Auditory prediction errors in sound frequency and duration generated different cortical activation patterns in the human brain: an ECoG study.

Sound frequency and duration are essential auditory components. The brain perceives deviations from the preceding sound context as prediction errors, allowing efficient reactions to the environment. Additionally, prediction error response to duration change is reduced in the initial stages of psychotic disorders. To compare the spatiotemporal profiles of responses to prediction errors, we conducted a human electrocorticography study with special attention to high gamma power in 13 participants who completed both frequency and duration oddball tasks. Remarkable activation in the bilateral superior temporal gyri in both the frequency and duration oddball tasks were observed, suggesting their association with prediction errors. However, the response to deviant stimuli in duration oddball task exhibited a second peak, which resulted in a bimodal response. Furthermore, deviant stimuli in frequency oddball task elicited a significant response in the inferior frontal gyrus that was not observed in duration oddball task. These spatiotemporal differences within the Parasylvian cortical network could account for our efficient reactions to changes in sound properties. The findings of this study may contribute to unveiling auditory processing and elucidating the pathophysiology of psychiatric disorders.

Utilizing Excitatory and Inhibitory Activity Derived from Interictal Intracranial Electroencephalography as Potential Biomarkers for Epileptogenicity.

Epileptogenic zones (EZs), where epileptic seizures cease after resection, are localized by assessing the seizure-onset zone using ictal electroencephalography (EEG). Owing to the difficulty in capturing unpredictable seizures, biomarkers capable of identifying EZs from interictal EEG are anticipated. Recent studies using intracranial EEG have identified several potential candidate biomarkers for epileptogenicity. High-frequency oscillation (HFO) was initially expected to be a robust biomarker of abnormal excitatory activity in the ictogenic region. However, HFO-guided resection failed to improve seizure prognosis. Meanwhile, the regularity of low-gamma oscillations (30-80 Hz) indicates inhibitory interneurons' hypersynchronization, which could be used to localize the EZ. Besides resting-state EEG assessments, evoked potentials elicited by single-pulse electrical stimulation, such as corticocortical evoked potentials (CCEP), became valuable tools for assessing epileptogenic regions. CCEP responses recorded in the cortex remote from the stimulation site indicate functional connectivity, revealing increased internal connectivity within the ictogenic region and elevated inhibitory input from the non-involved regions to the ictogenic region. Conversely, large responses close to the stimulation site reflect local excitability, manifesting as an increased N1 amplitude and overriding HFO. Further research is required to establish whether these novel electrophysiological methods, either individually or in combination, can function as robust biomarkers of epileptogenicity and hold promise for improving seizure prognosis.

Auditory prediction and prediction error responses evoked through a novel cascade roving paradigm: a human ECoG study.

Auditory sensory processing is assumed to occur in a hierarchical structure including the primary auditory cortex (A1), superior temporal gyrus, and frontal areas. These areas are postulated to generate predictions for incoming stimuli, creating an internal model of the surrounding environment. Previous studies on mismatch negativity have indicated the involvement of the superior temporal gyrus in this processing, whereas reports have been mixed regarding the contribution of the frontal cortex. We designed a novel auditory paradigm, the "cascade roving" paradigm, which incorporated complex structures (cascade sequences) into a roving paradigm. We analyzed electrocorticography data from six patients with refractory epilepsy who passively listened to this novel auditory paradigm and detected responses to deviants mainly in the superior temporal gyrus and inferior frontal gyrus. Notably, the inferior frontal gyrus exhibited broader distribution and sustained duration of deviant-elicited responses, seemingly differing in spatio-temporal characteristics from the prediction error responses observed in the superior temporal gyrus, compared with conventional oddball paradigms performed on the same participants. Moreover, we observed that the deviant responses were enhanced through stimulus repetition in the high-gamma range mainly in the superior temporal gyrus. These features of the novel paradigm may aid in our understanding of auditory predictive coding.

Paving the Way for Memory Enhancement: Development and Examination of a Neurofeedback System Targeting the Medial Temporal Lobe.

