Locus coeruleus microstructural integrity is associated with vigilance vulnerability to sleep deprivation.

Insufficient sleep compromises cognitive performance, diminishes vigilance, and disrupts daily functioning in hundreds of millions of people worldwide. Despite extensive research revealing significant variability in vigilance vulnerability to sleep deprivation, the underlying mechanisms of these individual differences remain elusive. Locus coeruleus (LC) plays a crucial role in the regulation of sleep-wake cycles and has emerged as a potential marker for vigilance vulnerability to sleep deprivation. In this study, we investigate whether LC microstructural integrity, assessed by fractional anisotropy (FA) through diffusion tensor imaging (DTI) at baseline before sleep deprivation, can predict impaired psychomotor vigilance test (PVT) performance during sleep deprivation in a cohort of 60 healthy individuals subjected to a rigorously controlled in-laboratory sleep study. The findings indicate that individuals with high LC FA experience less vigilance impairment from sleep deprivation compared with those with low LC FA. LC FA accounts for 10.8% of the variance in sleep-deprived PVT lapses. Importantly, the relationship between LC FA and impaired PVT performance during sleep deprivation is anatomically specific, suggesting that LC microstructural integrity may serve as a biomarker for vigilance vulnerability to sleep loss.

Phosphorylated Tau at T181 accumulates in the serum of hibernating Syrian hamsters and rapidly disappears after arousal.

The search for biomarkers for the early diagnosis of neurodegenerative diseases is a growing area. Numerous investigations are exploring minimally invasive and cost-effective biomarkers, with the detection of phosphorylated Tau (pTau) protein emerging as one of the most promising fields. pTau is the main component of the paired helical filaments found in the brains of Alzheimer's disease cases and serves as a precursor in the formation of neurofibrillary tangles (NFTs). Recent research has revealed that analysis of p-Tau181, p-Tau217 and p-Tau231 in blood may be an option for detecting the preclinical stage of Alzheimer's disease. In this study, we have analyzed the values of pTau 181 in the serum of Syrian hamsters during hibernation. Naturally, over the course of hibernation, these animals exhibit a reversible accumulation of pTau in the brain tissue, which rapidly disappears upon awakening. A biosensing system based on the interferometric optical detection method was used to measure the concentration of pTau181 protein in serum samples from Syrian hamsters. This method eliminates the matrix effect and amplifies the signal obtained by using silicon dioxide nanoparticles (SiO2 NPs) biofunctionalized with the αpTau181 antibody. Our results indicate a substantial increase in the serum concentration of pTau in threonine-181 during hibernation, which disappears completely 2-3 h after awakening. Investigating the mechanism by which pTau protein appears in the blood non-pathologically may enhance current diagnostic techniques. Furthermore, since this process is reversible, and no tangles are detected in the brains of hibernating hamsters, additional analysis may contribute to the discovery of improved biomarkers. Additionally, exploring drugs targeting pTau to prevent the formation of tangles or studying the outcomes of any pTau-targeted treatment could be valuable.

Feeling tired versus feeling relaxed: Two faces of low physiological arousal.

Human well-being and functioning depend on two fundamental mental states: Relaxation and sleepiness. Relaxation and sleepiness are both assumed to be states of low physiological arousal and negatively correlated. However, it is still unclear how consistent this negative relationship is across different settings and whether it changes before and after an intervention. Here we investigated this intricate relationship between subjective momentary sleepiness and relaxation states by meta-analytically analyzing several data sets from studies using the Relaxation State Questionnaire. We discovered that subjective sleepiness and relaxation were in fact anti-correlated pre-intervention. This anti-correlation provides a quantitative dissociation between sleepiness and relaxation. Thus, even though sleepiness and relaxation both implicate a low arousal level, the two mental states are subjectively experienced in a qualitatively different fashion, and thus reflect distinct underlying constructs. For the post-intervention relationship, this negative correlation could not be consistently found. This indicates that there are aspects of the experimental setting or intervention that introduce changes in the dynamics of the relationship of the two constructs.

