Linking visual gamma to task‐related brain networks—a simultaneous EEG‐fMRI study

There is a growing interest in human gamma‐band oscillatory activity due to its direct link to neuronal populations, its associations with many cognitive processes, and its positive relationship with fMRI BOLD signal. Visual gamma has been successfully detected using concurrent EEG‐fMRI recordings and linked to activity in the visual cortex using voxel‐wise regression analysis. As gamma‐band oscillations reflect predominantly feedforward projections between brain regions, its inclusion in functional connectivity analysis is highly recommended; however, very few studies have investigated this line of research. In the current study, we aimed to explore this gap by asking which fMRI brain network is related to gamma activity induced by the color discrimination task. Advanced denoising strategies and multitaper spectral decomposition were applied to EEG data to detect gamma oscillations, and group independent component analysis was performed on fMRI data to identify task‐related neural networks. Despite using only trials without motor response (50% of the trials), the two neural measures were successfully coupled. One of the six task‐related networks, the occipito‐parietal network, exhibited significant trial‐by‐trial covariations with gamma oscillations. In addition to the expected extrastriate visual cortex, the network encompasses extensive brain activations in the precuneus, bilateral intraparietal, and anterior insular cortices. We argue that the visual cortex is the source of gamma, whereas the remaining brain regions exhibit feedforward and feedback connections related to this oscillatory activity. Our findings provide evidence for the electrophysiological basis of the connectivity revealed by BOLD signal and impart novel insights into the neural mechanism of color discrimination.     

Modafinil augments oscillatory power in middle frequencies during rule selection

Control‐related cognitive processes are associated with cortical oscillations and modulated by catecholamine neurotransmitters. It remains unclear how catecholamine systems modulate control‐related oscillations. We tested modafinil effects on rule‐related 4–30 Hz oscillations, with double‐blind, placebo‐controlled (within‐subjects) testing of 22 healthy adults, using EEG during cognitive control task performance. EEG data underwent time‐frequency decomposition with Morlet wavelets to determine power of 4–30 Hz oscillations. Modafinil enhanced oscillatory power associated with high‐control rule selection in theta, alpha, and beta ranges, with a frontotemporal topography and minimal effects during rule maintenance. Augmentation of catecholamine signaling enhances middle‐frequency cortical oscillatory power associated with rule selection, which may subserve diverse subcomponent processes in proactive cognitive control.     

A mechanism of deficient interregional neural communication in schizophrenia

Cognitive interference control is disrupted in schizophrenia (SZ). Neuroimaging studies relate interference control to 4–7 Hz (theta) neural activity in a network spanning prefrontal, anterior cingulate (ACC), and parietal cortices. The mechanism of communication in this network and how it is disrupted in schizophrenia are unclear. Behavioral performance and EEG theta oscillations were examined in a Stroop color‐word interference task in 17 healthy controls (HC) and 14 SZ patients. Color‐word incongruence induced less theta power increase in SZ than in HC around 400 ms and 600–900 ms after word onset in ACC, left middle frontal gyrus (MFG), and inferior parietal regions. Coupling of ACC theta phase to MFG gamma amplitude, indexing interregional communication, was weaker in SZ than in HC. Results suggest ACC‐MFG theta power modulation as a mechanism of interference control that supports executive function and is disrupted in schizophrenia.     

Neurofeedback training of gamma band oscillations improves perceptual processing

In this study, a noninvasive electroencephalography-based neurofeedback method is applied to train volunteers to deliberately increase gamma band oscillations (40 Hz) in the visual cortex. Gamma band oscillations in the visual cortex play a functional role in perceptual processing. In a previous study, we were able to demonstrate that gamma band oscillations prior to stimulus presentation have a significant influence on perceptual processing of visual stimuli. In the present study, we aimed to investigate longer lasting effects of gamma band neurofeedback training on perceptual processing. For this purpose, a feedback group was trained to modulate oscillations in the gamma band, while a control group participated in a task with an identical design setting but without gamma band feedback. Before and after training, both groups participated in a perceptual object detection task and a spatial attention task. Our results clearly revealed that only the feedback group but not the control group exhibited a visual processing advantage and an increase in oscillatory gamma band activity in the pre-stimulus period of the processing of the visual object stimuli after the neurofeedback training. Results of the spatial attention task showed no difference between the groups, which underlines the specific role of gamma band oscillations for perceptual processing. In summary, our results show that modulation of gamma band activity selectively affects perceptual processing and therefore supports the relevant role of gamma band activity for this specific process. Furthermore, our results demonstrate the eligibility of gamma band oscillations as a valuable tool for neurofeedback applications.     

