Differentiation between finger, toe and tongue movement in man based on 40 Hz EEG

Primary somatosensory cortex sensitivity may increase upon completion of a motor task

OBJECTIVES

The electroencephalogram and magnetic field primary somatosensory cortex (S1)-derived components are attenuated before and during motor tasks compared to the resting state, a phenomenon called gating; however, the S1 response after a motor task has not been well studied. We aimed to investigate sensory information processing immediately after motor tasks using magnetoencephalography.

MATERIALS AND METHODS

We investigated sensory information processing immediately after finger movement using magnetoencephalography in 14 healthy adults. Volunteers performed a simple reaction task where they were required to press a button when they received a cue. In parallel, electrical stimulation to the right index finger was applied at regular intervals to detect the magnetic brain field changes. The end of the motor task timing was defined using the event-related synchronization (ERS) appearance latency in the brain magnetic field's beta band around the primary motor cortex. The ERS appearance latency and the sensory stimuli timing applied every 500 ms were synchronized over the experimental system timeline. We examined whether there was a difference in the S1 somatosensory evoked field responses between the ERS emergence and ERS disappearance phase, focusing on the N20m-P35m peak-to-peak amplitude (N20m-P35m amplitude) value. A control experiment was also conducted in which only sensory stimulation was applied with no motor task.

RESULTS

The N20m-P35m mean amplitude value was significantly higher in the ERS emergence phase (15.81 nAm; standard deviation [SD], 6.54 nAm) than in the ERS disappearance phase (13.54 nAm; SD, 5.12 nAm) (p < 0.05) and the control (12.08 nAm, SD 5.61 nAm) (p = 0.013). No statistically significant differences were identified between the ERS disappearance phase and the control (p = 0.281).

CONCLUSIONS

The S1 sensitivity may increase rapidly after exiting from the gating influence in S1 (after completing a motor task).

Age-related trends in neuromagnetic transient beta burst characteristics during a sensorimotor task and rest in the Cam-CAN open-access dataset

Non-invasive neurophysiological recordings, such as those measured by magnetoencelography (MEG), provide insight into the behaviour of neural networks and how these networks change with factors such as task performance, disease state, and age. Recently, there has been a trend in describing neurophysiological recordings as a series of transient bursts of neural activity rather than averaged sustained oscillations as burst characteristics may be more directly correlated with the neurological generators of brain activity. In this work, we investigate how beta burst characteristics change with age in a large open access dataset. The objectives are (1) to detect and characterize transient beta bursts over the ipsilateral and contralateral primary sensorimotor cortices during a unilateral motor task performance and during wakeful resting, and (2) to identify age-related changes in beta burst characteristics, in the context of earlier reports of age-related changes in beta suppression and the post-movement beta rebound. MEG data, acquired at the Cambridge Centre for Ageing and Neuroscience, of roughly 600 participants with a nearly uniform distribution of ages between 18 and 88 years old was used for analysis. We found that burst rate is the predominant factor related to age-related changes in the amplitude of the induced beta rhythm responses associated with a button press task. Furthermore, we present a cross-validation of burst parameters detected at the sensor- (peak sensor and sensor ROI) and source-level (beamformer spatial filter). This work is as an important step in characterizing transient bursts in neuromagnetic signals in the temporal domain, towards a better understanding of the healthy aging human brain.

Interpretable and lightweight convolutional neural network for EEG decoding: Application to movement execution and imagination

Convolutional neural networks (CNNs) are emerging as powerful tools for EEG decoding: these techniques, by automatically learning relevant features for class discrimination, improve EEG decoding performances without relying on handcrafted features. Nevertheless, the learned features are difficult to interpret and most of the existing CNNs introduce many trainable parameters. Here, we propose a lightweight and interpretable shallow CNN (Sinc-ShallowNet), by stacking a temporal sinc-convolutional layer (designed to learn band-pass filters, each having only the two cut-off frequencies as trainable parameters), a spatial depthwise convolutional layer (reducing channel connectivity and learning spatial filters tied to each band-pass filter), and a fully-connected layer finalizing the classification. This convolutional module limits the number of trainable parameters and allows direct interpretation of the learned spectral-spatial​ features via simple kernel visualizations. Furthermore, we designed a post-hoc gradient-based technique to enhance interpretation by identifying the more relevant and more class-specific features. Sinc-ShallowNet was evaluated on benchmark motor-execution and motor-imagery datasets and against different design choices and training strategies. Results show that (i) Sinc-ShallowNet outperformed a traditional machine learning algorithm and other CNNs for EEG decoding; (ii) The learned spectral-spatial features matched well-known EEG motor-related activity; (iii) The proposed architecture performed better with a larger number of temporal kernels still maintaining a good compromise between accuracy and parsimony, and with a trialwise rather than a cropped training strategy. In perspective, the proposed approach, with its interpretative capacity, can be exploited to investigate cognitive/motor aspects whose EEG correlates are yet scarcely known, potentially characterizing their relevant features.

Early gamma-oscillations as correlate of localized nociceptive processing in primary sensorimotor cortex

Recent studies put forward the idea that stimulus-evoked gamma-band oscillations (GBOs; 30-100 Hz) play a specific role in nociception. So far, evidence for the specificity of GBOs for nociception, their possible involvement in nociceptive sensory discriminatory abilities, and knowledge regarding their cortical sources is just starting to grow. To address these questions, we used electroencephalography (EEG) to record brain activity evoked by phasic nociceptive laser stimuli and tactile stimuli applied at different intensities to the right hand and foot of 12 healthy volunteers. The EEG was analyzed in the time domain to extract phase-locked event-related brain potentials (ERPs) and in three regions of interest in the time-frequency domain (delta/theta, 40-Hz gamma, 70-Hz gamma) to extract stimulus-evoked changes in the magnitude of non-phase-locked brain oscillations. Both nociceptive and tactile stimuli, matched with respect to subjective intensity, elicited phase locked ERPs of increasing amplitude with increasing stimulus intensity. In contrast, only nociceptive stimuli elicited a significant enhancement of GBOs (65-85 Hz, 150-230 ms after stimulus onset), whose magnitude encoded stimulus intensity, whereas tactile stimuli led to a GBO decrease. Following nociceptive hand stimulation, the topographical distribution of GBOs was maximal at contralateral electrode C3, whereas maximum activity following foot stimulation was recorded at the midline electrode Cz, compatible with generation of GBOs in the representations of the hand and foot of the primary sensorimotor cortex, respectively. The differential behavior of high-frequency GBOs and low-frequency 40-Hz GBOs is indicating different functional roles and regions in sensory processing.NEW & NOTEWORTHY Gamma-band oscillations show hand-foot somatotopy compatible with generation in primary sensorimotor cortex and are present following nociceptive but not tactile stimulation of the hand and foot in humans.