Physiological and communicative emotional disconcordance in children on the autism spectrum.

BACKGROUND: Individuals on the autism spectrum commonly have differences from non-autistic people in expressing their emotions using communicative behaviors, such as facial expressions. However, it is not yet clear if this reduced expressivity stems from reduced physiological reactivity in emotional contexts or if individuals react internally, but do not show these reactions externally to others. We hypothesized that autism is characterized by a discordance between in-the-moment internal psychophysiological arousal and external communicative expressions of emotion. METHODS: Forty-one children on the autism spectrum and 39 non-autistic, typically developing (TD) children of two age groups (2-4 and 8-12 years) participated in a low-level stress task whilst wearing a wireless electrocardiogram. Children's negative emotional expressions (facial, vocal, bodily) were coded following standardized protocols. Alexithymia traits were assessed using the Children's Alexithymia Measure with school-aged children only. Data analyses involved ANOVAs, correlations, and sensitivity analyses. RESULTS: There were no group differences in physiological arousal (heart rate) or in communicative expressions of stress to the stress task. For TD preschoolers, physiological arousal during the stress task was associated with vocal expressions and for TD school-aged children, they were associated with facial and bodily expressions. By contrast, for children on the autism spectrum, physiological arousal during the stress tasks was not associated with communicative expressions across age groups. CONCLUSIONS: Our findings suggest that children on the autism spectrum might experience emotional disconcordance, in that their physiological arousal does not align with their communicative expressions. Therefore, the internally experienced stress of children on the autism spectrum may be inadvertently missed by teachers and caregivers and, consequently, learning opportunities for teaching emotional communication and regulation may be also missed. Our results support the use of wearable biosensors to facilitate such interventions in children on the autism spectrum.

Exploration of resting-state brain functional connectivity as preclinical markers for arousal prediction in prolonged disorders of consciousness: A pilot study based on functional near-infrared spectroscopy.

BACKGROUND: There is no diagnostic assessment procedure with moderate or strong evidence of use, and evidence for current means of treating prolonged disorders of consciousness (pDOC) is sparse. This may be related to the fact that the mechanisms of pDOC have not been studied deeply enough and are not clear enough. Therefore, the aim of this study was to explore the mechanism of pDOC using functional near-infrared spectroscopy (fNIRS) to provide a basis for the treatment of pDOC, as well as to explore preclinical markers for determining the arousal of pDOC patients. METHODS: Five minutes resting-state data were collected from 10 pDOC patients and 13healthy adults using fNIRS. Based on the concentrations of oxyhemoglobin (HbO) and deoxyhemoglobin (HbR) in the time series, the resting-state cortical brain functional connectivity strengths of the two groups were calculated, and the functional connectivity strengths of homologous and heterologous brain networks were compared at the sensorimotor network (SEN), dorsal attention network (DAN), ventral attention network (VAN), default mode network (DMN), frontoparietal network (FPN), and visual network (VIS) levels. Univariate binary logistic regression analyses were performed on brain networks with statistically significant differences to identify brain networks associated with arousal in pDOC patients. The receiver operating characteristic (ROC) curves were further analyzed to determine the cut-off value of the relevant brain networks to provide clinical biomarkers for the prediction of arousal in pDOC patients. RESULTS: The results showed that the functional connectivity strengths of oxyhemoglobin (HbO)-based SEN∼SEN, VIS∼VIS, DAN∼DAN, DMN∼DMN, SEN∼VIS, SEN∼FPN, SEN∼DAN, SEN∼DMN, VIS∼FPN, VIS∼DAN, VIS∼DMN, HbR-based SEN∼SEN, and SEN∼DAN were significantly reduced in the pDOC group and were factors that could reflect the participants' state of consciousness. The cut-off value of resting-state functional connectivity strength calculated by ROC curve analysis can be used as a potential preclinical marker for predicting the arousal state of subjects. CONCLUSION: Resting-state functional connectivity strength of cortical networks is significantly reduced in pDOC patients. The cut-off values of resting-state functional connectivity strength are potential preclinical markers for predicting arousal in pDOC patients.

