Detection of Sustained Auditory Attention in Students with Visual Impairment

Recent advances in auditory attention detection from multichannel electroencephalography (EEG) signals encounter the challenges of the scarcity of available online EEG data and the detection of auditory attention with low latency. To this end, we propose a complete deep auditory generative adversarial network auxiliary, named auditory-GAN, designed to handle these challenges while generating EEG data and executing auditory spatial detection. The proposed auditory-GAN system consists of a spectro-spatial feature extraction (SSF) module and an auditory generative adversarial network auxiliary (AD-GAN) classifier. The SSF module extracts the spatial feature maps by learning the topographic specificity of alpha power from EEG signals. The designed AD-GAN network addresses the need for extensive training data by synthesizing augmented versions of original EEG data. We validated the proposed method on the widely used KUL dataset. The model assesses the quality of generated EEG images and the accuracy of auditory spatial attention detection. Results show that the proposed auditory-GAN can produce convincing EEG data and achieves a significant i.e., 98.5% spatial attention detection accuracy for a 10-s decision window of 64-channel EEG data. Comparative analysis reveals that the proposed neural approach outperforms existing state-of-the-art models across EEG data ranging from 64 to 32 channels. The Auditory-GAN model is available at https://github.com/tasleem-hello/Auditory-GAN-/tree/Auditory-GAN.

DGSD: Dynamical graph self-distillation for EEG-based auditory spatial attention detection

Rapid progress in information and communication technologies (ICTs) has fueled the popularity of e-learning for educational purposes. However, an e-learning environment is limited in that online instructors cannot monitor immediately whether students remain focus during online autonomous learning. Therefore, this study develops a novel attention aware system (AAS) capable of recognizing students' attention levels accurately based on EEG signals, thus having high potential to be applied in providing timely alert for conveying low-attention level feedback to online instructors in an e-learning environment. To construct AAS, attention responses of students and their corresponding EEG signals are gathered on a continuous performance test (CPT), i.e. An attention assessment test. Next, the AAS is constructed by using training and testing data by the NeuroSky brainwave detector and the support vector machine (SVM), a well-known machine learning model. Additionally, based on the discrete wavelet transform (DWT), the collected EEG signals are decomposed into five primary bands (i.e. Alpha, beta, gamma, theta and delta) as well as each primary band contains five statistical parameters (including approximate entropy, total variation, energy, skewness and standard deviation), thus generating twenty five potential brainwave features associated with students' attention level for constructing the AAS. An attempt based on genetic algorithm (GA) is also made to enhance the prediction performance of the proposed AAS in terms of identifying students' attention levels. According to GA, the seven most influential features are selected from twenty-five considered features, parameters of the proposed AAS are optimized as well. Analytical results indicate that the proposed AAS can accurately recognize individual student's attention state as either a high or low level, and the average accuracy rate reaches as high as 90.39 %. Moreover, the proposed AAS is integrated with a video lecture tagging system to examine whether the proposed AAS can accurately detect students' low-attention periods while learning about electrical safety in the workplace via a video lecture. An experiment is designed to assess the prediction performance of the proposed AAS in terms of identifying the periods of video lecture with high-or low-attention levels during learning processes. Analytical results indicate that the proposed AAS can accurately identify the low-attention periods of video lecture generated by students when engaging in a learning activity with video lecture. Results of this study demonstrate that the proposed AAS is an effective attention aware system, capable of assisting online instructors in evaluating students' attention levels to enhance their online learning performance.

Existing research has revealed that auditory attention can be tracked from ongoing electroencephalography (EEG) signals. The aim of this novel study was to investigate the identification of peoples’ attention to a specific auditory object from single-trial EEG signals via entropy measures and machine learning. Approximate entropy (ApEn), sample entropy (SampEn), composite multiscale entropy (CmpMSE) and fuzzy entropy (FuzzyEn) were used to extract the informative features of EEG signals under three kinds of auditory object-specific attention (Rest, Auditory Object1 Attention (AOA1) and Auditory Object2 Attention (AOA2)). The linear discriminant analysis and support vector machine (SVM), were used to construct two auditory attention classifiers. The statistical results of entropy measures indicated that there were significant differences in the values of ApEn, SampEn, CmpMSE and FuzzyEn between Rest, AOA1 and AOA2. For the SVM-based auditory attention classifier, the auditory object-specific attention of Rest, AOA1 and AOA2 could be identified from EEG signals using ApEn, SampEn, CmpMSE and FuzzyEn as features and the identification rates were significantly different from chance level. The optimal identification was achieved by the SVM-based auditory attention classifier using CmpMSE with the scale factor τ = 10. This study demonstrated a novel solution to identify the auditory object-specific attention from single-trial EEG signals without the need to access the auditory stimulus.

