Dynamic Cell Assemblies Research Articles

Dynamic assemblies of parvalbumin interneurons in brain oscillations

Brain oscillations are crucial for perception, memory, and behavior. Parvalbumin-expressing (PV) interneurons are critical for these oscillations, but their population dynamics remain unclear. Using voltage imaging, we simultaneously recorded membrane potentials in up to 26 PV interneurons invivo during hippocampal ripple oscillations in mice. We found that PV cells generate ripple-frequency rhythms by forming highly dynamic cell assemblies. These assemblies exhibit rapid and significant changes from cycle to cycle, varying greatly in both size and membership. Importantly, this variability is not just random spiking failures of individual neurons. Rather, the activities of other PV cells contain significant information about whether a PV cell spikes or not in a given cycle. This coordination persists without network oscillations, and it exists in subthreshold potentials even when the cells are not spiking. Dynamic assemblies of interneurons may provide a new mechanism to modulate postsynaptic dynamics and impact cognitive functions flexibly and rapidly.

Dynamic core-periphery structure of information sharing networks in entorhinal cortex and hippocampus.

Neural computation is associated with the emergence, reconfiguration, and dissolution of cell assemblies in the context of varying oscillatory states. Here, we describe the complex spatiotemporal dynamics of cell assemblies through temporal network formalism. We use a sliding window approach to extract sequences of networks of information sharing among single units in hippocampus and entorhinal cortex during anesthesia and study how global and node-wise functional connectivity properties evolve through time and as a function of changing global brain state (theta vs. slow-wave oscillations). First, we find that information sharing networks display, at any time, a core-periphery structure in which an integrated core of more tightly functionally interconnected units links to more loosely connected network leaves. However the units participating to the core or to the periphery substantially change across time windows, with units entering and leaving the core in a smooth way. Second, we find that discrete network states can be defined on top of this continuously ongoing liquid core-periphery reorganization. Switching between network states results in a more abrupt modification of the units belonging to the core and is only loosely linked to transitions between global oscillatory states. Third, we characterize different styles of temporal connectivity that cells can exhibit within each state of the sharing network. While inhibitory cells tend to be central, we show that, otherwise, anatomical localization only poorly influences the patterns of temporal connectivity of the different cells. Furthermore, cells can change temporal connectivity style when the network changes state. Altogether, these findings reveal that the sharing of information mediated by the intrinsic dynamics of hippocampal and entorhinal cortex cell assemblies have a rich spatiotemporal structure, which could not have been identified by more conventional time- or state-averaged analyses of functional connectivity.

A Basis Set of Elementary Operations Captures Recombination of Neocortical Cell Assemblies During Basal Conditions and Learning

Cell assemblies — subgroups within neuronal networks — are believed to serve as functional entities underlying cognitive capabilities such as categorical perception or memory formation and storage. However, little is known about their long-term dynamics. Using chronic in vivo calcium imaging in the mouse auditory cortex, we find that cell assemblies undergo continuous recombination, even under behaviorally stable conditions. We identify a basis set of elementary operations capturing the dynamics of cell assemblies, which involve plasticity of both the stimulus tuning of particular assemblies as well as the cellular composition of an assembly. Auditory fear conditioning introduces biases in the frequency of specific cell assembly operations that lead to increased association of representations and improved sound encoding. Thus, decomposing cell assembly dynamics during basal conditions into a basis set of elementary operations provides a general framework to explain learning-induced changes in global population activity.

Enhanced Theta and High-Gamma Coupling during Late Stage of Rule Switching Task in Rat Hippocampus

Hippocampal oscillations, particularly theta (6–12 Hz) and gamma (30–90 Hz) frequency bands, play an important role in several cognitive functions. Theta and gamma oscillations show cross-frequency coupling (CFC), wherein the phase of theta rhythm modulates the amplitude of the gamma oscillation, and this CFC is believed to reflect cell assembly dynamics in cognitive processes. Previous studies have reported that CFC strength correlates with the learning process. However, details on these dynamic correlations have not been elucidated. In the present study, we analyzed local field potentials recorded from the rat hippocampus during the learning of a rule-switching task. The modulation index, an index of the CFC strength, became higher in rule-guided behavior than in the no rule condition. The enhanced coupling between theta and high-gamma oscillations (60–90 Hz) changed during the late stage of learning. In contrast, the coupling between theta and low-gamma oscillations (30–60 Hz) did not show any changes during learning. These results suggest that the coupling between theta and gamma bands occurs during rule learning and that high- and low-gamma bands play different roles in rule switching.