Evaluation of electroencephalography analysis methods

The extraction of expressive features from an electroencephalography (EEG) signal is necessary for classification of movement and movement imagination of the limbs. We introduce different preprocessing and feature extraction algorithms for this purpose and develop an algorithm that selects features by their feature importance. This selection is used as an evaluation measure for features, their preprocessing algorithms and the EEG electrodes. Our results show that most influential features for signal interpretation are: common spatial patterns, fractal dimensions, as well as, variance and standard deviation of the preprocessed data. We show that preprocessing with continuous wavelet transforms outperforms the other tested preprocessing algorithms. Furthermore, we show that high gamma frequencies (70-90 Hz) contain more information than the lower μ-rhythms (8-12 Hz) where event-related-desynchronization (ERD) is known to occur. The important EEG electrodes for this classification task are located in the left and right back of the motor-cortex. The proposed algorithm can be further used to create subject-specific and performance models for real-time classification.

Evidence for age-related changes in sensorimotor neuromagnetic responses during cued button pressing in a large open-access dataset

Mu, beta, and gamma rhythms increase and decrease in amplitude during movement. This event-related synchronization (ERS) and desynchronization (ERD) can be readily recorded non-invasively using magneto- and electro-encephalography (M/EEG). In addition, event-related potentials and fields (i.e., evoked responses) can be elucidated during movement. There is some evidence that the frequency, amplitude and latency of the movement-related ERS/ERD changes with ageing, however the evidence surrounding this topic comes mainly from studies in sample sizes on the order of tens of participants. The objective of this study was to examine a large open-access MEG dataset for age-related changes in movement-related ERS/ERD and evoked responses. MEG data acquired at the Cambridge Centre for Ageing and Neuroscience during cued button pressing was used from 567 participants between the ages of 18 and 88 years. The characteristics movement-related ERD/ERS and evoked responses were calculated for each individual participant. Based on linear regression analysis, significant relationships were found between participant age and some response characteristics, although the predictive value of these relationships was low. Specifically, we conclude that peak beta rebound frequency and amplitude decreased with age, peak beta suppression amplitude increased with age, movement-related gamma burst amplitude decreased with age, and peak motor-evoked response amplitude increased with age. Given our current understanding of the underlying mechanisms of these responses, our findings suggest the existence of age-related changes in the neurophysiology of thalamocortical loops and local circuitry in the primary somatosensory and motor cortices.

Deep learning with convolutional neural networks for EEG decoding and visualization

Deep learning with convolutional neural networks (deep ConvNets) has revolutionized computer vision through end-to-end learning, that is, learning from the raw data. There is increasing interest in using deep ConvNets for end-to-end EEG analysis, but a better understanding of how to design and train ConvNets for end-to-end EEG decoding and how to visualize the informative EEG features the ConvNets learn is still needed. Here, we studied deep ConvNets with a range of different architectures, designed for decoding imagined or executed tasks from raw EEG. Our results show that recent advances from the machine learning field, including batch normalization and exponential linear units, together with a cropped training strategy, boosted the deep ConvNets decoding performance, reaching at least as good performance as the widely used filter bank common spatial patterns (FBCSP) algorithm (mean decoding accuracies 82.1% FBCSP, 84.0% deep ConvNets). While FBCSP is designed to use spectral power modulations, the features used by ConvNets are not fixed a priori. Our novel methods for visualizing the learned features demonstrated that ConvNets indeed learned to use spectral power modulations in the alpha, beta, and high gamma frequencies, and proved useful for spatially mapping the learned features by revealing the topography of the causal contributions of features in different frequency bands to the decoding decision. Our study thus shows how to design and train ConvNets to decode task-related information from the raw EEG without handcrafted features and highlights the potential of deep ConvNets combined with advanced visualization techniques for EEG-based brain mapping. Hum Brain Mapp 38:5391-5420, 2017. © 2017 Wiley Periodicals, Inc.

Beta-band activity and connectivity in sensorimotor and parietal cortex are important for accurate motor performance

Accurate motor performance may depend on the scaling of distinct oscillatory activity within the motor cortex and effective neural communication between the motor cortex and other brain areas. Oscillatory activity within the beta-band (13-30Hz) has been suggested to provide distinct functional roles for attention and sensorimotor control, yet it remains unclear how beta-band and other oscillatory activity within and between cortical regions is coordinated to enhance motor performance. We explore this open issue by simultaneously measuring high-density cortical activity and elbow flexor and extensor neuromuscular activity during ballistic movements, and manipulating error using high and low visual gain across three target distances. Compared with low visual gain, high visual gain decreased movement errors at each distance. Group analyses in 3D source-space revealed increased theta-, alpha-, and beta-band desynchronization of the contralateral motor cortex and medial parietal cortex in high visual gain conditions and this corresponded to reduced movement error. Dynamic causal modeling was used to compute connectivity between motor cortex and parietal cortex. Analyses revealed that gain affected the directionally-specific connectivity across broadband frequencies from parietal to sensorimotor cortex but not from sensorimotor cortex to parietal cortex. These new findings provide support for the interpretation that broad-band oscillations in theta, alpha, and beta frequency bands within sensorimotor and parietal cortex coordinate to facilitate accurate upper limb movement.

SUMMARY STATEMENT

Our findings establish a link between sensorimotor oscillations in the context of online motor performance in common source space across subjects. Specifically, the extent and distinct role of medial parietal cortex to sensorimotor beta connectivity and local domain broadband activity combine in a time and frequency manner to assist ballistic movements. These findings can serve as a model to examine whether similar source space EEG dynamics exhibit different time-frequency changes in individuals with neurological disorders that cause movement errors.