Working memory related gamma oscillations in schizophrenia patients.

Recent reports show that the capability of neural networks supporting high-frequency synchronization is reduced in the brain of schizophrenia patients. Specifically, deficits in gamma activity have been shown in schizophrenia patients during perception and simple cognitive tasks. However little is known about alterations in gamma responses during complex and higher cognitive processing. The main objective of this study was to investigate modulation of event-related gamma responses in tasks varying working memory (WM) load in schizophrenia patients (N=10) and healthy controls (N=10). Gamma amplitude values were obtained for a simple choice reaction task, a low WM demand task, and a high WM demand task. During all three tasks schizophrenia patients showed significantly slower reaction times and higher error rates than controls. A gradual increase of gamma amplitudes after stimulus onset was associated with increase of WM load in controls. In contrast, high amplitude gamma oscillations remained constant regardless of task difficulty in patients. These results suggest that healthy subjects used various cognitive strategies depending on task difficulty, while schizophrenia patients needed to initiate complex cognitive processes similar to those used during processing of novel contexts or stimuli even for the simple choice reaction task with low cognitive demand.     

Alpha oscillations and the control of voluntary saccadic behavior

The purpose of this review is to explore the dynamic properties of alpha oscillations as biological covariates of intra- and inter-individual variance in saccadic behavior. A preponderance of research suggests that oscillatory dynamics in the alpha band co-vary with performance on a number of visuo-spatial cognitive tasks. Here we discuss a growing body of research relating these measures to saccadic behavior, focusing also on how task related and spontaneous measures of alpha oscillations may serve as potential biomarkers for ocular motor dysfunction in clinical populations.     

Fast gamma oscillations in areas MT and MST occur during visual stimulation,but not during visually guided manual tracking

We studied the incidence of oscillatory activity in the gamma range (35–110 Hz) in single cell and multi-unit activity recorded from extrastriate areas MT (middle temporal) and MST (superior middle temporal) while rhesus monkeys performed different behavioural tasks. During full field stimulation by coherent motion of random dots, we observed gamma oscillations in approximately 20% of the cells. The average oscillation frequencies differed considerably between both animals (60 Hz vs 100 Hz). In both animals, oscillatory modulation was particularly strong at sites that showed a strong directional bias to visual stimulation. The amount of oscillatory activity was roughly the same whether stimulus movement was presented during fixation or whether the animal had to perform pursuit movements across a stationary visual pattern. If cells were engaged in gamma oscillations during visual stimulation, the amount of oscillatory modulation was dependent on stimulus direction, stimulus velocity and stimulus contrast. During a visually guided manual tracking task no gamma activity was detectable. Cells with clear oscillatory modulation during the full field stimulation failed to show oscillatory activity when the animal was involved in a motor task in which the visual motion information had to be evaluated for the correct movement of the hand. Our results reaffirm the ubiquitous presence of stimulus-induced gamma oscillation in extrastriate areas MT and MST of the awake monkey during various stimulus conditions, but they fail to support the notion that high-frequency gamma oscillations in this area play a specific role during cortical control of a motor response to visual stimulation. Electronic Publication     

Comparison of the effects of acute and chronic administration of ketamine on hippocampal oscillations: relevance for the NMDA receptor hypofunction model of schizophrenia