Extracellular GABA assisting in organizing dynamic cell assemblies to shorten reaction time to sensory stimulation

Until recently, glia, which exceeds the number of neurons, was considered to only have supportive roles in the central nervous system, providing homeostatic controls and metabolic supports. However, recent studies suggest that glia interacts with neurons and plays active roles in information processing within neuronal circuits. To elucidate how glia contributes to neuronal information processing, we simulated a sensory neuron-glia (neuron-astrocyte) network model. It was investigated in association with ambient (extracellular) GABA level, because the astrocyte has a major role in removing extracellular GABA molecules. In the network model, transporters, embedded in plasma membranes of astrocytes, modulated local ambient GABA levels by actively removing extracellular GABA molecules which persistently acted on receptors in membranes outside synapses and provided pyramidal cells with inhibitory currents. Gap-junction coupling between astrocytes mediated a concordant decrease in local ambient GABA levels, which solicited a prompt population response of pyramidal cells (i.e., activation of an ensemble of pyramidal cells) to a sensory stimulus. As a consequence, the reaction time of a motor network, to which axons of pyramidal cells of the sensory network project, to the sensory stimulus was shortened. We suggest that the astrocytic gap-junction coupling may assist in organizing dynamic cell assemblies by coordinating a reduction in local ambient GABA levels, thereby shortening reaction time to sensory stimulation.

Using neural response properties to draw the distinction between modal and amodal representations

ABSTRACTBarsalou has recently argued against the strategy of identifying amodal neural representations by using their cross-modal responses (i.e., their responses to stimuli from different modalities). I agree that there are indeed modal structures that satisfy this “cross-modal response” criterion (CM), such as distributed and conjunctive modal representations. However, I argue that we can distinguish between modal and amodal structures by looking into differences in their cross-modal responses. A component of a distributed cell assembly can be considered unimodal because its responses to stimuli from a given modality are stable, whereas its responses to stimuli from any other modality are not (i.e., these are lost within a short time, plausibly as a result of cell assembly dynamics). In turn, conjunctive modal representations, such as superior colliculus cells in charge of sensory integration, are multimodal because they have a stable response to stimuli from different modalities. Finally, some prefrontal cells constitute amodal representations because they exhibit what has been called ‘adaptive coding’. This implies that their responses to stimuli from any given modality can be lost when the context and task conditions are modified. We cannot assign them a modality because they have no stable relation with any input type.Abbreviatons: CM: cross-modal response criterion; CCR: conjuntive cross-modal representations; fMRI: functional magnetic resonance imaging; MVPA: multivariate pattern analysis; pre-SMA: pre-supplementary motor area; PFC: prefrontal cortex; SC: superior colliculus; GWS: global workspace

Transient and Persistent UP States during Slow-wave Oscillation and their Implications for Cell-Assembly Dynamics

The membrane potentials of cortical neurons in vivo exhibit spontaneous fluctuations between a depolarized UP state and a resting DOWN state during the slow-wave sleeps or in the resting states. This oscillatory activity is believed to engage in memory consolidation although the underlying mechanisms remain unknown. Recently, it has been shown that UP-DOWN state transitions exhibit significantly different temporal profiles in different cortical regions, presumably reflecting differences in the underlying network structure. Here, we studied in computational models whether and how the connection configurations of cortical circuits determine the macroscopic network behavior during the slow-wave oscillation. Inspired by cortical neurobiology, we modeled three types of synaptic weight distributions, namely, log-normal, sparse log-normal and sparse Gaussian. Both analytic and numerical results suggest that a larger variance of weight distribution results in a larger chance of having significantly prolonged UP states. However, the different weight distributions only produce similar macroscopic behavior. We further confirmed that prolonged UP states enrich the variety of cell assemblies activated during these states. Our results suggest the role of persistent UP states for the prolonged repetition of a selected set of cell assemblies during memory consolidation.

Cell Assembly Dynamics of Sparsely-Connected Inhibitory Networks: A Simple Model for the Collective Activity of Striatal Projection Neurons.