Spatial Inhibition of Return promotes changes in response-related mu and beta oscillatory patterns

The possible role that response processes play in Inhibition of Return (IOR), traditionally associated with reduced or inhibited attentional processing of spatially cued target stimuli presented at cue-target intervals longer than 300 ms, is still under debate. Previous psychophysiological studies on response-related Electroencephalographic (EEG) activity and IOR have found divergent results. Considering that the ability to optimize our behavior not only resides in our capacity to inhibit the focus of attention from irrelevant information but also to inhibit or reduce motor activation associated with responses to that information, it is conceivable that response processes are also affected by IOR. In the present study, time-frequency (T-F) analyses were performed on EEG oscillatory activity between 2 and 40 Hz to check whether spatial IOR affects response preparation and execution during a visuospatial attention task. To avoid possible spatial stimulus-response compatibility effects and their interaction with the IOR effects, the stimuli were presented along the vertical meridian of the visual field. The results differed between lower and upper visual fields. In the lower visual field spatial IOR was related to a synchronization in the pre-movement mu band at bilateral precentral and central electrodes, and in the post-movement beta band at contralateral precentral and central electrodes, which may be associated with an attention-driven reduction of somatomotor processing prior to the execution of responses to relevant stimuli presented at previously cued locations followed by a post-movement deactivation of motor areas. In the upper visual field, spatial IOR was associated with a decrease in desynchronization around response execution in the beta band at contralateral postcentral electrodes that might indicate a late (last moment) reduction of motor activation when responding to spatially cued targets. The present results suggest that different response processes are affected by spatial IOR depending on the visual field where the target is presented.

Classification of hand movement direction based on EEG high-gamma activity

The Electroencephalogram (EEG) is a non-invasive technique used in the medical field to record and analyze brain activity. In particular, Brain Machine Interfaces (BMI) create this bridge between brain signals and the external world through prosthesis, visual interfaces and other physical devices. This paper investigates the relation between particular hand movement directions while using a BMI and the EEG recordings during such movement. The Common Spatial Pattern method (CSP) over the high-γ frequency band is utilized in order to discriminate opposite hand movement directions. The experiment is performed with three subjects and the average classification accuracy is obtained for two different cases.

Electrophysiological CNS-processes related to associative learning in humans

The neurophysiology of human associative memory has been studied with electroencephalographic techniques since the 1930s. This research has revealed that different types of electrophysiological processes in the human brain can be modified by conditioning: sensory evoked potentials, sensory induced gamma-band activity, periods of frequency-specific waves (alpha and beta waves, the sensorimotor rhythm and the mu-rhythm) and slow cortical potentials. Conditioning of these processes has been studied in experiments that either use operant conditioning or repeated contingent pairings of conditioned and unconditioned stimuli (classical conditioning). In operant conditioning, the appearance of a specific brain process is paired with an external stimulus (neurofeedback) and the feedback enables subjects to obtain varying degrees of control of the CNS-process. Such acquired self-regulation of brain activity has found practical uses for instance in the amelioration of epileptic seizures, Autism Spectrum Disorders (ASD) and Attention Deficit Hyperactivity Disorder (ADHD). It has also provided communicative means of assistance for tetraplegic patients through the use of brain computer interfaces. Both extra and intracortically recorded signals have been coupled with contingent external feedback. It is the aim for this review to summarize essential results on all types of electromagnetic brain processes that have been modified by classical or operant conditioning. The results are organized according to type of conditioned EEG-process, type of conditioning, and sensory modalities of the conditioning stimuli.

3D Cortical electrophysiology of ballistic upper limb movement in humans

Precise motor control requires the ability to scale the parameters of movement. Theta oscillations across the cortex have been associated with changes in memory, attention, and sensorimotor processing. What has proven more elusive is pinpointing the region-specific frequency band oscillations that are associated with specific parameters of movement during the acceleration and deceleration phases. We report a study using 3D analytic techniques for high density electroencephalography that examines electrocortical dynamics while participants produce upper limb movements to different distances at varying rates. During fast ballistic movements, we observed increased theta band activity in the left motor area contralateral to the moving limb during the acceleration phase of the movement, and theta power correlated with the acceleration of movement. In contrast, beta band activity scaled with the type of movement during the deceleration phase near the end of the movement and correlated with movement time. In the ipsilateral motor and somatosensory area, alpha band activity decreased with the type of movement near the end of the movement, and gamma band activity in visual cortex increased with the type of movement near the end of the movement. Our results suggest that humans use distinct lateralized cortical activity for distance and speed dependent arm movements. We provide new evidence that a temporary increase in theta band power relates to movement acceleration and is important during movement execution. Further, the theta power increase is coupled with desychronization of beta band power and alpha band power which are modulated by the task near the end of movement.

Quality parameters for a multimodal EEG/EMG/kinematic brain-computer interface (BCI) aiming to suppress neurological tremor in upper limbs

Tremor is the most common movement disorder encountered during daily neurological practice. Tremor in the upper limbs causes functional disability and social inconvenience, impairing daily life activities. The response of tremor to pharmacotherapy is variable. Therefore, a combination of drugs is often required. Surgery is considered when the response to medications is not sufficient. However, about one third of patients are refractory to current treatments. New bioengineering therapies are emerging as possible alternatives. Our study was carried out in the framework of the European project “Tremor” (ICT-2007-224051). The main purpose of this challenging project was to develop and validate a new treatment for upper limb tremor based on the combination of functional electrical stimulation (FES; which has been shown to reduce upper limb tremor) with a brain-computer interface (BCI). A BCI-driven detection of voluntary movement is used to trigger FES in a closed-loop approach. Neurological tremor is detected using a matrix of EMG electrodes and inertial sensors embedded in a wearable textile. The identification of the intentionality of movement is a critical aspect to optimize this complex system. We propose a multimodal detection of the intentionality of movement by fusing signals from EEG, EMG and kinematic sensors (gyroscopes and accelerometry). Parameters of prediction of movement are extracted in order to provide global prediction plots and trigger FES properly. In particular, quality parameters (QPs) for the EEG signals, corticomuscular coherence and event-related desynchronization/synchronization (ERD/ERS) parameters are combined in an original algorithm which takes into account the refractoriness/responsiveness of tremor. A simulation study of the relationship between the threshold of ERD/ERS of artificial EEG traces and the QPs is also provided. Very interestingly, values of QPs were much greater than those obtained for the corticomuscular module alone.