Neurofeedback (NF) shows promise in enhancing memory, but its application to the medial temporal lobe (MTL) still needs to be studied. Therefore, we aimed to develop an NF system for the memory function of the MTL and examine neural activity changes and memory task score changes through NF training. We created a memory NF system using intracranial electrodes to acquire and visualise the neural activity of the MTL during memory encoding. Twenty trials of a tug-of-war game per session were employed for NF and designed to control neural activity bidirectionally (Up/Down condition). NF training was conducted with three patients with drug-resistant epilepsy, and we observed an increasing difference in NF signal between conditions (Up-Down) as NF training progressed. Similarities and negative correlation tendencies between the transition of neural activity and the transition of memory function were also observed. Our findings demonstrate NF's potential to modulate MTL activity and memory encoding. Future research needs further improvements to the NF system to validate its effects on memory functions. Nonetheless, this study represents a crucial step in understanding NF's application to memory and provides valuable insights into developing more efficient memory enhancement strategies.

Intracranial Neurofeedback Modulating Neural Activity in the Mesial Temporal Lobe During Memory Encoding: A Pilot Study.

Removal of the mesial temporal lobe (MTL) is an established surgical procedure that leads to seizure freedom in patients with intractable MTL epilepsy; however, it carries the potential risk of memory damage. Neurofeedback (NF), which regulates brain function by converting brain activity into perceptible information and providing feedback, has attracted considerable attention in recent years for its potential as a novel complementary treatment for many neurological disorders. However, no research has attempted to artificially reorganize memory functions by applying NF before resective surgery to preserve memory functions. Thus, this study aimed (1) to construct a memory NF system that used intracranial electrodes to feedback neural activity on the language-dominant side of the MTL during memory encoding and (2) to verify whether neural activity and memory function in the MTL change with NF training. Two intractable epilepsy patients with implanted intracranial electrodes underwent at least five sessions of memory NF training to increase the theta power in the MTL. There was an increase in theta power and a decrease in fast beta and gamma powers in one of the patients in the late stage of memory NF sessions. NF signals were not correlated with memory function. Despite its limitations as a pilot study, to our best knowledge, this study is the first to report that intracranial NF may modulate neural activity in the MTL, which is involved in memory encoding. The findings provide important insights into the future development of NF systems for the artificial reorganization of memory functions.

Efficient, continual, and generalized learning in the brain - neural mechanism of Mental Schema 2.0.

There has been tremendous progress in artificial neural networks (ANNs) over the past decade; however, the gap between ANNs and the biological brain as a learning device remains large. With the goal of closing this gap, this paper reviews learning mechanisms in the brain by focusing on three important issues in ANN research: efficiency, continuity, and generalization. We first discuss the method by which the brain utilizes a variety of self-organizing mechanisms to maximize learning efficiency, with a focus on the role of spontaneous activity of the brain in shaping synaptic connections to facilitate spatiotemporal learning and numerical processing. Then, we examined the neuronal mechanisms that enable lifelong continual learning, with a focus on memory replay during sleep and its implementation in brain-inspired ANNs. Finally, we explored the method by which the brain generalizes learned knowledge in new situations, particularly from the mathematical generalization perspective of topology. Besides a systematic comparison in learning mechanisms between the brain and ANNs, we propose "Mental Schema 2.0," a new computational property underlying the brain's unique learning ability that can be implemented in ANNs.

Spatiotemporal target selection for intracranial neural decoding of abstract and concrete semantics.