Striatal projection neurons form a sparsely-connected inhibitory network, and this arrangement may be essential for the appropriate temporal organization of behavior. Here we show that a simplified, sparse inhibitory network of Leaky-Integrate-and-Fire neurons can reproduce some key features of striatal population activity, as observed in brain slices. In particular we develop a new metric to determine the conditions under which sparse inhibitory networks form anti-correlated cell assemblies with time-varying activity of individual cells. We find that under these conditions the network displays an input-specific sequence of cell assembly switching, that effectively discriminates similar inputs. Our results support the proposal that GABAergic connections between striatal projection neurons allow stimulus-selective, temporally-extended sequential activation of cell assemblies. Furthermore, we help to show how altered intrastriatal GABAergic signaling may produce aberrant network-level information processing in disorders such as Parkinson’s and Huntington’s diseases.

Neural constraints and flexibility in language processing.

Humans process language with their neurons. Memory in neurons is supported by neural firing and by short- and long-term synaptic weight change; the emergent behaviour of neurons, synchronous firing, and cell assembly dynamics is also a form of memory. As the language signal moves to later stages, it is processed with different mechanisms that are slower but more persistent.

Cell assembly dynamics of sparse inhibitory networks: a simple model for the activity of the Medium Spiny Neurons

Here we show that a simple inhibitory network model, made of sparsely connected Leaky Integrate and Fire (LIF) neurons, is able to retrieve some of the relevant dynamical features of a striatal network, in particular the appearance of cell assembly dynamics as it has been reported in in-vitro experiments of rats striatum [1]. In a first approach, we discuss how our simple model is consistent with the findings in [2,3]. For an optimal choice of the model parameters, the response of the neurons to uniformly distributed constant inputs, occurs in a bursting fashion. As shown in Figure. 1B, the neurons organize their dynamics in groups with correlated bursting-like activity, displaying typical recurrent patterns, similarly to the dynamics of Medium Spiny Neurons (MSNs). Furthermore, the firing of the neurons taking part in this structured cell assembly dynamics is characterized by a high variability, as shown in Figure 1A for a few representative neurons. This high variability is reflected in a coefficient of variation of the interspike-interval (ISI) larger than one. An important aspect of the dynamics of the MSNs is the emergence of coexisting correlated and anti-correlated assemblies, as reported in the experimental work by Carrillo-Reid et al. [1]. Indeed also in our system this aspect is present, as revealed by examining the cross-correlation matrix of the firing rates shown in Figure 1C. Here the neurons are grouped in assemblies accordingly to their level of correlation (as in Figure 1B) and it is evident that the correlated activities within the neuronal assemblies can be highly anti-correlated with other cells in the network. Figure 1 A. Firing rates of 5 selected neurons; B. raster plot activity, the firing times are colored accordingly to the assembly the neurons belong to; C. cross-correlation matrix of the firing rates. The neurons in panel B and C are clusterized accordingly to ...

Interaction between memories in an abstract mathematical model based on the Hebbian cell assembly hypothesis

Learning and memory are essential properties of adaptive neural circuits. Thereby, the general hypothesis (Synaptic-Plasticity-and-Memory hypothesis; [1,2]) is that the adaptive process of synaptic plasticity induces changes at the synapses connecting the neurons. These changes lead to the formation of strongly interconnected subgroups of neurons, so-called cell assemblies [3]. It has been suggested that such cell assemblies represent the learned memory items. However, as known from everyday life, after learning, humans and animals show the remarkable ability to connect, generalize, and discriminate old and new memories. How these memory interactions are realized on a neuronal level based on the idea of cell assemblies is still unknown. In this work, we use a network model dependent on the interaction between synaptic plasticity and synaptic scaling [4,5]. Amongst others, this interaction yields the formation of cell assemblies showing dynamics comparable to human memories [6]. For simplicity, here, we further abstract this complex network model by the methods of mean-field theory. Thereby, the dynamics of each cell assembly in the network are reduced to two differential equations; one for the average neuronal activity and one for the average synaptic weight. Given this simplified model, we show that, in contrast to loosely connected groups of neurons, the formation of cell assemblies lets the activity of the neuronal population follow a hysteresis which creates complex network dynamics, which depend on the initial conditions. Given this hysteresis, in the next step, we connect two such assemblies with each other by plastic connections also adapted by the interaction between synaptic plasticity and scaling. Amongst others, we analyzed under which circumstances the dynamics of this two-cell-assembly system are comparable to the above mentioned dynamics of human and animal memories and how different parameters of the system (e.g., cell assembly size) influence them. In summary, this work is one of the first using a mathematical model to relate the dynamics of memories on a psychological scale to the hypothesized dynamics of cell assemblies on a neuronal scale.

