might mediate the processing of PTD in the cognitive spatial representation of time. Future proximity also activates cerebellum and medial temporal areas, known to be involved in imaginative and constructive cognitive functions.

Operational encoding enhances action knowledge integration: insights from event-related potential analysis.

In this study, we conducted an examination of knowledge integration concerning action information and assessed the impact of operational on this process. Additionally, we delved into the underlying mechanisms of how operational encoding influences the processing of knowledge integration of action information, utilizing the event-related potential technique. The results of our investigation revealed that operational encoding, encompassing the observed operational encoding and the imagined operational encoding, exhibited superior performance in the integration of action knowledge compared to verbal encoding. This distinction may be attributed to the greater efficiency of operant encoding in activating motor cortical areas, thereby inducing more robust brain activity. These findings suggest the potential advantages of operational encoding in facilitating the integration of knowledge related to movement information at both cognitive and neural levels, underscoring its significant role in the processing of such information. Future studies can further explore the applications of operational encoding in domains, such as motor learning, skill training, and rehabilitation therapy. Such investigations may offer novel insights into enhancing human behavior and motor control.

Multi-scale spatiotemporal attention network for neuron based motor imagery EEG classification.

BACKGROUND: In recent times, the expeditious expansion of Brain-Computer Interface (BCI) technology in neuroscience, which relies on electroencephalogram (EEG) signals associated with motor imagery, has yielded outcomes that rival conventional approaches, notably due to the triumph of deep learning. Nevertheless, the task of developing and training a comprehensive network to extract the underlying characteristics of motor imagining EEG data continues to pose challenges. NEW METHOD: This paper presents a multi-scale spatiotemporal self-attention (SA) network model that relies on an attention mechanism. This model aims to classify motor imagination EEG signals into four classes (left hand, right hand, foot, tongue/rest) by considering the temporal and spatial properties of EEG. It is employed to autonomously allocate greater weights to channels linked to motor activity and lesser weights to channels not related to movement, thus choosing the most suitable channels. Neuron utilises parallel multi-scale Temporal Convolutional Network (TCN) layers to extract feature information in the temporal domain at various scales, effectively eliminating temporal domain noise. RESULTS: The suggested model achieves accuracies of 79.26%, 85.90%, and 96.96% on the BCI competition datasets IV-2a, IV-2b, and HGD, respectively. COMPARISON WITH EXISTING METHODS: In terms of single-subject classification accuracy, this strategy demonstrates superior performance compared to existing methods. CONCLUSION: The results indicate that the proposed strategy exhibits favourable performance, resilience, and transfer learning capabilities.

The effect of self-focused attention during mirror gazing on body image evaluations, appearance-related imagery, and urges to mirror gaze.

BACKGROUND AND OBJECTIVES: Mirror gazing has been linked to poor body image. Cognitive-behavioral models propose that mirror gazing induces self-focused attention. This activates appearance-related imagery, increases body dissatisfaction, and promotes further mirror gazing. However, evidence for these relationships remains scarce. Our study experimentally investigated how self-focused attention impacts overall and facial appearance satisfaction, perceived attractiveness, distress about appearance and disliked features, vividness and emotional quality of appearance-related imagery, and urges to mirror gaze. Baseline body dysmorphic concerns were studied as a moderator. METHODS: Singaporean undergraduates (Mage = 21.22, SDage = 1.62; 35 females, 28 males) were randomly assigned to high or low self-focused attention during a mirror gazing task. Dependent variables were measured with visual analogue scales, and body dysmorphic concerns with the Body Image Disturbance Questionnaire (BIDQ). Analysis of variance and moderation analyses were conducted. RESULTS: Self-focused attention lowered overall and facial appearance satisfaction. Perceived attractiveness decreased only in individuals with high baseline body dysmorphic concerns. Contrary to predictions, distress, appearance-related imagery, and urges to mirror gaze were unaffected. LIMITATIONS: This study used a non-clinical sample. The BIDQ has not been psychometrically validated in Singaporean samples. CONCLUSIONS: Self-focused attention during mirror gazing lowers positive body image evaluations. Individuals with higher body dysmorphic concerns are particularly vulnerable to low perceived attractiveness.

Semi-supervised multi-source transfer learning for cross-subject EEG motor imagery classification.

Electroencephalogram (EEG) motor imagery (MI) classification refers to the use of EEG signals to identify and classify subjects' motor imagery activities; this task has received increasing attention with the development of brain-computer interfaces (BCIs). However, the collection of EEG data is usually time-consuming and labor-intensive, which makes it difficult to obtain sufficient labeled data from the new subject to train a new model. Moreover, the EEG signals of different individuals exhibit significant differences, leading to a significant drop in the performance of a model trained on the existing subjects when directly classifying EEG signals acquired from new subjects. Therefore, it is crucial to make full use of the EEG data of the existing subjects and the unlabeled EEG data of the new target subject to improve the MI classification performance achieved for the target subject. This research study proposes a semi-supervised multi-source transfer (SSMT) learning model to address the above problems; the model learns informative and domain-invariant representations to address cross-subject MI-EEG classification tasks. In particular, a dynamic transferred weighting schema is presented to obtain the final predictions by integrating the weighted features derived from multi-source domains. The average accuracies achieved on two publicly available EEG datasets reach 83.57 % and 85.09 % , respectively, validating the effectiveness of the SSMT process. The SSMT process reveals the importance of informative and domain-invariant representations in MI classification tasks, as they make full use of the domain-invariant information acquired from each subject.