Decoding the inner representation of a word meaning from human cortical activity is a substantial challenge in the development of speech brain-machine interfaces (BMIs). The semantic aspect of speech is a novel target of speech decoding that may enable versatile communication platforms for individuals with impaired speech ability; however, there is a paucity of electrocorticography studies in this field. We decoded the semantic representation of a word from single-trial cortical activity during an imageability-based property identification task that required participants to discriminate between the abstract and concrete words. Using high gamma activity in the language-dominant hemisphere, a support vector machine classifier could discriminate the 2-word categories with significantly high accuracy (73.1 ± 7.5%). Activities in specific time components from two brain regions were identified as significant predictors of abstract and concrete dichotomy. Classification using these feature components revealed that comparable prediction accuracy could be obtained based on a spatiotemporally targeted decoding approach. Our study demonstrated that mental representations of abstract and concrete word processing could be decoded from cortical high gamma activities, and the coverage of implanted electrodes and time window of analysis could be successfully minimized. Our findings lay the foundation for the future development of semantic-based speech BMIs.

Vagus Nerve Stimulation Amplifies Task-Induced Cerebral Blood Flow Increase.

BACKGROUND: Vagus nerve stimulation (VNS) is an established palliative surgical treatment for refractory epilepsy. Recently, pairing VNS with rehabilitation received growing attention for their joint effect on neural plasticity. However, objective biological measurements proving the interaction between VNS effects and cortical recruitment are lacking. Studies reported that VNS induced little blood flow increase in the cerebral cortex. OBJECTIVE: This study tested the hypothesis that pairing VNS with a cognitive task amplifies task-induced cerebral blood flow (CBF). METHODS: This study included 21 patients implanted with vagus nerve stimulator to treat refractory epilepsy. Near-infrared spectroscopy (NIRS) with sensors on the forehead measured CBF changes in the frontal cortices in response to VNS. Cerebral blood flow was measured when VNS was delivered during a resting state or a verbal fluency task. We analyzed the VNS effect on CBF in relation to stimulation intensity and clinical responsiveness. RESULTS: We observed no CBF change when VNS was delivered during rest, irrespective of stimulation intensity or responsiveness. Cerebral blood flow changed significantly when a verbal fluency task was paired with VNS in a stimulation intensity-dependent manner. Cerebral blood flow changes in the non-responders showed no intensity-dependency. CONCLUSION: Our results could be an important biological proof of the interaction between VNS effects and cortical recruitment, supporting the validity of pairing VNS with rehabilitation.

Global and Parallel Cortical Processing Based on Auditory Gamma Oscillatory Responses in Humans.

Gamma oscillations are physiological phenomena that reflect perception and cognition, and involve parvalbumin-positive γ-aminobutyric acid-ergic interneuron function. The auditory steady-state response (ASSR) is the most robust index for gamma oscillations, and it is impaired in patients with neuropsychiatric disorders such as schizophrenia and autism. Although ASSR reduction is known to vary in terms of frequency and time, the neural mechanisms are poorly understood. We obtained high-density electrocorticography recordings from a wide area of the cortex in 8 patients with refractory epilepsy. In an ASSR paradigm, click sounds were presented at frequencies of 20, 30, 40, 60, 80, 120, and 160 Hz. We performed time-frequency analyses and analyzed intertrial coherence, event-related spectral perturbation, and high-gamma oscillations. We demonstrate that the ASSR is globally distributed among the temporal, parietal, and frontal cortices. The ASSR was composed of time-dependent neural subcircuits differing in frequency tuning. Importantly, the frequency tuning characteristics of the late-latency ASSR varied between the temporal/frontal and parietal cortex, suggestive of differentiation along parallel auditory pathways. This large-scale survey of the cortical ASSR could serve as a foundation for future studies of the ASSR in patients with neuropsychiatric disorders.

Musical Auditory Alpha Wave Neurofeedback: Validation and Cognitive Perspectives.