Pupil contagion variation with gaze, arousal, and autistic traits.

Pupillary contagion occurs when one's pupil size unconsciously adapts to the pupil size of an observed individual and is presumed to reflect the transfer of arousal. Importantly, when estimating pupil contagion, low level stimuli properties need to be controlled for, to ensure that observations of pupillary changes are due to internal change in arousal rather than the external differences between stimuli. Here, naturalistic images of children's faces depicting either small or large pupils were presented to a group of children and adolescents with a wide range of autistic traits, a third of whom had been diagnosed with autism. We examined the extent to which pupillary contagion reflects autonomic nervous system reaction through pupil size change, heart rate and skin conductance response. Our second aim was to determine the association between arousal reaction to stimuli and degree of autistic traits. Results show that pupil contagion and concomitant heart rate change, but not skin conductance change, was evident when gaze was restricted to the eye region of face stimuli. A positive association was also observed between pupillary contagion and autistic traits when participants' gaze was constrained to the eye region. Findings add to a broader understanding of the mechanisms underlying pupillary contagion and its association with autism.

Orexin-mediated motivated arousal and reward seeking.

The neuromodulator orexin has been identified as a key factor for motivated arousal including recent evidence that sleep deprivation-induced enhancement of reward behavior is modulated by orexin. While orexin is not necessary for either reward or arousal behavior, orexin neurons' broad projections, ability to sense the internal state of the animal, and high plasticity of signaling in response to natural rewards and drugs of abuse may underlie heightened drug seeking, particularly in a subset of highly motivated reward seekers. As such, orexin receptor antagonists have gained deserved attention for putative use in addiction treatments. Ongoing and future clinical trials are expected to identify individuals most likely to benefit from orexin receptor antagonist treatment to promote abstinence, such as those with concurrent sleep disorders or high craving, while attention to methodological considerations will aid interpretation of the numerous preclinical studies investigating disparate aspects of the role of orexin in reward and arousal.

The aperiodic exponent of neural activity varies with vigilance state in mice and men.

Recently the 1/f signal of human electroencephalography has attracted attention, as it could potentially reveal a quantitative measure of neural excitation and inhibition in the brain, that may be relevant in a clinical setting. The purpose of this short article is to show that the 1/f signal depends on the vigilance state of the brain in both humans and mice. Therefore, proper labelling of the EEG signal is important as improper labelling may obscure disease-related changes in the 1/f signal. We demonstrate this by comparing EEG results from a longitudinal study in a genetic mouse model for synaptic dysfunction in schizophrenia and autism spectrum disorders to results from a large European cohort study with schizophrenia and mild Alzheimer's disease patients. The comparison shows when the 1/f is corrected for vigilance state there is a difference between groups, and this effect disappears when vigilance state is not corrected for. In conclusion, more attention should be paid to the vigilance state during analysis of EEG signals regardless of the species.

Low arousal threshold is associated with altered functional connectivity of the ascending reticular activating system in patients with obstructive sleep apnea.