The proper organization and function of GABAergic interneuron networks is essential for many cognitive processes and abnormalities in these systems have been documented in schizophrenic patients. The memory function of the hippocampus depends on two major patterns of oscillations in the theta and gamma ranges, both requiring the intact functioning of the network of fast-firing interneurons expressing parvalbumin. We examined the ability of acute and chronic administration of NMDA receptor (NMDA-R) antagonists to recapitulate the oscillatory dysfunctions observed in schizophrenia. In freely moving rats, acute injection of MK801 or ketamine increased gamma power in both CA1 and dentate gyrus of the hippocampus. Theta peak shifted to higher frequencies whereas the average 5–10 Hz theta power decreased by 24% in CA1 and remained high in the dentate gyrus. Strong increase in CA1 gamma and decrease in theta power triggered by brainstem stimulation were found under urethane anesthesia. In contrast to acute experiments, chronic administration of ketamine caused a steady decline in both gamma and theta oscillations, 2–4 weeks after treatment. A further important difference between the two models was that the effects of acute injection were more robust than the changes after chronic treatment. Chronic administration of ketamine also leads to decrease in the number of detectable parvalbumin interneurons. Histological examination of interindividual differences indicated, however, that within the ketamine treated group a further decrease in parvalbumin neurons correlated with strengthening of oscillations. The findings are consistent with abnormalities of oscillations in human schizophrenia and further validate the NMDA-R hypofunction hypothesis.     

Sensory encoding in Neuregulin 1 mutants

Schizophrenic patients show altered sensory perception as well as changes in electrical and magnetic brain responses to sustained, frequency-modulated sensory stimulation. Both the amplitude and temporal precision of the neural responses differ in patients as compared to control subjects, and these changes are most pronounced for stimulation at gamma frequencies (20–40 Hz). In addition, patients display enhanced spontaneous gamma oscillations, which has been interpreted as ‘neural noise’ that may interfere with normal stimulus processing. To investigate electrophysiological markers of aberrant sensory processing in a model of schizophrenia, we recorded neuronal activity in primary somatosensory cortex of mice heterozygous for the schizophrenia susceptibility gene Neuregulin 1. Sensory responses to sustained 20–70 Hz whisker stimulation were analyzed with respect to firing rates, spike precision (phase locking) and gamma oscillations, and compared to baseline conditions. The mutants displayed elevated spontaneous firing rates, a reduced gain in sensory-evoked spiking and gamma activity, and reduced spike precision of 20–40 Hz responses. These findings present the first in vivo evidence of the linkage between a genetic marker and altered stimulus encoding, thus suggesting a novel electrophysiological endophenotype of schizophrenia.     

40‐Hz oscillations underlying perceptual binding in young and older adults

Auditory object perception requires binding of elementary features of complex stimuli. Synchronization of high‐frequency oscillation in neural networks has been proposed as an effective alternative to binding via hard‐wired connections because binding in an oscillatory network can be dynamically adjusted to the ever‐changing sensory environment. Previously, we demonstrated in young adults that gamma oscillations are critical for sensory integration and found that they were affected by concurrent noise. Here, we aimed to support the hypothesis that stimulus evoked auditory 40‐Hz responses are a component of thalamocortical gamma oscillations and examined whether this oscillatory system may become less effective in aging. In young and older adults, we recorded neuromagnetic 40‐Hz oscillations, elicited by monaural amplitude‐modulated sound. Comparing responses in quiet and under contralateral masking with multitalker babble noise revealed two functionally distinct components of auditory 40‐Hz responses. The first component followed changes in the auditory input with high fidelity and was of similar amplitude in young and older adults. The second, significantly smaller in older adults, showed a 200‐ms interval of amplitude and phase rebound and was strongly attenuated by contralateral noise. The amplitude of the second component was correlated with behavioral speech‐in‐noise performance. Concurrent noise also reduced the P2 wave of auditory evoked responses at 200‐ms latency, but not the earlier N1 wave. P2 modulation was reduced in older adults. The results support the model of sensory binding through thalamocortical gamma oscillations. Limitation of neural resources for this process in older adults may contribute to their speech‐in‐noise understanding deficits.     