Neurofeedback through visual, auditory, or tactile sensations improves cognitive functions and alters the activities of daily living. However, some people, such as children and the elderly, have difficulty concentrating on neurofeedback for a long time. Constant stressless neurofeedback for a long time may be achieved with auditory neurofeedback using music. The primary purpose of this study was to clarify whether music-based auditory neurofeedback increases the power of the alpha wave in healthy subjects. During neurofeedback, white noise was superimposed on classical music, with the noise level inversely correlating with normalized alpha wave power. This was a single-blind, randomized control crossover trial in which 10 healthy subjects underwent, in an assigned order, normal and random feedback (NF and RF), either of which was at least 4 weeks long. Cognitive functions were evaluated before, between, and after each neurofeedback period. The secondary purpose was to assess neurofeedback-induced changes in cognitive functions. A crossover analysis showed that normalized alpha-power was significantly higher in NF than in RF; therefore, music-based auditory neurofeedback facilitated alpha wave induction. A composite category-based analysis of cognitive functions revealed greater improvements in short-term memory in subjects whose alpha-power increased in response to NF. The present study employed a long period of auditory alpha neurofeedback and achieved successful alpha wave induction and subsequent improvements in cognitive functions. Although this was a pilot study that validated a music-based alpha neurofeedback system for healthy subjects, the results obtained are encouraging for those with difficulty in concentrating on conventional alpha neurofeedback.Trial registration: 2018077NI, date of registration: 2018/11/27.

Convolutional neural network with autoencoder-assisted multiclass labelling for seizure detection based on scalp electroencephalography.

OBJECTIVE: In long-term video-monitoring, automatic seizure detection holds great promise as a means to reduce the workload of the epileptologist. A convolutional neural network (CNN) designed to process images of EEG plots demonstrated high performance for seizure detection, but still has room for reducing the false-positive alarm rate. METHODS: We combined a CNN that processed images of EEG plots with patient-specific autoencoders (AE) of EEG signals to reduce the false alarms during seizure detection. The AE automatically logged abnormalities, i.e., both seizures and artifacts. Based on seizure logs compiled by expert epileptologists and errors made by AE, we constructed a CNN with 3 output classes: seizure, non-seizure-but-abnormal, and non-seizure. The accumulative measure of number of consecutive seizure labels was used to issue a seizure alarm. RESULTS: The second-by-second classification performance of AE-CNN was comparable to that of the original CNN. False-positive seizure labels in AE-CNN were more likely interleaved with "non-seizure-but-abnormal" labels than with true-positive seizure labels. Consequently, "non-seizure-but-abnormal" labels interrupted runs of false-positive seizure labels before triggering an alarm. The median false alarm rate with the AE-CNN was reduced to 0.034 h-1, which was one-fifth of that of the original CNN (0.17 h-1). CONCLUSIONS: A label of "non-seizure-but-abnormal" offers practical benefits for seizure detection. The modification of a CNN with an AE is worth considering because AEs can automatically assign "non-seizure-but-abnormal" labels in an unsupervised manner with no additional demands on the time of the epileptologist.

Spatiotemporal Differentiation of MMN From N1 Adaptation: A Human ECoG Study.

Auditory mismatch negativity (MMN) is an electrophysiological response to a deviation from regularity. This response is considered pivotal to understanding auditory processing, particularly in the pre-attentive phase. However, previous findings suggest that MMN is a product of N1 adaptation/enhancement, which reflects lower-order auditory processing. The separability of these two components remains unclear and is considered an important issue in the field of neuroscience. The aim of the present study was to spatiotemporally differentiate MMN from N1 adaptation using human electrocorticography (ECoG). Auditory evoked potentials under the classical oddball (OD) task as well as the many standards (MS) task were recorded in three patients with epilepsy whose lateral cortices were widely covered with high-density electrodes. Close observation identified an electrode at which N1 adaptation was temporally separated from MMN, whereas N1 adaptation was partially incorporated into MMN at other electrodes. Since N1 adaptation occurs in the N1 population, we spatially compared MMN with N1 obtained from the MS task instead of N1 adaptation. As a result, N1 was observed in a limited area around the Sylvian fissure adjacent to A1, whereas MMN was noted in wider areas, including the temporal, frontal, and parietal lobes. MMN was thus considered to be differentiated from N1 adaptation. The results suggest that MMN is not merely a product of the neural adaptation of N1 and instead represents higher-order processes in auditory deviance detection. These results will contribute to strengthening the foundation of future research in this field.

