Search

Brain–computer interfaces (BCIs) target specific brain activity for neuropsychological rehabilitation, and also allow patients with motor disabilities to control mobility and communication devices. Motor imagery of single-handed actions is used in BCIs but many users cannot control the BCIs effectively, limiting applications in the health systems. Crosstalk is unintended brain activations that interfere with bimanual actions and could also occur during motor imagery. To test if crosstalk impaired BCI user performance, we recorded EEG in 46 participants while they imagined movements in four experimental conditions using motor imagery: left hand (L), right hand (R), tongue (T) and feet (F). Pairwise classification accuracies of the tasks were compared (LR, LF, LT, RF, RT, FT), using common spatio-spectral filters and linear discriminant analysis. We hypothesized that LR classification accuracy would be lower than every other combination that included a hand imagery due to crosstalk. As predicted, classification accuracy for LR (58%) was reliably the lowest. Interestingly, participants who showed poor LR classification also demonstrated at least one good TR, TL, FR or FL classification; and good LR classification was detected in 16% of the participants. For the first time, we showed that crosstalk occurred in motor imagery, and affected BCI performance negatively. Such effects are effector-sensitive regardless of the BCI methods used; and likely not apparent to the user or the BCI developer. This means that tasks choice is crucial when designing BCI. Critically, the effects of crosstalk appear mitigatable. We conclude that understanding crosstalk mitigation is important for improving BCI applicability. SUPPLEMENTARY INFORMATION: The online version of this article contains supplementary material available (10.1007/s13755-021-00142-y).

Like socio-economic status and cognitive abilities, personality traits predict important life outcomes. Traits that reflect unusually low or high approach motivations, such as low extraversion and high disinhibition, are linked to various forms of mental disorder. Similarly, the dopamine system is theoretically linked to approach motivation traits and to various forms of mental disorder. Identifying neural contributions to extremes of such traits should map to neural sources of psychopathology, with dopamine a prime candidate. Notably, dopamine cells fire in response to unexpected reward, which suggests that the size of non-invasive, scalp-recorded potentials evoked by unexpected reward could reflect sensitivity in approach motivation traits. Here, we evaluated the validity of evoked electroencephalography (EEG) responses to unexpected reward in a monetary gain/loss task to assess approach motivation traits in 137 participants, oversampled for externalizing psychopathology symptoms. We demonstrated that over the 0–400 ms period in which feedback on the outcome was presented, responses evoked by unexpected reward contributed to all theoretically relevant approach motivation trait domains (disinhibition, extraversion and the behavioural activation system); and did so only at times when dopamine responses normally peak and reportedly code salience (70–100 ms) and valuation (200–300 ms). In particular, we linked “dopaminergic” salience and valuation to the psychopathology-related constructs of low extraversion (social anxiety) and high disinhibition (impulsivity) respectively, making the evoked potential components biomarker candidates for indexing aberrant processing of unexpected reward.

Purpose: This study aims to identify EEG biomarkers that predict the level of depressive personality (where extreme scores indicate disorder), as opposed to the presence or absence of a depressive state or a depression diagnosis.Methods: Fourier features were extracted from 2-second epochs of resting state EEG and used by LSBoost to maximise the correlation with depressive trait tendencies (PID-5 depressivity index).Results: Our method accounted for 25.75% of the variance in PID-5 scores, albeit in females only. The recording channel C3 and frequencies in the gamma band were the most important contributors to the prediction. The findings are consistent with previous psychological studies and suggest that our method is a feasible strategy for developing quantitative EEG biomarkers for trait depressivity in a neuropsychologically interpretable form. We have also shown that there might be different markers for depressivity between males and females.

EEG signals are often contaminated with artefacts, particularly with mains power from electrical equipment. Low-pass filtering and notch filtering can lose valuable data. Here we describe a novel mains power noise removal method based on the fact that mains power noise is a sine wave component with an essentially fixed frequency and the same phase across all channels. This removes the blink component, leaving uncontaminated EEG largely unchanged. Blink removal had a success rate of >99.9% recovered variance of the original EEG when removing synthesised mains power components. We compared this method with two other popular mains power noise removal methods. Our method was better than Cleanline which is an ICA based method and multiple PLI which is an adaptive notch filter based method. With the higher recovery, our method shows clear advantages over Cleanline and multiple PLI for removing mains power noise. From the aspect of data loss, it is obviously better than low-pass filter and normal notch filter.

In this paper, we propose and implement an EEG-based three-dimensional Convolutional Neural Network architecture, ‘AnxietyDecoder’, to predict anxious personality and decode its potential biomarkers from the participants. Since Goal-Conflict-Specific-Rhythmicity (GCSR) in the EEG is a sign of an anxiety-related system working, we first propose a two-dimensional Conflict-focused CNN (2-D CNN). It simulates the GCSR extraction process but with the advantages of automatic frequency band selection and functional contrast calculation optimization, thus providing more comprehensive trait anxiety predictions. Then, to generate more targeted hierarchical features from local spatio-temporal scale to global, we propose a three-dimensional Conflict-focused CNN (3-D CNN), which simultaneously integrates information in the temporal and brain-topology-related spatial dimensions. In addition, we embed Layer-wise Relevance Propagation (LRP) into our model to reveal the essential brain areas that are correlated to anxious personality. The experimental results show that the percentage variance accounted for by our three-dimensional Conflict-focused CNN is 33%, which is almost four times higher than the previous theoretically derived GCSR contrast (7%). Meanwhile, it also outperforms the 2-D model (26%) and the t-test difference between the 3-D and 2-D models is significant (t(4) = 5.4962, p = 0.0053). What’s more, the reverse engineering results provide an interpretable way to understand the prediction decision-making and participants’ anxiety personality. Our proposed AnxietyDecoder not only sets a new benchmark for EEG-based anxiety prediction but also reveals essential EEG components that contribute to the decision-making, and thus sheds some light on the anxiety biomarker research.