A low arousal threshold (LAT) is a pathophysiological trait of obstructive sleep apnea (OSA) that may be associated with brainstem ascending reticular activating system-cortical functional connectivity changes. We evaluated resting-state connectivity between the brainstem nuclei and 105 cortical/subcortical regions in OSA patients with or without a LAT and healthy controls. Twenty-five patients with moderate to severe OSA with an apnea-hypopnea index between 20 and 40/hr (15 with and 10 without a LAT) and 15 age- and sex-matched controls were evaluated. Participants underwent functional magnetic resonance imaging after overnight polysomnography. Three brainstem nuclei-the locus coeruleus (LC), laterodorsal tegmental nucleus (LDTg), and ventral tegmental area (VTA)-associated with OSA in our previous study were used as seeds. Functional connectivity values of the two brainstem nuclei (LC and LDTg) significantly differed among the three groups. The connectivity of the LC with the precuneus was stronger in OSA patients than in controls regardless of the concomitant LAT. The connectivity between the LDTg and the posterior cingulate cortex was also stronger in OSA patients regardless of the LAT. Moreover, OSA patients without a LAT showed stronger LDTg-posterior cingulate cortex connectivity than those with a LAT (post hoc p = 0.013), and this connectivity strength was negatively correlated with the minimum oxygen saturation in OSA patients (r = - 0.463, p = 0.023). The LAT in OSA patients was associated with altered LDTg-posterior cingulate cortex connectivity. This result may suggested that cholinergic activity may play a role in the LAT in OSA patients.

Examining the effects of caffeine during an auditory attention task.

Participants completed two sessions of an auditory attention task and intermittently responded to thought probes asking about their level of mind-wandering. After the first session one group received 200 mg of caffeinated chewing gum (n = 61) and another group received regular (placebo) chewing gum (n = 66). The gum was chewed for 20-minutes and then disposed of before beginning the second session. Participants who received caffeine showed a performance benefit as well as reported being more on task and fewer instances of spontaneous mind-wandering compared to those in the placebo group. Participants who received caffeine also reported greater positive affect and arousal, as well as less feelings of boredom, sleepiness, and mental effort required to stay on task compared to those who received placebo. These results suggest that caffeine may benefit attentional engagement as well as performance during a sustained attention task.

Norms for 718 Persian Words in Emotional Dimensions, Animacy, and Familiarity.

Research frequently uses words as stimuli to assess cognitive and psychological processes. However, various attributes of these words, such as their semantic and emotional aspects, could potentially confound study results if not properly controlled. This study aims to establish a reliable foundation for the semantic and emotional aspects of words for research in Persian. To this end, the present study provided norms for 718 Persian nouns in arousal, valence, familiarity, and animacy dimensions. The words were selected from a previous English dataset (Warriner et al. in Behav Res Methods 45(4):1191-1207, 2013), translated into Persian, and rated by a total of 463 native Persian-speaking participants. The ratings were obtained through an online questionnaire using a 9-point Likert scale for emotional dimensions (i.e., valence and arousal) and a 5-point Likert scale for semantic dimensions (i.e., familiarity and animacy). The reliability of the ratings was measured using the split-half method, and the result indicated a high consistency of ratings in all dimensions. To assess the relationship between the emotional and semantic dimensions, Pearson correlation coefficient was conducted. Gender differences were investigated through the Mann-Whitney U test, and significant differences were observed in all dimensions. These results are compared with findings from previous studies that were conducted in various languages.

Sensing emotional valence and arousal dynamics through automated facial action unit analysis.

Information about the concordance between dynamic emotional experiences and objective signals is practically useful. Previous studies have shown that valence dynamics can be estimated by recording electrical activity from the muscles in the brows and cheeks. However, whether facial actions based on video data and analyzed without electrodes can be used for sensing emotion dynamics remains unknown. We investigated this issue by recording video of participants' faces and obtaining dynamic valence and arousal ratings while they observed emotional films. Action units (AUs) 04 (i.e., brow lowering) and 12 (i.e., lip-corner pulling), detected through an automated analysis of the video data, were negatively and positively correlated with dynamic ratings of subjective valence, respectively. Several other AUs were also correlated with dynamic valence or arousal ratings. Random forest regression modeling, interpreted using the SHapley Additive exPlanation tool, revealed non-linear associations between the AUs and dynamic ratings of valence or arousal. These results suggest that an automated analysis of facial expression video data can be used to estimate dynamic emotional states, which could be applied in various fields including mental health diagnosis, security monitoring, and education.