Two types of network oscillations and their odor responses in the primary olfactory center of a terrestrial mollusk

We identified two classes of network oscillations with different frequency ranges in the tentacle ganglion (TG), the primary olfactory center of the terrestrial mollusk Limax marginatus, and investigated the responses of these oscillations to odor inputs. A recent study indicated that there are serotonergic terminals in the TG. We found that when serotonin was applied to the TG, the spontaneous network oscillation of about 1.5 Hz in the TG changed its oscillatory frequency to 0.5 Hz. These two oscillations are distinct, because 1) in most cases, the application of serotonin to the TG initially inhibited the 1.5-Hz oscillation and subsequently generated the slow 0.5-Hz oscillation; and 2) occasionally, the application of serotonin did not inhibit the spontaneous 1.5-Hz oscillation, resulting in the coexistence of two network oscillations. Thus the TG has two different oscillatory dynamics. We named the spontaneous 1.5-Hz oscillation the fast oscillation (FO), and the serotonin-induced 0.5-Hz oscillation the slow oscillation (SO). By calculating the spatial coherence of the TG oscillations, we found that the FO is a noncoherent oscillatory mode and the SO is a coherent oscillatory mode. Finally, odor presentation to the olfactory receptors selectively modulated the SO by decreasing the oscillatory amplitude, but the FO was not modulated by the odor input. These results indicate that 1) the TG has two oscillatory states (FO and SO) and these states are changed by the extracellular level of serotonin, and 2) these two oscillatory states have different responses to odors.     

Simultaneous activation of gamma and theta network oscillations in rat hippocampal slice cultures

Hippocampal activity in vivo is characterized by concurrent oscillations at theta (4–15 Hz) and gamma (20–80 Hz) frequencies. Here we show that cholinergic receptor activation (methacholine 10–20 nm) in hippocampal slice cultures induces an oscillatory mode of activity, in which the intrinsic network oscillator (located in the CA3 area) expresses simultaneous theta and gamma network oscillations. Pyramidal cells display synaptic theta oscillations, characterized by cycles consisting of population EPSP-IPSP sequences that are dominated by population IPSPs. These rhythmic IPSPs most probably result from theta-modulated spiking activity of several interneurons. At the same time, the majority of interneurons consistently display synaptic gamma oscillations. These oscillatory cycles consist of fast depolarizing rhythmic events that are likely to reflect excitatory input from CA3 pyramidal cells. Interneurons comprising this functional group were identified morphologically. They include four known types of interneurons (basket, O-LM, bistratified and str. lucidum-specific cells) and one new type of CA3 interneuron (multi-subfield cell). The oscillatory activity of these interneurons is only weakly correlated between neighbouring cells, and in about half of these (44 %) is modulated by depolarizing theta rhythmicity. The overall characteristics of acetylcholine-induced oscillations in slice cultures closely resemble the rhythmicity observed in hippocampal field and single cell recordings in vivo . Both rhythmicities depend on intrinsic synaptic interactions, and are expressed by different cell types. The fact that these oscillations persist in a network lacking extra-hippocampal connections emphasizes the importance of intrinsic mechanisms in determining this form of hippocampal activity.     

Odorant modulation of neuronal activity and local field potential in sensory-deprived olfactory bulb

Neuronal discharge and local field potential (LFP) oscillations in the olfactory bulb (OB) are modulated by odorant stimulation. The LFP oscillations have been proposed as the mechanism that facilitates synchronization of OB output neurons and the representation of similar odorants. Gamma LFP oscillations depend on the OB inhibitory network and early sensory deprivation modifies this inhibitory network. However, little is known about the LFP oscillations and neuronal discharge in the deprived OB. We examined the mitral/tufted (MT) cells' oscillatory activity and LFP oscillations in both sensory-deprived and normal OBs in urethane anesthetized rats. We found that MT cells in deprived and normal OBs have similar basal mean firing rate; 44% of the recorded cells in deprived OB and only 8% of the cells in normal OB showed firing rate modulation by odorants, both exhibiting a similar ratio of excitatory to inhibitory responses. A fraction of MT cells exhibited oscillatory discharge centered on gamma (60–70 Hz) and beta (20 Hz) frequencies, although this feature was not consistently dependent on odorant stimulation. Odorants decreased the LFP oscillatory power in the gamma band (35–90 Hz) and increased the power in the beta band (12–30 Hz). The modulation of LFP oscillations by odorants was also predominant in the deprived (53%) compared to the normal OB (17%). In contrast, a higher fraction of MT cells' discharge was locked to the gamma LFP cycle in the normal OB. These results suggest that early unilateral olfactory deprivation increases the OB sensitivity to odorants and reduce the temporal synchrony between unitary activity and gamma LFP oscillations without altering the basal neuronal discharge.     