The performance of university students during their academic session are vital to their overall grade throughout their term in the university. There are multiple factors that could lead to the loss of performance but the foremost factor is their level of emotions. Previous research has shown that to determine the performance of the students, the best way to do so is by analysing their attention levels. With the development of portable Electroencephalogram (EEG) devices and machine learning algorithms, it is easy to obtain the students attention and emotion level during their academic sessions. This paper aims to present a method of obtaining the EEG signals using a portable EEG device and classifying it into the type of emotions that are present in the human brain. The EEG device will obtain the attention level and EEG signals during two scenarios which are lectures/tutorials and exams/quizzes. The signals are then compiled and analysed to determine the emotion labels based on a normalization process that categories the signals into positive or negative emotions. The dataset and labels are then used to train and evaluate multiple machine learning models and a deep learning model in order to determine which model has the best accuracy and performance. The chosen model is then used to predict the emotions of several students during both scenarios and the average emotions are then compared with their average attention to determine the effect of emotions on the students’ performance. Hence, this paper will first provide a method on obtaining the emotion labels, followed by the models’ development and finally correlating the predicted emotions with the students’ performance during their academic sessions.

Rapid progress in information and communication technologies ( ICTs ) has fueled the popularity of e‐learning. However, an e‐learning environment is limited in that online instructors cannot monitor immediately whether students remain focus during online autonomous learning. Therefore, this study tries to develop a novel attention aware system ( AAS ) capable of recognizing students' attention levels accurately based on electroencephalography ( EEG ) signals, thus having high potential to be applied in providing timely alert for conveying low‐attention level feedback to online instructors in an e‐learning environment. To construct AAS , attention responses of students and their corresponding EEG signals are gathered based on a continuous performance test ( CPT ), ie, an attention assessment test. Next, the AAS is constructed by using training and testing data by the N euro S ky brainwave detector and the support vector machine ( SVM ), a well‐known machine learning model. Additionally, based on the discrete wavelet transform ( DWT ), the collected EEG signals are decomposed into five primary bands (ie, alpha, beta, gamma, theta, and delta). Each primary band contains five statistical parameters (including approximate entropy, total variation, energy, skewness, and standard deviation), thus generating 25 potential brainwave features associated with students' attention level for constructing the AAS . An attempt based on genetic algorithm ( GA ) is also made to enhance the prediction performance of the proposed AAS in terms of identifying students' attention levels. According to GA , the seven most influential features are selected from 25 considered features; parameters of the proposed AAS are also optimized. Analytical results indicate that the proposed AAS can accurately recognize individual student's attention state as either a high or low level, and the average accuracy rate reaches as high as 89.52%. Moreover, the proposed AAS is integrated with a video lecture tagging system to examine whether the proposed AAS can accurately detect students' low‐attention periods while learning about electrical safety in the workplace via a video lecture. Four experiments are designed to assess the prediction performance of the proposed AAS in terms of identifying the periods of video lecture with high‐ or low‐attention levels during learning processes. Analytical results indicate that the proposed AAS can accurately identify the low‐attention periods of video lecture generated by students when engaging in a learning activity with video lecture. Meanwhile, the proposed AAS can also accurately identify the low‐attention periods of video lecture generated by students to some degree even when students engage in a learning activity by a video lecture with random disturbances. Furthermore, strong negative correlations are found between the students' learning performance (ie, posttest score and progressive score) and the low‐attention periods of video lecture identified by the proposed AAS . Results of this study demonstrate that the proposed AAS is effective, capable of assisting online instructors in evaluating students' attention levels to enhance their online learning performance.