Contrastive fine-grained domain adaptation network for EEG-based vigilance estimation.

Vigilance state is crucial for the effective performance of users in brain-computer interface (BCI) systems. Most vigilance estimation methods rely on a large amount of labeled data to train a satisfactory model for the specific subject, which limits the practical application of the methods. This study aimed to build a reliable vigilance estimation method using a small amount of unlabeled calibration data. We conducted a vigilance experiment in the designed BCI-based cursor-control task. Electroencephalogram (EEG) signals of eighteen participants were recorded in two sessions on two different days. And, we proposed a contrastive fine-grained domain adaptation network (CFGDAN) for vigilance estimation. Here, an adaptive graph convolution network (GCN) was built to project the EEG data of different domains into a common space. The fine-grained feature alignment mechanism was designed to weight and align the feature distributions across domains at the EEG channel level, and the contrastive information preservation module was developed to preserve the useful target-specific information during the feature alignment. The experimental results show that the proposed CFGDAN outperforms the compared methods in our BCI vigilance dataset and SEED-VIG dataset. Moreover, the visualization results demonstrate the efficacy of the designed feature alignment mechanisms. These results indicate the effectiveness of our method for vigilance estimation. Our study is helpful for reducing calibration efforts and promoting the practical application potential of vigilance estimation methods.

Neuronal rhythmicity and cortical arousal in a mouse model of absence epilepsy.

OBJECTIVES: Absence seizures impair psychosocial function, yet their detailed neuronal basis remains unknown. Recent work in a rat model suggests that cortical arousal state changes prior to seizures and that single neurons show diverse firing patterns during seizures. Our aim was to extend these investigations to a mouse model with studies of neuronal activity and arousal state to facilitate future fundamental investigations of absence epilepsy. METHODS: We performed in vivo extracellular single unit recordings on awake head-fixed C3H/HeJ mice. Mice were implanted with tripolar electrodes for cortical electroencephalography (EEG). Extracellular single unit recordings were obtained with glass micropipettes in the somatosensory barrel cortex, while animals ambulated freely on a running wheel. Signals were digitized and analyzed during seizures and at baseline. RESULTS: Neuronal activity was recorded from 36 cortical neurons in 19 mice while EEG showed characteristic 7-8 Hz spike-wave discharges. Different single neurons showed distinct firing patterns during seizures, but the overall mean population neuronal firing rate during seizures was no different from pre-seizure baseline. However, the rhythmicity of neuronal firing during seizures was significantly increased (p < 0.001). In addition, beginning 10s prior to seizure initiation, we observed a progressive decrease in cortical high frequency (>40 Hz) EEG and an increase in lower frequency (1-39 Hz) activity suggesting decreased arousal state. SIGNIFICANCE: We found that the awake head-fixed C3H/HeJ mouse model demonstrated rhythmic neuronal firing during seizures, and a decreased cortical arousal state prior to seizure onset. Unlike the rat model we did not observe an overall decrease in neuronal firing during seizures. Similarities and differences across species strengthen the ability to investigate fundamental key mechanisms. Future work in the mouse model will identify the molecular basis of neurons with different firing patterns, their role in seizure initiation and behavioral deficits, with ultimate translation to human absence epilepsy.

Effects of early midlife ovarian removal on sleep: Polysomnography-measured cortical arousal, homeostatic drive, and spindle characteristics.