Transient alterations in slow oscillations of hippocampal networks by low-frequency stimulations on multi-electrode arrays

Slow oscillations in the hippocampus are correlated with memory consolidation and brain diseases. The characteristic firings of the hippocampal network in vitro are still poorly understood. Here, spontaneous oscillations (~0.004 Hz) were found in high-density hippocampal networks by multi-electrode arrays after 30 days in vitro. This kind of spontaneous activity was characterized by periodic synchronized superbursts, which persisted for approximately 60 s at long intervals. Additionally, 1-Hz stimulation (duration <120 s) could regulate these network-wide oscillatory activities by triggering the next synchronized superbursts prematurely. The results demonstrated that the slow oscillatory activities in hippocampal cultures could be regulated by external stimulation, which indicates that multi-electrode arrays provide a well-suited platform for studying the dynamics of slow oscillations in vitro and may help to elucidate the mechanism of electrical stimulation therapy.     

Oscillatory gamma activity in humans: a possible role for object representation.

The coherent representation of an object has been suggested to be established by the synchronization in the gamma range (20-100 Hz) of a distributed neural network. So-called '40-Hz' activity in humans could reflect such a mechanism. We have presented here experimental evidence supporting this hypothesis, both in the visual and auditory modalities. However, different types of gamma activity should be distinguished, mainly the evoked 40-Hz response and the induced gamma activities. Only induced gamma activities seem to be related to coherent object representations. In addition, their topography depends on sensory modality and task, which is in line with the idea that they reflect the oscillatory synchronization of task-dependent networks. They can also be functionally and topographically distinguished from the classical evoked potentials and from the alpha rhythm. It was also proposed that the functional role of gamma oscillations is not restricted to object representation established through bottom-up mechanisms of feature binding, but also extends to the cases of internally driven representations and to the maintenance of information in memory.     

Verbal working memory‐related neural network communication in schizophrenia

Impaired working memory (WM) in schizophrenia is associated with reduced hemodynamic and electromagnetic activity and altered network connectivity within and between memory‐associated neural networks. The present study sought to determine whether schizophrenia involves disruption of a frontal‐parietal network normally supporting WM and/or involvement of another brain network. Nineteen schizophrenia patients (SZ) and 19 healthy comparison subjects (HC) participated in a cued visual‐verbal Sternberg task while dense‐array EEG was recorded. A pair of item arrays each consisting of 2–4 consonants was presented bilaterally for 200 ms with a prior cue signaling the hemifield of the task‐relevant WM set. A central probe letter 2,000 ms later prompted a choice reaction time decision about match/mismatch with the target WM set. Group and WM load effects on time domain and time‐frequency domain 11–15 Hz alpha power were assessed for the cue‐to‐probe time window, and posterior 11–15 Hz alpha power and frontal 4–8 Hz theta power were assessed during the retention period. Directional connectivity was estimated via Granger causality, evaluating group differences in communication. SZ showed slower responding, lower accuracy, smaller overall time‐domain alpha power increase, and less load‐dependent alpha power increase. Midline frontal theta power increases did not vary by group or load. Network communication in SZ was characterized by temporal‐to‐posterior information flow, in contrast to bidirectional temporal‐posterior communication in HC. Results indicate aberrant WM network activity supporting WM in SZ that might facilitate normal load‐dependent and only marginally less accurate task performance, despite generally slower responding.     

Carbachol induces fast oscillations in the medial but not in the lateral entorhinal cortex of the isolated guinea pig brain