No somatotopy of sensorimotor alpha-oscillation responses to differential finger stimulation

The somatotopic layout of the primary somatosensory cortex is known for its fine spatial structure as delineated in single cell recordings and macroscopic EEG evoked responses. While a gross somatotopic layout has been revealed also for neuronal oscillations responding to sensorimotor stimulation of distant body parts (e.g. hand vs. foot), it is still unclear whether these oscillatory dynamics exhibit fine spatial layout comparable to those found in evoked responses. In twelve healthy subjects we applied electric stimuli to the first (D1) and fifth finger (D5) of the same hand while performing high-density electroencephalography. We used Common Spatial Pattern analysis to optimally extract components showing the strongest Event-Related Desynchronization (ERD) in neuronal alpha oscillations. In agreement with the previous studies, dipole locations of Somatosensory Evoked Potentials (SEPs) confirmed the existence of spatially distinct representations of each finger. In contrast, dipole locations of alpha-ERD patterns did not yield spatially different source locations, indicating that the stimulation of different fingers most likely resulted in oscillatory activity of overlapping neuronal populations. When both fingers were stimulated simultaneously the SEP dipole strength was found increased in comparison to a stimulation of either finger alone, in agreement with spatially distinct SEP to finger stimulation. The strength of ERD, on the other hand, was the same regardless of whether either one or both fingers were stimulated. Our findings might reflect anatomical constraints on the sequential temporal activation of fingers' skin where almost simultaneous activation of many fingers usually occurs in everyday activities, such as grasping or holding objects. Such simultaneity is unlikely to benefit from slow amplitude modulation of alpha oscillations, which would rather be beneficial for contrasting somatosensory processing of distinct body parts.

Enhanced phase and amplitude synchronization toward focal seizure offset

Recent studies involving individual neurons in the seizure focal and surrounding areas have established heterogeneous firing patterns in single cells. However, the patterns become more homogeneous approaching the seizure offset. In this article, we show that similar observations are possible from intracranial recording if the right quantitative or engineering techniques are used. We have observed an increase in Hilbert transformation-based phase synchronization in the focal electrocorticoencehalogram (ECoG) in the gamma band (30-40 Hz) towards the end of the majority of focal epileptic seizures. An amplitude correlation measure shows an enhanced principal component (and hence enhanced correlation among the channels involved) approaching the offset of the large majority of seizures. Surprisingly, there are seizures which show the enhanced phase synchronization approaching offset but no enhanced amplitude correlation during the same period and vice versa. This study shows that suitable computational tools can sometimes compensate for more expensive and technologically demanding data acquisition systems. A possible neurophysiological explanation behind the observed phenomenon is also presented.

Audio-vocal monitoring system revealed by mu-rhythm activity

Understanding the neural mechanisms underlying speech production has a number of potential practical applications. Speech production involves multiple feedback loops. An audio-vocal monitoring system plays an important role in speech production, based on auditory feedback about the speaker’s own voice. Here we investigated the mu-rhythm activity associated with speech production by examining event-related desynchronization and synchronization in conditions of delayed auditory feedback (DAF) and noise feedback (Lombard). In Experiment 1, we confirmed that the mu-rhythms were detectable for a conventional finger-tapping task, and vocalization. In Experiment 2, we examined the mu-rhythms for imagined speech production. We tested whether the same motor-related mu-rhythm activity was exhibited while participants listened to their own voice, and while reading. The mu-rhythms were observed for overt vocalization and covert reading, while listening to simulated auditory feedback of the participants’ own voice reading text. In addition, we found that the mu-rhythm associated with listening was boosted and attenuated under the DAF and Lombard conditions, respectively. This is consistent with the notion that auditory feedback is important for the audio-vocal monitoring system in speech production. This paradigm may help clarify the way in which auditory feedback supports motor planning, as indexed by the motor-related mu-rhythm.

A quality parameter for the detection of the intentionality of movement in patients with neurological tremor performing a finger-to-nose test

The identification of the intentionality of movement is a key-aspect for the development of brain-computer interfaces (BCIs) applicable to daily life in neurological patients. We present a novel method of processing of electroencephalography (EEG) signals for the extraction of movement intention in neurological patients with upper limb tremor. This method is based on event-related EEG desynchronization, considering α (8-12 Hz), β (13-30 Hz), and γ (30-40 Hz) bands. We have analyzed the EEG signals from the sensorimotor areas of 4 neurological patients presenting an upper limb tremor (grade 1 to 3/4) and executing successive finger-to-nose movements. A Quality Parameter (QP) for the detection of intentionality of movement has been extracted, by considering: (a) the changes in the β²/α and β/α ratio (representing bursts of β-γ frequencies) during the pre-movement period; (b) an appropriate threshold predicting the movement; (c) the number of movements executed. This QP allows the prediction of the voluntary movement with a probability between 70% and 90%. This method could be implemented in a wearable BCI to detect the intentionality of movement and could be used, for instance, to trigger the electrical stimulation in selected muscles of upper limbs with the aim of blocking the emergence of tremor.

Sparse gamma rhythms arising through clustering in adapting neuronal networks

Gamma rhythms (30-100 Hz) are an extensively studied synchronous brain state responsible for a number of sensory, memory, and motor processes. Experimental evidence suggests that fast-spiking interneurons are responsible for carrying the high frequency components of the rhythm, while regular-spiking pyramidal neurons fire sparsely. We propose that a combination of spike frequency adaptation and global inhibition may be responsible for this behavior. Excitatory neurons form several clusters that fire every few cycles of the fast oscillation. This is first shown in a detailed biophysical network model and then analyzed thoroughly in an idealized model. We exploit the fact that the timescale of adaptation is much slower than that of the other variables. Singular perturbation theory is used to derive an approximate periodic solution for a single spiking unit. This is then used to predict the relationship between the number of clusters arising spontaneously in the network as it relates to the adaptation time constant. We compare this to a complementary analysis that employs a weak coupling assumption to predict the first Fourier mode to destabilize from the incoherent state of an associated phase model as the external noise is reduced. Both approaches predict the same scaling of cluster number with respect to the adaptation time constant, which is corroborated in numerical simulations of the full system. Thus, we develop several testable predictions regarding the formation and characteristics of gamma rhythms with sparsely firing excitatory neurons.