Despite recent advances, the decoding of auditory attention from brain signals remains a challenge. A key solution is the extraction of discriminative features from high-dimensional data, such as multi-channel electroencephalography (EEG). However, to our knowledge, topological relationships between individual channels have not yet been considered in any study. In this work, we introduced a novel architecture that exploits the topology of the human brain to perform auditory spatial attention detection (ASAD) from EEG signals. We propose EEG-Graph Net, an EEG-graph convolutional network, which employs a neural attention mechanism. This mechanism models the topology of the human brain in terms of the spatial pattern of EEG signals as a graph. In the EEG-Graph, each EEG channel is represented by a node, while the relationship between two EEG channels is represented by an edge between the respective nodes. The convolutional network takes the multi-channel EEG signals as a time series of EEG-graphs and learns the node and edge weights from the contribution of the EEG signals to the ASAD task. The proposed architecture supports the interpretation of the experimental results by data visualization. We conducted experiments on two publicly available databases. The experimental results showed that EEG-Graph Net significantly outperforms the state-of-the-art methods in terms of decoding performance. In addition, the analysis of the learned weight patterns provides insights into the processing of continuous speech in the brain and confirms findings from neuroscientific studies. We showed that modeling brain topology with EEG-graphs yields highly competitive results for auditory spatial attention detection. The proposed EEG-Graph Net is more lightweight and accurate than competing baselines and provides explanations for the results. Also, the architecture can be easily transferred to other brain-computer interface (BCI) tasks.

Humans are able to localize the source of a sound. This enables them to direct attention to a particular speaker in a cocktail party. Psycho-acoustic studies show that the sensory cortices of the human brain respond to the location of sound sources differently, and the auditory attention itself is a dynamic and temporally based brain activity. In this work, we seek to build a computational model which uses both spatial and temporal information manifested in EEG signals for auditory spatial attention detection (ASAD). We propose an end-to-end spatiotemporal attention network, denoted as STAnet, to detect auditory spatial attention from EEG. The STAnet is designed to assign differentiated weights dynamically to EEG channels through a spatial attention mechanism, and to temporal patterns in EEG signals through a temporal attention mechanism. We report the ASAD experiments on two publicly available datasets. The STAnet outperforms other competitive models by a large margin under various experimental conditions. Its attention decision for 1-second decision window outperforms that of the state-of-the-art techniques for 10-second decision window. Experimental results also demonstrate that the STAnet achieves competitive performance on EEG signals ranging from 64 to as few as 16 channels. This study provides evidence suggesting that efficient low-density EEG online decoding is within reach. This study also marks an important step towards the practical implementation of ASAD in real life applications.

Humans show a remarkable perceptual ability to select the speech stream of interest among multiple competing speakers. Previous studies demonstrated that auditory attention detection (AAD) can infer which speaker is attended by analyzing a listener's electroencephalography (EEG) activities. However, previous AAD approaches perform poorly on short signal segments, more advanced decoding strategies are needed to realize robust real-time AAD. In this study, we propose a novel approach, i.e., cross-modal attention-based AAD (CMAA), to exploit the discriminative features and the correlation between audio and EEG signals. With this mechanism, we hope to dynamically adapt the interactions and fuse cross-modal information by directly attending to audio and EEG features, thereby detecting the auditory attention activities manifested in brain signals. We also validate the CMAA model through data visualization and comprehensive experiments on a publicly available database. Experiments show that the CMAA achieves accuracy values of 82.8, 86.4, and 87.6% for 1-, 2-, and 5-s decision windows under anechoic conditions, respectively; for a 2-s decision window, it achieves an average of 84.1% under real-world reverberant conditions. The proposed CMAA network not only achieves better performance than the conventional linear model, but also outperforms the state-of-the-art non-linear approaches. These results and data visualization suggest that the CMAA model can dynamically adapt the interactions and fuse cross-modal information by directly attending to audio and EEG features in order to improve the AAD performance.