Bilateral salpingo-oophorectomy (BSO; removal of ovaries and fallopian tubes) prior to age 48 is associated with elevated risk for both Alzheimer's disease (AD) and sleep disorders such as insomnia and sleep apnea. In early midlife, individuals with BSO show reduced hippocampal volume, function, and hippocampal-dependent verbal episodic memory performance associated with changes in sleep. It is unknown whether BSO affects fine-grained sleep measurements (sleep microarchitecture) and how these changes might relate to hippocampal-dependent memory. We recruited thirty-six early midlife participants with BSO. Seventeen of these participants were taking 17ß-estradiol therapy (BSO+ET) and 19 had never taken ET (BSO). Twenty age-matched control participants with intact ovaries (AMC) were also included. Overnight at-home polysomnography recordings were collected, along with subjective sleep quality and hot flash frequency. Multivariate Partial Least Squares (PLS) analysis was used to assess how sleep varied between groups. Compared to AMC, BSO without ET was associated with significantly decreased time spent in non-rapid eye movement (NREM) stage 2 sleep as well as increased NREM stage 2 and 3 beta power, NREM stage 2 delta power, and spindle power and maximum amplitude. Increased spindle maximum amplitude was negatively correlated with verbal episodic memory performance. Decreased sleep latency, increased sleep efficiency, and increased time spent in rapid eye movement sleep were observed for BSO+ET. Findings suggest there is an association between ovarian hormone loss and sleep microarchitecture, which may contribute to poorer cognitive outcomes and be ameliorated by ET.

The influence of transcranial direct current stimulation to the trigeminal nerve on attention and arousal.

One mechanism by which transcranial direct current stimulation (tDCS) has been proposed to improve attention is by transcutaneous stimulation of cranial nerves, thereby activating the locus coeruleus (LC). Specifically, placement of the electrodes over the frontal bone and mastoid is thought to facilitate current flow across the face as a path of least resistance. The face is innervated by the trigeminal nerve, and the trigeminal nerve is interconnected with the LC. In this study, we tested whether stimulating the trigeminal nerve impacts indices of LC activity and performance on a sustained attention task. We replicated previous research that shows deterioration in task performance, increases in the rate of task-unrelated thoughts, and reduced pupil responses due to time on task irrespective of tDCS condition (sham, anodal, and cathodal stimulation). Importantly, tDCS did not influence pupil dynamics (pretrial or stimulus-evoked), self-reported attention state, nor task performance in active versus sham stimulation conditions. The findings reported here are consistent with theories about arousal centered on a hypothesized link between LC activity indexed by pupil size, task performance, and self-reported attention state but fail to support hypotheses that tDCS over the trigeminal nerve influences indices of LC function.

Does Physiological Arousal Increase Social Transmission of Information? Two Replications of Berger (2011).

People share information for many reasons. For example, Berger (2011, N = 40) found that undergraduate participants manipulated to have higher physiological arousal were more likely to share a news article with others via email than people who had low arousal. Berger's research is widely cited as evidence of the causal role of arousal in sharing information and has been used to explain why information that induces high-arousal emotions is shared more than information that induces low-arousal emotions. We conducted two replications (N = 111, N = 160) of Berger's study, using the same arousal manipulation but updating the sharing measure to reflect the rise of information sharing through social media. Both studies failed to find an impact of incidental physiological arousal on undergraduate participants' willingness to share news articles on social media. Our studies cast doubt on the idea that incidental physiological arousal-in the absence of other factors-impacts people's decisions to share information on social networking sites.

Application of Facial Expression Analysis Technology in Violence Risk Assessment of Individuals with Mental Disorders in Supervised Settings.