Fast oscillations at 25-80 Hz (gamma activity) have been proposed to play a role in attention-related mechanisms and synaptic plasticity in cortical structures. Recently, it has been demonstrated that the preservation of the entorhinal cortex is necessary to maintain gamma oscillations in the hippocampus. Because gamma activity can be reproduced in vitro by cholinergic activation, this study examined the characteristics of gamma oscillations induced by arterial perfusion or local intracortical injections of carbachol in the entorhinal cortex of the in vitro isolated guinea pig brain preparation. Shortly after carbachol administration, fast oscillatory activity at 25.2-28.2 Hz was observed in the medial but not in the lateral entorhinal cortex. Such activity was transiently associated with oscillations in the theta range that showed a variable pattern of distribution in the entorhinal cortex. No oscillatory activity was observed when carbachol was injected in the lateral entorhinal cortex. Gamma activity in the medial entorhinal cortex showed a phase reversal at 200-400 microm, had maximal amplitude at 400-500 microm depth, and was abolished by arterial perfusion of atropine (5 microM). Local carbachol application in the medial entorhinal cortex induced gamma oscillations in the hippocampus, whereas no oscillations were observed in the amygdala and in the piriform, periamygdaloid, and perirhinal cortices ipsilateral and contralateral to the carbachol injection. Hippocampal oscillations had higher frequency than the gamma activity recorded in the entorhinal cortex, suggesting the presence of independent generators in the two structures. The selective ability of the medial but not the lateral entorhinal cortex to generate gamma activity in response to cholinergic activation suggests a differential mode of signal processing in entorhinal cortex subregions.     

Abnormal-induced theta activity supports early directed-attention network deficits in progressive MCI

The electroencephalography (EEG) theta frequency band reacts to memory and selective attention paradigms. Global theta oscillatory activity includes a posterior phase-locked component related to stimulus processing and a frontal-induced component modulated by directed attention. To investigate the presence of early deficits in the directed attention-related network in elderly individuals with mild cognitive impairment (MCI), time–frequency analysis at baseline was used to assess global and induced theta oscillatory activity (4–6 Hz) during n-back working memory tasks in 29 individuals with MCI and 24 elderly controls (EC). At 1-year follow-up, 13 MCI patients were still stable and 16 had progressed. Baseline task performance was similar in stable and progressive MCI cases. Induced theta activity at baseline was significantly reduced in progressive MCI as compared to EC and stable MCI in all n-back tasks, which were similar in terms of directed attention requirements. While performance is maintained, the decrease of induced theta activity suggests early deficits in the directed-attention network in progressive MCI, whereas this network is functionally preserved in stable MCI.     

Effects of neurocognitive enhancement therapy in schizophrenia: Normalisation of memory performance

Introduction. A preponderance of research indicates that cognitive function in schizophrenia can be improved through cognitive remediation. However, few studies have attempted to characterise the extent of improvement relative to nonpsychiatric controls.

Method. Cognitive performance on reaction time, digit recall, and word recall of 58 schizophrenia patients at baseline and after 6 months of cognitive remediation was compared to the performance on these tasks of 39 community controls. Schizophrenia patients participated in Neurocognitive Enhancement Therapy (NET) and received hierarchical training on the memory tasks, but not on the reaction time task, which was only administered at intake and follow‐up.

Results. The schizophrenia sample showed significantly poorer performances than the community control sample on all three tasks at baseline. NET led to significant improvements in performance on trained memory tasks, but not the untrained reaction time task. There was a significant increase in the proportion of schizophrenia patients who achieved normal range performance on the memory tasks.

Conclusions. 52% of schizophrenia patients who were impaired on at least one of the memory tasks normalised their performance on at least one of those tasks as a result of cognitive training. Results suggest that clinically meaningful improvement may be possible using cognitive remediation.     

Feature integration in visual working memory: parietal gamma activity is related to cognitive coordination

The mechanism by which distinct subprocesses in the brain are coordinated is a central conundrum of systems neuroscience. The parietal lobe is thought to play a key role in visual feature integration, and oscillatory activity in the gamma frequency range has been associated with perception of coherent objects and other tasks requiring neural coordination. Here, we examined the neural correlates of integrating mental representations in working memory and hypothesized that parietal gamma activity would be related to the success of cognitive coordination. Working memory is a classic example of a cognitive operation that requires the coordinated processing of different types of information and the contribution of multiple cognitive domains. Using magnetoencephalography (MEG), we report parietal activity in the high gamma (80-100 Hz) range during manipulation of visual and spatial information (colors and angles) in working memory. This parietal gamma activity was significantly higher during manipulation of visual-spatial conjunctions compared with single features. Furthermore, gamma activity correlated with successful performance during the conjunction task but not during the component tasks. Cortical gamma activity in parietal cortex may therefore play a role in cognitive coordination.