Effect of the small-world structure on encoding performance in the primary visual cortex: an electrophysiological and modeling analysis.

The biological networks have been widely reported to present small-world properties. However, the effects of small-world network structure on population's encoding performance remain poorly understood. To address this issue, we applied a small world-based framework to quantify and analyze the response dynamics of cell assemblies recorded from rat primary visual cortex, and further established a population encoding model based on small world-based generalized linear model (SW-GLM). The electrophysiological experimental results show that the small world-based population responses to different topological shapes present significant variation (t test, p < 0.01; effect size: Hedge's g > 0.8), while no significant variation was found for control networks without considering their spatial connectivity (t test, p > 0.05; effect size: Hedge's g < 0.5). Furthermore, the numerical experimental results show that the predicted response under SW-GLM is more accurate and reliable compared to the control model without small-world structure, and the decoding performance is also improved about 10 % by taking the small-world structure into account. The above results suggest the important role of the small-world neural structure in encoding visual information for the neural population by providing electrophysiological and theoretical evidence, respectively. The study helps greatly to well understand the population encoding mechanisms of visual cortex.

Deficient GABAergic Gliotransmission May Cause Broader Sensory Tuning in Schizophrenia

We examined how the depression of intracortical inhibition due to a reduction in ambient GABA concentration impairs perceptual information processing in schizophrenia. A neural network model with a gliotransmission-mediated ambient GABA regulatory mechanism was simulated. In the network, interneuron-to-glial-cell and principal-cell-to-glial-cell synaptic contacts were made. The former hyperpolarized glial cells and let their transporters import (remove) GABA from the extracellular space, thereby lowering ambient GABA concentration, reducing extrasynaptic GABAa receptor-mediated tonic inhibitory current, and thus exciting principal cells. In contrast, the latter depolarized the glial cells and let the transporters export GABA into the extracellular space, thereby elevating the ambient GABA concentration and thus inhibiting the principal cells. A reduction in ambient GABA concentration was assumed for a schizophrenia network. Multiple dynamic cell assemblies were organized as sensory feature columns. Each cell assembly responded to one specific feature stimulus. The tuning performance of the network to an applied feature stimulus was evaluated in relation to the level of ambient GABA. Transporter-deficient glial cells caused a deficit in GABAergic gliotransmission and reduced ambient GABA concentration, which markedly deteriorated the tuning performance of the network, broadening the sensory tuning. Interestingly, the GABAergic gliotransmission mechanism could regulate local ambient GABA levels: it augmented ambient GABA around stimulus-irrelevant principal cells, while reducing ambient GABA around stimulus-relevant principal cells, thereby ensuring their selective responsiveness to the applied stimulus. We suggest that a deficit in GABAergic gliotransmission may cause a reduction in ambient GABA concentration, leading to a broadening of sensory tuning in schizophrenia. The GABAergic gliotransmission mechanism proposed here may have an important role in the regulation of local ambient GABA levels, thereby improving the sensory tuning performance of the cortex.

A model of hippocampal cell assembly dynamics based on single-cell theta phase precession

Neural oscillations are associated with a wide variety of cognitive and perceptual processes, both in health and disease. In the hippocampus of rodents during exploratory behaviours, prominent theta and gamma oscillations are observed in the local field potential (LFP). These rhythms are linked to spatial cognition and working memory, but the underlying mechanisms are unclear. At the single-cell and cell assembly levels, two salient features emerge during the theta rhythm - phase precession and spatiotemporal spike sequences. The relationship between these two phenomena is yet to be fully characterised. For example, it is unclear whether the sequential structure of hippocampal cell assemblies is fully explained through independent phase coding at the single-cell level, or whether further coordination is required to account for the observed multi-cell behaviour. We developed a descriptive model of phase coding in individual place cells and used this model to investigate the cell assembly dynamics on a linear track. Under the assumption of independent phase coding, key experimental quantities were derived analytically and their relationship to behavioural variables was analysed and compared to experimental data (e.g., [1]). We showed that experimentally established relationships between behavioural variables such as running speed and cell assembly metrics such as the compression factor and lookahead can be reproduced and understood analytically in terms of the collective behaviour of independent phase coding units.