OBJECTIVES: To explore the association between violent behaviors and emotions in individuals with mental disorders, to evaluate the application value of facial expression analysis technology in violence risk assessment of individuals with mental disorders in supervised settings, and to provide a reference for violence risk assessment. METHODS: Thirty-nine male individuals with mental disorders in supervised settings were selected, the participant risk of violence, cognitive function, psychiatric symptoms and severity were assessed using the Modified Overt Aggression Scale (MOAS), the Historical, Clinical, Risk Management-Chinese version(HCR-CV), the Positive and Negative Syndrome Scale (PANSS) and the Brief Psychiatric Rating Scale (BPRS). An emotional arousal was performed on the participants and the intensity of their emotions and facial expression action units was recorded before, during and after the arousal. One-way analysis of variance (ANOVA) was used to compare the differences in the intensity of emotions and facial expression action units before, during and after the arousal. Pearson correlation analysis was used to calculate the correlations between the intensity of the seven basic emotional facial expressions and the scores of the assessment scales. RESULTS: The intensity difference of sadness, surprise and fear in different time periods was statistically significant (P<0.05). The intensity of the left medial eyebrow lift action unit was found significantly different before and after the emotional arousal (P<0.05). The intensity of anger was positively correlated with the Modified Overt Aggression Scale score throughout the experiment (P<0.05). CONCLUSIONS: Eye action units such as eyebrow lifting, eyelid tightening and upper eyelid lifting can be used as effective action units to identify sadness, anger and other negative emotions associated with violent behaviors. Facial expression analysis technology can be used as an auxiliary tool to assess the potential risk of violence in individuals with mental disorders in supervised settings.

Inter- and intra-subject similarity in network functional connectivity across a full narrative movie.

Naturalistic paradigms, such as watching movies during functional magnetic resonance imaging, are thought to prompt the emotional and cognitive processes typically elicited in real life situations. Therefore, naturalistic viewing (NV) holds great potential for studying individual differences. Previous studies have primarily focused on using shorter movie clips, geared toward eliciting specific and often isolated emotions, while the potential behind using full narratives depicted in commercial movies as a proxy for real-life experiences has barely been explored. Here, we offer preliminary evidence that a full narrative movie (FNM), that is, a movie covering a complete narrative arc, can capture complex socio-affective dynamics and their links to individual differences. Using the studyforrest dataset, we investigated inter- and intra-subject similarity in network functional connectivity (NFC) of 14 meta-analytically defined networks across a full narrative, audio-visual movie split into eight consecutive movie segments. We characterized the movie segments by valence and arousal portrayed within the sequences, before utilizing a linear mixed model to analyze which factors explain inter- and intra-subject similarity. Our results show that the model best explaining inter-subject similarity comprised network, movie segment, valence and a movie segment by valence interaction. Intra-subject similarity was influenced significantly by the same factors and an additional three-way interaction between movie segment, valence and arousal. Overall, inter- and intra-subject similarity in NFC were sensitive to the ongoing narrative and emotions in the movie. We conclude that FNMs offer complex content and dynamics that might be particularly valuable for studying individual differences. Further characterization of movie features, such as the overarching narratives, that enhance individual differences is needed for advancing the potential of NV research.

CATM: A Multi-Feature-Based Cross-Scale Attentional Convolutional EEG Emotion Recognition Model.

Aiming at the problem that existing emotion recognition methods fail to make full use of the information in the time, frequency, and spatial domains in the EEG signals, which leads to the low accuracy of EEG emotion classification, this paper proposes a multi-feature, multi-frequency band-based cross-scale attention convolutional model (CATM). The model is mainly composed of a cross-scale attention module, a frequency-space attention module, a feature transition module, a temporal feature extraction module, and a depth classification module. First, the cross-scale attentional convolution module extracts spatial features at different scales for the preprocessed EEG signals; then, the frequency-space attention module assigns higher weights to important channels and spatial locations; next, the temporal feature extraction module extracts temporal features of the EEG signals; and, finally, the depth classification module categorizes the EEG signals into emotions. We evaluated the proposed method on the DEAP dataset with accuracies of 99.70% and 99.74% in the valence and arousal binary classification experiments, respectively; the accuracy in the valence-arousal four-classification experiment was 97.27%. In addition, considering the application of fewer channels, we also conducted 5-channel experiments, and the binary classification accuracies of valence and arousal were 97.96% and 98.11%, respectively. The valence-arousal four-classification accuracy was 92.86%. The experimental results show that the method proposed in this paper exhibits better results compared to other recent methods, and also achieves better results in few-channel experiments.