Cortical γ responses: searching high and low

In this paper, a brief, preliminary attempt is made to frame a scientific debate about how functional responses at gamma frequencies in electrophysiological recordings (EEG, MEG, ECoG, and LFP) should be classified and interpreted. In general, are all gamma responses the same, or should they be divided into different classes according to criteria such as their spectral characteristics (frequency range and/or shape), their spatial-temporal patterns of occurrence, and/or their responsiveness under different task conditions? In particular, are the responses observed in intracranial EEG at a broad range of "high gamma" frequencies (~60-200Hz) different from gamma responses observed at lower frequencies (~30-80Hz), typically in narrower bands? And if they are different, how should they be interpreted? Does the broad spectral shape of high gamma responses arise from the summation of many different narrow-band oscillations, or does it reflect something completely different? If we are not sure, should we refer to high gamma activity as oscillations? A variety of theories have posited a mechanistic role for gamma activity in cortical function, often assuming narrow-band oscillations. These theories continue to influence the design of experiments and the interpretation of their results. Do these theories apply to all electrophysiological responses at gamma frequencies? Although no definitive answers to these questions are immediately anticipated, this paper will attempt to review the rationale for why they are worth asking and to point to some of the possible answers that have been proposed.

Supplementary motor area exerts proactive and reactive control of arm movements

Adaptive behavior requires the ability to flexibly control actions. This can occur either proactively to anticipate task requirements, or reactively in response to sudden changes. Here we report neuronal activity in the supplementary motor area (SMA) that is correlated with both forms of behavioral control. Single-unit and multiunit activity and intracranial local field potentials (LFPs) were recorded in macaque monkeys during a stop-signal task, which elicits both proactive and reactive behavioral control. The LFP power in high- (60-150 Hz) and low- (25-40 Hz) frequency bands was significantly correlated with arm movement reaction time, starting before target onset. Multiunit and single-unit activity also showed a significant regression with reaction time. In addition, LFPs and multiunit and single-unit activity changed their activity level depending on the trial history, mirroring adjustments on the behavioral level. Together, these findings indicate that neuronal activity in the SMA exerts proactive control of arm movements by adjusting the level of motor readiness. On trials when the monkeys successfully canceled arm movements in response to an unforeseen stop signal, the LFP power, particularly in a low (10-50 Hz) frequency range, increased early enough to be causally related to the inhibition of the arm movement on those trials. This indicated that neuronal activity in the SMA is also involved in response inhibition in reaction to sudden task changes. Our findings indicate, therefore, that SMA plays a role in the proactive control of motor readiness and the reactive inhibition of unwanted movements.

A generalized framework for quantifying the dynamics of EEG event-related desynchronization

Brains were built by evolution to react swiftly to environmental challenges. Thus, sensory stimuli must be processed ad hoc, i.e., independent—to a large extent—from the momentary brain state incidentally prevailing during stimulus occurrence. Accordingly, computational neuroscience strives to model the robust processing of stimuli in the presence of dynamical cortical states. A pivotal feature of ongoing brain activity is the regional predominance of EEG eigenrhythms, such as the occipital alpha or the pericentral mu rhythm, both peaking spectrally at 10 Hz. Here, we establish a novel generalized concept to measure event-related desynchronization (ERD), which allows one to model neural oscillatory dynamics also in the presence of dynamical cortical states. Specifically, we demonstrate that a somatosensory stimulus causes a stereotypic sequence of first an ERD and then an ensuing amplitude overshoot (event-related synchronization), which at a dynamical cortical state becomes evident only if the natural relaxation dynamics of unperturbed EEG rhythms is utilized as reference dynamics. Moreover, this computational approach also encompasses the more general notion of a “conditional ERD,” through which candidate explanatory variables can be scrutinized with regard to their possible impact on a particular oscillatory dynamics under study. Thus, the generalized ERD represents a powerful novel analysis tool for extending our understanding of inter-trial variability of evoked responses and therefore the robust processing of environmental stimuli.

When Hans Berger described the human EEG in the 1920s, a pivotal finding was the demonstration of prominent oscillations in the frequency range between 8 and 12 Hz, which he called alpha wave rhythm. He also described for the first time the so-called “alpha blockade,” i.e., the suppression of the ongoing alpha activity when the subject opens his eyes. Based on these early findings, induced changes of macroscopic EEG oscillations have been reported for diverse physiological manipulations and processing of sensory information. The magnitude and the latency of these induced changes are, however, subject to variations, even if identical stimuli are processed. In order to enable investigations of the underlying neural mechanisms of these variations, we here establish a mathematical framework which allows one to scrutinize candidate explanatory factors with regard to their possible impact on the characteristics of the induced oscillatory dynamics.

Cross-frequency phase coupling of brain rhythms during the orienting response

A critical function of the brain's orienting response is to evaluate novel environmental events in order to prepare for potential behavioral action. Here, measures of synchronization (power, coherence) and nonlinear cross-frequency phase coupling (m:n phase locking measured with bicoherence and cross-bicoherence) were computed on 62-channel electroencephalographic (EEG) data during a paradigm in which unexpected, highly-deviant, novel sounds were randomly intermixed with frequent standard and infrequent target tones. Low frequency resolution analyses showed no significant changes in phase coupling for any stimulus type, though significant changes in power and synchrony did occur. High frequency resolution analyses, on the other hand, showed significant differences in phase coupling, but only for novel sounds compared to standard tones. Novel sounds elicited increased power and coherence in the delta band together with m:n phase locking (bicoherence) of delta:theta (1:3) and delta:alpha (1:4) rhythms in widespread fronto-central, right parietal, temporal, and occipital regions. Cross-bicoherence revealed that globally synchronized delta oscillations were phase coupled to theta oscillations in central regions and to alpha oscillations in right parietal and posterior regions. These results suggest that globally synchronized low frequency oscillations with phase coupling to more localized higher frequency oscillations provide a neural mechanism for the orienting response.