PURPOSE: Yoga is associated with psychological and physiological benefits, including relaxation and increased attention. Electroencephalograms (EEG) signals reflect the electrical activity of the brain and transformed Alpha and Beta brainwaves into meditation and attention values. This study primarily aimed to measure college students’ attention and meditation level in yoga courses. The secondary aim was to explore yoga’s effect on students’ cognitive ability. METHODS: The study selected two yoga classes in the first semester of 2023 in an academic university in China. Using convenience sampling methods, 16 college students who attended yoga courses voluntarily participated in this study. All learning phases, teaching content, and course time were maintained the same between the two classes during the study. Selected yoga poses included one previously learned pose and seven new poses. Each lesson lasted 90 min with 30-40 min review phase and 40-45 min in the teaching phase. EEG signals were processed during the review phase and the teaching phase in yoga courses. Collected data were analyzed by one-way ANOVA tests using R 4.3.1. RESULTS: For the attention level, EEG signal in the review phase was (mean ± standard deviation) 48.12±1.40, while that in the teaching phase was 46.24±2.57. For meditation, EEG signal in the review phase was 50.05±3.41, while that in the teaching phase was 49.37±4.85. The attention level in the teaching phase was significantly lower than that in the review phase (F(1,19)=4.68, p < 0.05, η2=0.2). In contrast, for the meditation level, there was no significant difference between the teaching phase and the review phase (F(1,19)=0.15, p=0.7, η2=0.008). Among different yoga poses, there was no significant difference observed in either attention or meditation. CONCLUSION: Students’ attention was significantly higher in the review phase than that in the teaching phase. As the time of two phases was relatively the same, students may lose their attention when learning new poses. To enhance class efficiency, the teaching phase is recommended to be placed before the review phase, as students may have more energy at the beginning of the class. In addition, including a review practice during the teaching phase can enhance students’ attention and achieve better effect.

Attention is one of many human cognitive functions that are essential in everyday life. Given our limited processing capacity, attention helps us focus only on what matters. Focusing attention on one speaker in an environment with many speakers is a critical ability of the human auditory system. This paper proposes a new end-to-end method based on the combined transformer and graph convolutional neural network (TraGCNN) that can effectively detect auditory attention from electroencephalograms (EEGs). This approach eliminates the need for manual feature extraction, which is often time-consuming and subjective. Here, the first EEG signals are converted to graphs. We then extract attention information from these graphs using spatial and temporal approaches. Finally, our models are trained with these data. Our model can detect auditory attention in both the spatial and temporal domains. Here, the EEG input is first processed by transformer layers to obtain a sequential representation of EEG based on attention onsets. Then, a family of graph convolutional layers is used to find the most active electrodes using the spatial position of electrodes. Finally, the corresponding EEG features of active electrodes are fed into the graph attention layers to detect auditory attention. The Fuglsang 2020 dataset is used in the experiments to train and test the proposed and baseline systems. The new TraGCNN approach, as compared with state-of-the-art attention classification methods from the literature, yields the highest performance in terms of accuracy (80.12%) as a classification metric. Additionally, the proposed model results in higher performance than our previously graph-based model for different lengths of EEG segments. The new TraGCNN approach is advantageous because attenuation detection is achieved from EEG signals of subjects without requiring speech stimuli, as is the case with conventional auditory attention detection methods. Furthermore, examining the proposed model for different lengths of EEG segments shows that the model is faster than our previous graph-based detection method in terms of computational complexity. The findings of this study have important implications for the understanding and assessment of auditory attention, which is crucial for many applications, such as brain–computer interface (BCI) systems, speech separation, and neuro-steered hearing aid development.

In this study, two electroencephalogram (EEG) feedback experiments were completed to measure the different levels of visual attention. In order to assess different visual attention levels, EEG data were processed with nonlinear dynamics parameters based on sequence complexity which involves approximate entropy, sample entropy and multi-scale entropy. According to the statistical analysis of 14 subjects’ EEG signal by using entropy as parameters, significant differences in attention intensity have been found in most of the electrodes in frontal regions and some of the electrodes in temporal regions. The values of entropy indicate a declining tendency with the decreasing level of attention and among all the parameters, sample entropy achieves the highest sensitivity in the classification performance of visual attention. We also applied a classifier based on support vector machine (SVM) to discriminate the different levels of attention which finally achieved a reasonable recognition ratio of 85.24%.