Visualization of AMPA receptors in living human brain with positron emission tomography.

Although aberrations in the number and function of glutamate AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid) receptors are thought to underlie neuropsychiatric disorders, no methods are currently available for visualizing AMPA receptors in the living human brain. Here we developed a positron emission tomography (PET) tracer for AMPA receptors. A derivative of 4-[2-(phenylsulfonylamino)ethylthio]-2,6-difluoro-phenoxyacetamide radiolabeled with 11C ([11C]K-2) showed specific binding to AMPA receptors. Our clinical trial with healthy human participants confirmed reversible binding of [11C]K-2 in the brain according to Logan graphical analysis (UMIN000020975; study design: non-randomized, single arm; primary outcome: dynamics and distribution volumes of [11C]K-2 in the brain; secondary outcome: adverse events of [11C]K-2 during the 4-10 d following dosing; this trial met prespecified endpoints). In an exploratory clinical study including patients with epilepsy, we detected increased [11C]K-2 uptake in the epileptogenic focus of patients with mesial temporal lobe epilepsy, which was closely correlated with the local AMPA receptor protein distribution in surgical specimens from the same individuals (UMIN000025090; study design: non-randomized, single arm; primary outcome: correlation between [11C]K-2 uptake measured with PET before surgery and AMPA receptor protein density examined by biochemical study after surgery; secondary outcome: adverse events during the 7 d following PET scan; this trial met prespecified endpoints). Thus, [11C]K-2 is a potent PET tracer for AMPA receptors, potentially providing a tool to examine the involvement of AMPA receptors in neuropsychiatric disorders.

Gamma-Band Auditory Steady-State Response as a Neurophysiological Marker for Excitation and Inhibition Balance: A Review for Understanding Schizophrenia and Other Neuropsychiatric Disorders.

Altered gamma oscillations have attracted considerable attention as an index of the excitation/inhibition (E/I) imbalance in schizophrenia and other neuropsychiatric disorders. The auditory steady-state response (ASSR) has been the most robust probe of abnormal gamma oscillatory dynamics in schizophrenia. Here, we review recent ASSR studies in patients with schizophrenia and other neuropsychiatric disorders. Preclinical ASSR research, which has contributed to the elucidation of the underlying pathophysiology of these diseases, is also discussed. The developmental trajectory of the ASSR has been explored and may show signs of the maturation and disruption of E/I balance in adolescence. Animal model studies have shown that synaptic interactions between parvalbumin-positive GABAergic interneurons and pyramidal neurons contribute to the regulation of E/I balance, which is related to the generation of gamma oscillation. Therefore, ASSR alteration may be a significant electrophysiological finding related to the E/I imbalance in neuropsychiatric disorders, which is a cross-disease feature and may reflect clinical staging. Future studies regarding ASSR generation, especially in nonhuman primate models, will advance our understanding of the brain circuit and the molecular mechanisms underlying neuropsychiatric disorders.

Autoencoding of long-term scalp electroencephalogram to detect epileptic seizure for diagnosis support system.

INTRODUCTION: Epileptologists could benefit from a diagnosis support system that automatically detects seizures because visual inspection of long-term electroencephalograms (EEGs) is extremely time-consuming. However, the diversity of seizures among patients makes it difficult to develop universal features that are applicable for automatic seizure detection in all cases, and the rarity of seizures results in a lack of sufficient training data for classifiers. METHODS: To overcome these problems, we utilized an autoencoder (AE), which is often used for anomaly detection in the field of machine learning, to perform seizure detection. We hypothesized that multichannel EEG signals are compressible by AE owing to their spatio-temporal coupling and that the AE should be able to detect seizures as anomalous events from an interictal EEG. RESULTS: Through experiments, we found that the AE error was able to classify seizure and nonseizure states with a sensitivity of 100% in 22 out of 24 available test subjects and that the AE was better than the commercially available software BESA and Persyst for half of the test subjects. CONCLUSIONS: These results suggest that the AE error is a feasible candidate for a universal seizure detection feature.