Hippocampal cell assembly dynamics to represent sequential order information

Hippocampal cell assembly dynamics to represent sequential order information

Input dependent cell assembly dynamics in a model of the striatal medium spiny neuron network.

The striatal medium spiny neuron (MSN) network is sparsely connected with fairly weak GABAergic collaterals receiving an excitatory glutamatergic cortical projection. Peri-stimulus time histograms (PSTH) of MSN population response investigated in various experimental studies display strong firing rate modulations distributed throughout behavioral task epochs. In previous work we have shown by numerical simulation that sparse random networks of inhibitory spiking neurons with characteristics appropriate for UP state MSNs form cell assemblies which fire together coherently in sequences on long behaviorally relevant timescales when the network receives a fixed pattern of constant input excitation. Here we first extend that model to the case where cortical excitation is composed of many independent noisy Poisson processes and demonstrate that cell assembly dynamics is still observed when the input is sufficiently weak. However if cortical excitation strength is increased more regularly firing and completely quiescent cells are found, which depend on the cortical stimulation. Subsequently we further extend previous work to consider what happens when the excitatory input varies as it would when the animal is engaged in behavior. We investigate how sudden switches in excitation interact with network generated patterned activity. We show that sequences of cell assembly activations can be locked to the excitatory input sequence and outline the range of parameters where this behavior is shown. Model cell population PSTH display both stimulus and temporal specificity, with large population firing rate modulations locked to elapsed time from task events. Thus the random network can generate a large diversity of temporally evolving stimulus dependent responses even though the input is fixed between switches. We suggest the MSN network is well suited to the generation of such slow coherent task dependent response which could be utilized by the animal in behavior.

Input dependent population dynamics in a model of the inhibitory spiking network of medium spiny neurons in the striatum.

Event Abstract Back to Event Input dependent population dynamics in a model of the inhibitory spiking network of medium spiny neurons in the striatum. A. Ponzi1* and J. Wickens1 1 OIST, Japan We investigate a model of an inhibitory spiking network of striatal GABAergic medium spiny neurons (MSN) by computer simulation. We show that the network exhibits complex spatial temporal population dynamics produced by the occurrence of episodically firing cell assemblies even when the inputs from the excitatory cortical cells are composed of many independent Poisson spiking processes, providing the excitatory inputs obey certain conditions. We also show that cell assembly dynamics can occur for several different choices of MSN cell model and is thus a network generated property, while details of MSN cell spiking statistics do depend on the cell model. When cell input firing rates are dynamically variable we show the network can display input dependent population dynamics and investigate how this depends on excitatory input parameters and also on inhibitory network parameters such as connectivity. We discuss how our results may relate to striatal cognitive functions in behavioural tasks. Conference: EMBO workshop: Gaba Signalling and Brain Networks , Amsterdam, Netherlands, 30 Jun - 2 Jul, 2010. Presentation Type: Poster Presentation Topic: Posters Citation: Ponzi A and Wickens J (2010). Input dependent population dynamics in a model of the inhibitory spiking network of medium spiny neurons in the striatum.. Conference Abstract: EMBO workshop: Gaba Signalling and Brain Networks . doi: 10.3389/conf.fnins.2010.15.00018 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 24 Jun 2010; Published Online: 24 Jun 2010. * Correspondence: A. Ponzi, OIST, Okinawa, Japan, aponz1@oist.jp Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers A. Ponzi J. Wickens Google A. Ponzi J. Wickens Google Scholar A. Ponzi J. Wickens PubMed A. Ponzi J. Wickens Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

Dopamine-modulated dynamic cell assemblies generated by the GABAergic striatal microcircuit

Frontiers Events is a rapidly growing calendar management system dedicated to the scheduling of academic events. This includes announcements and invitations, participant listings and search functionality, abstract handling and publication, related events and post-event exchanges. Whether an organizer or participant, make your event a Frontiers Event!

Input dependent cell assembly dynamics in an inhibitory spiking network model

Input dependent cell assembly dynamics in an inhibitory spiking network model

A dual information task for analyzing cell assembly dynamics underlying different internal cognitive process

A dual information task for analyzing cell assembly dynamics underlying different internal cognitive process