Purpose: Extended periods of desk-based study in schools often result in diminished attention and reduced learning efficiency among students. This study aims to investigate exercise methods and implementation approaches that can enhance the attention and academic performance of primary school students. Methods: This study randomly selected 90 students（44 in the experimental group and 46 in the control group）from two classes of Grade 3 in Shenxhen Yantian Foreign Language primary School Donghe Branch. The students from the experimental group received two rounds of 12-week 10 minutes of micro-exercise every day. The attention index (D2) and the academic performance (Chinese, Math, and English) of the experimental group and the control group before and after the experiment were compared and analyzed. Results: Before the experiment, there was no significant difference between the experimental group and the control group. After the experiment, there was a significant difference in the attention index between the experimental group and the control group (tTN=7.274, PTN < 0.01). There was a positive correlation between students' math performance and attention index of the experimental group before and after the experiment (rE1=-0.413, PE1= 0.005 < 0.01; rE2=-0.423, PE2=0.003 < 0.01). Conclusion: The current research indicates that incorporating ten minutes of micro-exercise between classes can activate students' attention systems, enhance attention processing speed, and elevate overall attention levels. Consistent implementation of micro-exercises during breaks has been found to enhance students' computational skills, logical thinking, and other cognitive abilities, leading to improvements in their academic performance, particularly in math. The sustained use of micro-exercises between classes emerges as an effective strategy for cultivating students' attention levels and boosting academic performance.

In online learning environments, the need to monitor student engagement to achieve desired learning outcomes is foundational to any pedagogy for effective online teaching. With the absence of in-person supervision, keeping track of student attention in remote classes is quite arduous, if not impossible. This paper proposes an IoT-based EEG device for the real-time assessment of students' brain activity to measure and possibly improve students' attention levels in virtual classes. An ensemble machine learning (ML) study, specifically an artificial neural network (ANN) approach, is employed to investigate the relationship between student performance and EEG data. The study shows the performance of students is negatively correlated to the delta power and the theta/alpha ratio, the common EEG metrics for mental fatigue and drowsiness, respectively, for the student. An IoT-enabled EEG device provides teachers with real-time, precise, and unbiased data regarding the student's cognitive attention levels, via a reporting system. Thus, this research demonstrates that the proposed system, utilizing EEG and ANN ensemble-ML methods, can predict attention levels in real-time, enabling timely intervention for students who are disconnected. The study opens the door to the use of advanced BCI systems in teaching to maximize student attention.

Enhancements in the rehabilitation of motor and cognitive functions are significantly attainable through proactive patient engagement. The difficulty of rehabilitation tasks and the environment in which they are conducted directly impact patient motivation. Consequently, this study introduces a dynamic difficulty adjustment method for rehabilitation training tasks based on attention levels, designed to adjust task difficulty in real-time and augment the focus of participants on their training tasks. 
Approach: EEG signals from participants were harnessed to train an attention classification model, enabling the acquisition of real-time attention level signals. Task difficulty levels were adjusted based on the fluctuating attention levels. A cohort of 30 participants was engaged to evaluate: (1) the impact on engagement when attention levels are utilized as dynamic difficulty 18 triggers; (2) the influence of various task environments on concentration. The experiment was assessed through EEG signals and questionnaire data, with frequency domain analysis conducted on EEG signals to calculate concentration values and statistical analysis performed on additional data. 
Main Results: The findings reveal that within an identical virtual reality (VR) environment, leveraging attention levels as triggers for difficulty adjustment markedly improves participants' task concentration. Compared to 2D environments, VR environments substantially enhance participants' sense of immersion, interest, and flow state, albeit with increased physical exertion during training. The integration of VR and attention level feedback is deemed the most effective strategy. 
Significance: These exploratory insights indicate that the proposed method paves a novel path for boosting patient engagement in rehabilitation. Immersive rehabilitation training, driven by attention levels, promises a more effective and captivating patient experience. This study advances the field by offering data-driven, personalized rehabilitation approaches, potentially culminating in superior patient outcomes and enhanced quality of life.