Mismatch negativity (MMN) as a tool for translational investigations into early psychosis: A review.

Mismatch negativity (MMN) reduction is one of the most robust findings among several neurophysiological and neurocognitive measures in patients with schizophrenia. MMN is a promising biomarker for schizophrenia because of the following properties: 1) its relationship with early psychosis, including clinical high-risk (CHR); 2) its relationship with the functional abilities of patients; and 3) its translatability into basic research using animal models. Specifically, the utility of the passive auditory oddball paradigm that does not require subjects to make behavioral responses enables identical physiological activities to be obtained from both experimental animals and patients. This advantage has contributed to clarifying the generating mechanism of MMN in various animal studies. We reviewed clinical reports focused on early psychosis; specifically differential effects of deviance type and relationships to clinical and functional outcome. For the utility of MMN as a tool for translational research, we next reviewed recent MMN studies in rodents and nonhuman primates (NHP) as well as studies using intracranial recordings in humans, a rare opportunity to detect neural signals in vivo in humans. Neural computations of MMN, such as adaptation, deviance detection, and predictive coding, have been recent topics for understanding MMN generating mechanisms. Finally, several significant research questions were provided for future directions. MMN research could contribute to innovative, novel, therapeutic strategies in the future by becoming a bridge between basic and clinical research.

Seizure detection by convolutional neural network-based analysis of scalp electroencephalography plot images.

We hypothesized that expert epileptologists can detect seizures directly by visually analyzing EEG plot images, unlike automated methods that analyze spectro-temporal features or complex, non-stationary features of EEG signals. If so, seizure detection could benefit from convolutional neural networks because their visual recognition ability is comparable to that of humans. We explored image-based seizure detection by applying convolutional neural networks to long-term EEG that included epileptic seizures. After filtering, EEG data were divided into short segments based on a given time window and converted into plot EEG images, each of which was classified by convolutional neural networks as 'seizure' or 'non-seizure'. These resultant labels were then used to design a clinically practical index for seizure detection. The best true positive rate was obtained using a 1-s time window. The median true positive rate of convolutional neural networks labelling by seconds was 74%, which was higher than that of commercially available seizure detection software (20% by BESA and 31% by Persyst). For practical use, the median of detected seizure rate by minutes was 100% by convolutional neural networks, which was higher than the 73.3% by BESA and 81.7% by Persyst. The false alarm of convolutional neural networks' seizure detection was issued at 0.2 per hour, which appears acceptable for clinical practice. Moreover, we demonstrated that seizure detection improved when training was performed using EEG patterns similar to those of testing data, suggesting that adding a variety of seizure patterns to the training dataset will improve our method. Thus, artificial visual recognition by convolutional neural networks allows for seizure detection, which otherwise currently relies on skillful visual inspection by expert epileptologists during clinical diagnosis.

Deviance detection is the dominant component of auditory contextual processing in the lateral superior temporal gyrus: A human ECoG study.

Auditory contextual processing has been assumed to be based on a hierarchical structure consisting of the primary auditory cortex, superior temporal gyrus (STG), and frontal lobe. Recent invasive studies on mismatch negativity (MMN) have revealed functional segregation for auditory contextual processing such as neural adaptation in the primary auditory cortex and prediction in the frontal lobe. However, the role of the STG remains unclear. We obtained induced activity in the high gamma band as mismatch response (MMR), an electrocorticographic (ECoG) counterpart to scalp MMN, and the components of MMR by analyzing ECoG data from patients with refractory epilepsy in an auditory oddball task paradigm. We found that MMR localized mainly in the bilateral posterior STGs, and that deviance detection largely accounted for MMR. Furthermore, adaptation was identified in a limited number of electrodes on the superior temporal plane. Our findings reveal a mixed contribution of deviance detection and adaptation depending on location in the STG. Such spatial considerations could lead to further understanding of the pathophysiology of relevant psychiatric disorders.