In this paper, an emotional EEG-specific threedimensional Convolutional Neural Network, EmotioNet, is proposed and implemented to accurately recognize emotion states. For the first time, raw data in the benchmark emotional EEG database, i.e., DEAP, are used as the input to a CNN architecture. In order to investigate the spatio-temporal character of emotional features, the effectiveness of 2-D and 3-D convolution kernels, which extract spatial and temporal features separately and simultaneously, are compared in detail. Furthermore, two major problems of EEG-based emotion recognition, namely, covariance shift and the unreliability of emotional ground truth, are described, and the effectiveness of batch normalization and dense prediction, which alleviate these problems respectively, are also investigated. Experimental results show that 3-D kernels, batch normalization, and dense prediction are all essential techniques for the emotional EEG-specific CNN architecture. The proposed EmotioNet, namely, a 3-D covariance shift adaptation-based CNN with a dense prediction layer, achieves classification rates of 73.3% and 72.1% for arousal and valence, equivalent to the best performance of several previous studies. Importantly, our results are based on automatic feature extraction, which is in contrast to previous handcrafted features. Therefore, EmotioNet provides a new method for EEG-based emotion recognition.

Background Syndromes of fear/anxiety are currently ill-defined, with no accepted human biomarkers for anxiety-specific processes. A unique common neural action of different classes of anxiolytic drugs may provide such a biomarker. In rodents, a reduction in low frequency (4–12 Hz; “theta”) brain rhythmicity is produced by all anxiolytics (even those lacking panicolytic or antidepressant action) and not by any non-anxiolytics. This rhythmicity is a key property of the Behavioural Inhibition System (BIS) postulated to be one neural substrate of anxiety. We sought homologous anxiolytic-sensitive changes in human surface EEG rhythmicity. Method Thirty-four healthy volunteers in parallel groups were administered double blind single doses of triazolam 0.25 mg, buspirone 10 mg or placebo 1 hour prior to completing the stop-signal task. Right frontal conflict-specific EEG power (previously shown to correlate with trait anxiety and neuroticism in this task) was extracted as a contrast between trials with balanced approach–avoidance (stop-go) conflict and the average of trials with net approach and net avoidance. Results Compared with placebo, both triazolam and buspirone decreased right-frontal, 9–10 Hz, conflict-specific-power. Limitations Only one dose of each of only two classes of anxiolytic and no non-anxiolytics were tested, so additional tests are needed to determine generality. Conclusions There is a distinct rhythmic system in humans that is sensitive to both classical/GABAergic and novel/serotonergic anxiolytics. This conflict-specific rhythmicity should provide a biomarker, with a strong pre-clinical neuropsychology, for a novel approach to classifying anxiety disorders.

The medial right frontal cortex is implicated in fast stopping of an initiated motor action in the stop-signal task (SST). To assess whether this region is also involved in the slower behavioural inhibition induced by goal conflict, we tested for effects of goal conflict (when stop and go tendencies are balanced) on low-frequency rhythms in the SST. Stop trials were divided, according to the delays at which the stop signal occurred, into short-, intermediate-, and long-delay trials. Consistent with goal-conflict processing, intermediate-delay trials were associated with greater 7–8 Hz EEG power than short- or long-delay trials at medial right frontal sites (Fz, F4, and F8). At F8, 7–8 Hz power was linked to high trait anxiety and neuroticism. A separate 4–7 Hz power increase was also seen in stop, relative to go, trials, but this was independent of delay, was maximal at the central midline site Cz, and predicted faster stopping. Together with previous data on the SST, these results suggest that the right frontal region could be involved in multiple inhibition mechanisms. We propose a hierarchical model of the control of stopping that integrates the literature on the neural control of fast motor stopping with that on slower, motive-directed behavioural inhibition.

Approach–avoidance conflict is thought to generate negative affective bias, mediated by theta rhythms. This process is distinct from, and adds to, the effects of simple aversive input. We assessed this distinction by holding gain constant and increasing loss value so that conflict and simple aversion peaked in the conflict (gain equals loss) and loss (net loss) conditions, respectively. Right frontal areas showed increases in both conflict- and loss-induced theta power. However, loss, but not conflict, power was correlated with avoidance and neuroticism, showing a Gender x Hemisphere interaction. We concluded that multiple aversive processes converge in lateral frontal networks and that individual differences in theta response in these networks may reflect differences in behavioural and emotional reactivity to aversive events.

Bilateral vestibular lesions cause atrophy of the hippocampus in humans and subsequent deficits in spatial memory and the processing of emotional stimuli in both rats and humans. Vestibular lesions also impair hippocampal theta rhythm in rats. The aim of the present study was to investigate whether restoring theta rhythm to the hippocampus of a rat, via stimulation of the medial septum, would repair the deficits caused by vestibular lesions. It was hypothesized that the restoration of theta would repair the deficits and the vestibular rats would exhibit behavior and EEG similar to that of the sham rats. Rats were given either sham surgery or bilateral vestibular deafferentation (BVD) followed in a later operation by electrode implants. Half of the lesioned rats received stimulation. Subjects were tested in open field, elevated T-maze and spatial nonmatching to sample tests. BVD caused a deficit in hippocampal theta rhythm. Stimulation restored theta power at a higher frequency in the vestibular-lesioned rats, however, the stimulation did not repair the cognitive and emotional deficits caused by the lesions. It was concluded that stimulation, at least in the form used here, would not be a viable treatment option for vestibular damaged humans. © 2011 Wiley Periodicals, Inc.