A survey of signal processing algorithms in brain-computer interfaces based on electrical brain signals

Brain-computer interfaces (BCIs) aim at providing a non-muscular channel for sending commands to the external world using the electroencephalographic activity or other electrophysiological measures of the brain function. An essential factor in the successful operation of BCI systems is the methods used to process the brain signals. In the BCI literature, however, there is no comprehensive review of the signal processing techniques used. This work presents the first such comprehensive survey of all BCI designs using electrical signal recordings published prior to January 2006. Detailed results from this survey are presented and discussed. The following key research questions are addressed: (1) what are the key signal processing components of a BCI, (2) what signal processing algorithms have been used in BCIs and (3) which signal processing techniques have received more attention?

EMG and EOG artifacts in brain computer interface systems: A survey

It is widely accepted in the brain computer interface (BCI) research community that neurological phenomena are the only source of control in any BCI system. Artifacts are undesirable signals that can interfere with neurological phenomena. They may change the characteristics of neurological phenomena or even be mistakenly used as the source of control in BCI systems. Electrooculography (EOG) and electromyography (EMG) artifacts are considered among the most important sources of physiological artifacts in BCI systems. Currently, however, there is no comprehensive review of EMG and EOG artifacts in BCI literature. This paper reviews EOG and EMG artifacts associated with BCI systems and the current methods for dealing with them. More than 250 refereed journal and conference papers are reviewed and categorized based on the type of neurological phenomenon used and the methods employed for handling EOG and EMG artifacts. This study reveals weaknesses in BCI studies related to reporting the methods of handling EMG and EOG artifacts. Most BCI papers do not report whether or not they have considered the presence of EMG and EOG artifacts in the brain signals. Only a small percentage of BCI papers report automated methods for rejection or removal of artifacts in their systems. As the lack of dealing with artifacts may result in the deterioration of the performance of a particular BCI system during practical applications, it is necessary to develop automatic methods to handle artifacts or to design BCI systems whose performance is robust to the presence of artifacts.

Learning in human neural networks on microelectrode arrays

This paper describes experiments involving the growth of human neural networks of stem cells on a MEA (microelectrode array) support. The microelectrode arrays (MEAs) are constituted by a glass support in which a set of tungsten electrodes are inserted. The artificial neural network (ANN) paradigm was used by stimulating the neurons in parallel with digital patterns distributed on eight channels, then by analyzing a parallel multichannel output. In particular, the microelectrodes were connected following two different architectures, one inspired by the Kohonen's SOM, the other by the Hopfield network. The output signals have been analyzed in order to evaluate the possibility of organized reactions by the natural neurons.f The results show that the network of human neurons reacts selectively to the subministered digital signals, i.e., it produces similar output signals referred to identical or similar patterns, and clearly differentiates the outputs coming from different stimulations. Analyses performed with a special artificial neural network called ITSOM show the possibility to codify the neural responses to different patterns, thus to interpret the signals coming from the network of biological neurons, assigning a code to each output. It is straightforward to verify that identical codes are generated by the neural reactions to similar patterns. Further experiments are to be designed that improve the hybrid neural networks' capabilities and to test the possibility of utilizing the organized answers of the neurons in several ways.

Estimation of short-time cross-correlation between frequency bands of event related EEG

Simultaneous variations of the event-related power changes (ERD/ERS) are often observed in a number of frequency bands. ERD/ERS measures are usually based on the relative changes of power in a given single frequency band. Within such an approach one cannot answer questions concerning the mutual relations between the band-power variations observed in different frequency bands. This paper addresses the problem of estimating and assessing the significance of the average cross-correlation between ERD/ERS phenomena occurring in two frequency bands. The cross-correlation function in a natural way also provides estimation of the delay between ERD/ERS in those bands. The proposed method is based on estimating the short-time cross-correlation function between relative changes of power in two selected frequency bands. The cross-correlation function is estimated in each trial separately and then averaged across trials. The significance of those mean cross-correlation functions is evaluated by means of a nonparametric test. The basic properties of the method are presented on simulated signals, and an example application to real EEG and ECoG signals is given.

A model of event-related EEG synchronization changes in beta and gamma frequency bands

During preparation, execution and recovery from simple movements, the EEG power spectrum undergoes a sequence of changes. The power in the beta band (13-25 Hz) decreases during preparation and execution of movement, but during recovery it reaches a level higher than that in the reference period (not affected by the event). These effects are known as event-related beta desynchronization and beta rebound. The power in the gamma band (>30 Hz) increases significantly just before the onset of the movement. This effect is known as event-related gamma synchronization. There are numerous observations concerning these effects but the underlying physiological mechanisms and functional role are not clear. We propose a lumped computational model of a cortical circuit. The model consists only of a pyramidal and an interneuronal population. Each population represents averaged properties of constituting neurons. The output of the model represents a local field potential, with a power spectrum peak either in the beta or in the gamma band. The model elucidates the mechanisms of transition between slower and faster rhythms, gamma synchronization and beta desynchronization and rebound effects. The sufficient conditions to observe the effects in the model are changes of the external excitation level and of the connection strength between excitatory and inhibitory populations attributed to short-time plasticity. The present model presents the role of the pyramidal neurons to interneuron connection in the oscillatory behavior of the two populations. We conclude that the pronounced facilitation of the pyramidal to fast spiking interneuron connections, initiated by robust excitation of the motor cortex neurons, may be essential for the effect of beta rebound. Further experiments concerning short-time plasticity during behavioral tasks would be of great value in studies of functional local cortical circuits.

High-frequency gamma oscillations and human brain mapping with electrocorticography

Invasive EEG recordings with depth and/or subdural electrodes are occasionally necessary for the surgical management of patients with epilepsy refractory to medications. In addition to their vital clinical utility, electrocorticographic (ECoG) recordings provide an unprecedented opportunity to study the electrophysiological correlates of functional brain activation in greater detail than non-invasive recordings. The proximity of ECoG electrodes to the cortical sources of EEG activity enhances their spatial resolution, as well as their sensitivity and signal-to-noise ratio, particularly for high-frequency EEG activity. ECoG recordings have, therefore, been used to study the event-related dynamics of brain oscillations in a variety of frequency ranges, and in a variety of functional-neuroanatomic systems, including somatosensory and somatomotor systems, visual and auditory perceptual systems, and cortical networks responsible for language. These ECoG studies have confirmed and extended the original non-invasive observations of ERD/ERS phenomena in lower frequencies, and have discovered novel event-related responses in gamma frequencies higher than those previously observed in non-invasive recordings. In particular, broadband event-related gamma responses greater than 60 Hz, extending up to approximately 200 Hz, have been observed in a variety of functional brain systems. The observation of these "high gamma" responses requires a recording system with an adequate sampling rate and dynamic range (we use 1000 Hz at 16-bit A/D resolution) and is facilitated by event-related time-frequency analyses of the recorded signals. The functional response properties of high-gamma activity are distinct from those of ERD/ERS phenomena in lower frequencies. In particular, the timing and spatial localization of high-gamma ERS often appear to be more specific to the putative timing and localization of functional brain activation than alpha or beta ERD/ERS. These findings are consistent with the proposed role of synchronized gamma oscillations in models of neural computation, which have in turn been inspired by observations of gamma activity in animal preparations, albeit at somewhat lower frequencies. Although ECoG recordings cannot directly measure the synchronization of action potentials among assemblies of neurons, they may demonstrate event-related interactions between gamma oscillations in macroscopic local field potentials (LFP) generated by different large-scale populations of neurons engaged by the same functional task. Indeed, preliminary studies suggest that such interactions do occur in gamma frequencies, including high-gamma frequencies, at latencies consistent with the timing of task performance. The neuronal mechanisms underlying high-gamma activity and its unique response properties in humans are still largely unknown, but their investigation through invasive methods is expected to facilitate and expand their potential clinical and research applications, including functional brain mapping, brain-computer interfaces, and neurophysiological studies of human cognition.

Differential conditioning of alpha amplitude: A fresh look at an old phenomenon

OBJECTIVE

To determine the latency and development of conditional suppression of alpha amplitude and its relationship to behaviour, alpha amplitude (8-13 Hz) was measured in a differential conditioning procedure.

METHODS

The CS+/- were tones and the US was a photic checkerboard. Alpha amplitude, CNV, RT and verbal responses were recorded from 12 participants.

RESULTS

The CS+/- difference in acquisition was greatest from 250 ms before the US. It was greatest from the trial where RT declined and participants could report the CS+/US relationship. There was an amplitude increase in lower band activity 230 ms after the US. This looked like a VEP but was produced by phase-locked activity starting before the US.

CONCLUSIONS

Predicting the US led to cortical priming. Amplitude change in acquisition is congruent with CNV, RT and verbal performance.

SIGNIFICANCE

Prediction, expectancy and motor preparation are reflected in changes in alpha activity. These results provide converging evidence for the functional role of 8-10 Hz activity. They complement the emerging picture of the role of alpha activity in cognition, indicating that it extends to the acquisition of predictive knowledge.

Motor cortex dynamics in visuomotor production of speech and non-speech mouth movements

We investigated timing and hemispheric balance of motor cortex activation when kinetically similar speech and non-speech mouth movements and sequences of such movements were triggered by visually presented letter- and symbol-strings. As an index of motor cortex activation, we used magnetoencephalographic recording of task-related change of precentral 20 Hz (16-24 Hz) activity. Suppression of the 20 Hz rhythm revealed pre-movement activation in the face representation areas that was tied to visual instruction, not movement onset. The 20 Hz rhythm remained suppressed throughout the preparation and execution of mouth movements and was followed by post-movement rebound. Left hemisphere preceded the right at the onset and offset of the suppression, similarly for isolated and sequential speech and non-speech movements. Pattern of task-related change in 20 Hz activity was otherwise symmetrical. In the face areas, the overall modulation of 20 Hz activity increased with sequence length and motor demands. Hand representation areas showed also weak reactivity, with systematically larger modulation of 20 Hz activity for non-speech than speech movements. Our results suggest an active role for the motor cortex in cognitive control of visually triggered mouth movements, not limited to movement execution.

Computationally efficient approaches to calculating significant ERD/ERS changes in the time-frequency plane

This paper addresses some practical issues related to the calculation, display and assessment of the significance of changes in the average time-frequency energy density of event-related brain activity. Using scalp EEG and subdural ECoG example datasets, parametric tests are evaluated as a replacement for previously applied computer-intensive resampling methods. The performance of different estimates of energy density, based on matching pursuit, scalogram and spectrogram, and their Box-Cox transformations is evaluated with respect to the assumption of normality required for the t-test, and the consistency of the final results.

A neuroanatomical model of passivity phenomena

Any attempt to elucidate the nature and mechanism of passivity phenomena, i.e., experiences that one's conscious actions or thoughts have not been 'willed' by oneself, requires an integrative philosophical-neurobiological approach. The model proposed here adopts some fundamental positions that have long been advocated by philosophers and theoretical psychologists and have now found support from functional neuroanatomy. First, we experience our actions not from the standpoint of the executive but through the perception of its effects. Second, the 'self' is not an agent of behaviour. Third, behaviour is energized and integrated by basic drives (instincts). Fourth, the view that the experience of an acting self is related to drive reduction associated with voluntary actions is perhaps less well developed. The model thus proposes that passivity phenomena are actions that are induced by the perception of salient events but that are not integrated with or conducive to the overall motivational state of the organism. It has been suggested that, following the perception of salient events, competition arises between automatic response tendencies seeking expression. The prefrontal cortex appears to play an important role not only in determining which events are to be perceived but also which of the corresponding response dispositions is to be selected and actualized in overt behaviour. Thus, action selection is the outcome of competition between response tendencies in the context of prefrontal biasing signals that represent drives and strivings for goals. Action selection may be uncoupled from drives and strivings as a result of a lowering of the threshold for action selection--as is suggested to be the case in schizophrenic passivity phenomena--or due to disconnection from prefrontal regions--as may be the case in the alien limb syndrome.

Reciprocal modulation of neuromagnetic induced gamma activity by attention in the human visual and auditory cortex

For attentional control of behavior, the brain permanently resolves a competition between the impressions supplied by different senses. Here, using a dual-modality temporal order detection task, we studied attentional modulation of oscillatory neuromagnetic activity in the human cerebral cortex. On each trial, after simultaneous exposure to visual and auditory noise, subjects were presented with an asynchronous pair of a visual and an auditory stimulus. Either of the two stimuli could occur first equally often, their order was not cued. Subjects had to determine the leading stimulus in a pair and attentively monitor it to respond upon its offset. With the attended visual or auditory stimuli, spectral power analysis revealed marked enhancements of induced gamma activity within 250 ms post-stimulus onset over the modality-specific cortices (occipital at 64 Hz, right temporal at 53 Hz). When unattended, however, the stimuli led to a significantly decreased (beneath baseline) gamma response in these cortical regions. The gamma decreases occurred at lower frequencies ( approximately 30 Hz) than did the gamma increases. An increase in the gamma power and frequency for the attended modality and their decrease for the unattended modality suggest that attentional regulation of multisensory processing involves reciprocal changes in synchronization of respective cortical networks. We assume that the gamma decrease reflects an active suppression of the task-irrelevant sensory input. This suppression occurs at lower frequencies, suggesting an involvement of larger scale cell assemblies.

Halothane augments event-related γ oscillations in rat visual cortex

Cortical gamma oscillations have been associated with neural processes supporting cognition and the state of consciousness but the effect of general anesthesia on gamma oscillations is controversial. Here we studied the concentration-dependent effect of halothane on gamma (20-60 Hz) power of event-related potentials (ERP) in rat primary visual cortex. ERP to light flashes repeated at 5-s intervals was recorded with chronically implanted, bipolar, intracortical electrodes at selected steady-state halothane concentrations between 0 and 2%. gamma-Band power was calculated for 0-1000, 0-300 and 300-1000 ms poststimulus periods and corresponding prestimulus (PS) periods. Multitaper power spectral analysis was used to estimate gamma power from both single-trial and average ERP in order to differentiate between phase-locked (evoked) and non-phase-locked (induced) gamma activities. Significant PS gamma power was present at all halothane concentrations. Flash elicited an increase in gamma power that lasted up to 1 s poststimulus at all halothane concentrations. Halothane at intermediate concentrations (0.5-1.2%) augmented both PS and ERP gamma power two to four times relative to the waking baseline. gamma Power was not different between waking and deeply anesthetized (2%) levels. gamma Power reached maximum, as predicted by a Gaussian fit of power-concentration data, at halothane concentration (0.86%) similar to the concentration (0.73%) that abolished the righting reflex, a behavioral index of loss of consciousness. Evoked, i.e. stimulus-locked, gamma power was present during the first 300 ms poststimulus but not later, and was approximately 50% of single-trial ERP gamma power. Single-trial gamma power was present also at 300-1000 ms poststimulus, reflecting ERP not phase-locked to the stimulus. In summary, these observations suggest that (1) gamma activity is present in states ranging from waking to deep halothane anesthesia, (2) halothane does not prevent the transfer of visual input to striate cortex even at surgical plane of anesthesia, and (3) anesthetic-induced loss of consciousness, as reflected by the loss of righting reflex, is not correlated with a reduction in gamma power. Variance with other studies may be due to an underestimation of gamma power by ERP signal averaging as compared with single-trial analysis.

A general framework for brain-computer interface design

The Brain-Computer Interface (BCI) research community has acknowledged that researchers are experiencing difficulties when they try to compare the BCI techniques described in the literature. In response to this situation, the community has stressed the need for objective methods to compare BCI technologies. Suggested improvements have included the development and use of benchmark applications and standard data sets. However, as a young, multidisciplinary research field, the BCI community lacks a common vocabulary. As a result, this deficiency leads to poor intergroup communication, which hinders the development of the desired methods of comparison. One of the principle reasons for the lack of common vocabulary is the absence of a common functional model of a BCI System. This paper proposes a new functional model for BCI System design. The model supports many features that facilitate the comparison of BCI technologies with other BCI and non-BCI user interface technologies. From this model, taxonomy for BCI System design is developed. Together the model and taxonomy are considered a general framework for BCI System design. The representational power of the proposed framework was evaluated by applying it to a set of existing BCI technologies. The framework could effectively describe all of the BCI System designs tested.

Synchronous gamma activity: a review and contribution to an integrative neuroscience model of schizophrenia

Synchronous high frequency (Gamma band) activity has been proposed as a candidate mechanism for the integration or 'binding' of distributed brain activities. Since the first descriptions of schizophrenia, attempts to characterize this disorder have focused on disturbances in such integrative processing. Here, we review both micro- and macroscopic neuroscience research into Gamma synchrony, and its application to understanding schizophrenia. The review encompasses evidence from both animal and human studies for the functional significance of Gamma activity, the association between Gamma dysfunction and information processing disturbances, and the relevance of specific Gamma dysfunctions to the integration and extension of previous disconnection models of schizophrenia. Attention is given to the relationship between Gamma activity and the heterogeneous symptoms of schizophrenia. Existing studies show that measures of Gamma activity have the potential to explain far more of the variance in schizophrenia performance than previous neurophysiological measures. It is concluded that measures of Gamma synchrony offer a valuable window into the core integrative disturbance in schizophrenia cognition.

MEG study of long-term cortical reorganization of sensorimotor areas with respect to using chopsticks

The movements required to use chopsticks are overlearned and routine in Asians. Most non-Asians, on the other hand, typically have difficulty performing this unfamiliar manual activity, and have to focus their attention on the movements required to use chopsticks adequately. Using magnetoencephalography (MEG) we compared the cortical activation of highly trained Asian chopstick users to the activation of Europeans who only occasionally used chopsticks, while they performed the same tasks with chopsticks or a control task of simple tapping of the same fingers. The data were analyzed using the new method of synthetic aperture magnetometry (SAM). In Europeans there was a significantly higher ratio of spectral power in the higher gamma frequency band (60-80 Hz) over the sensorimotor area compared to the Asian subjects. From these results we conclude that the high gamma band activity in the sensorimotor area may reflect focused attention and functional reorganization of the cortical network with respect to sensorimotor experience.