Hippocampal sharp wave-ripples and the associated sequence replay emerge from structured synaptic interactions in a network model of area CA3

Unraveling Multiregional Neural Patterns during Consciousness Transition Using Flexible Microelectrode Arrays Integrated with Neuropixels Chips

Consciousness transitions, including awakening from anesthesia or falling asleep, involve complex neural dynamics across multiple brain regions. Understanding these transitions requires simultaneous and stable monitoring of large-scale neural activity in freely moving animals. Here, a flexible microelectrode array system (FlexiPixels probe) is demonstrated that integrates a multishank flexible microelectrode array with Neuropixels chips. This lightweight FlexiPixels probe enables stable and long-term neural signal recording across multiple brain regions in freely moving rats and tracking of neuronal activities during consciousness transitions from anesthesia to wakefulness and subsequent sleep states. Distinct state-dependent firing patterns emerge across different brain regions and neuronal types. CA1 neurons show similar activity during wakefulness and sleep, while DG neurons exhibit unique anesthesia sensitivity. These findings demonstrate FlexiPixels' capabilities for stable multiregion neural recording in freely moving animals and potential to unravel region-specific signatures in consciousness studies.

Spontaneous alternation of place-cell sequences in the open field through spike frequency adaptation

Spatial sequences encoded by cells in the hippocampal-entorhinal region have been observed to spontaneously alternate across the animal's midline during navigation in the open field, but it is unknown how this occurs. We show that sinusoidal sampling patterns emerge rapidly and robustly in a simple model of the hippocampal place-cell sequences based on spike frequency adaptation that makes no assumptions about sequence direction. We corroborate our findings using hippocampal data from rats performing a spatial memory task in the open field.

Integrating multimodal data to understand cortical circuit architecture and function

In recent years there has been a tremendous growth in new technologies that allow large-scale investigation of different characteristics of the nervous system at an unprecedented level of detail. There is a growing trend to use combinations of these new techniques to determine direct links between different modalities. In this Perspective, we focus on the mouse visual cortex, as this is one of the model systems in which much progress has been made in the integration of multimodal data to advance understanding. We review several approaches that allow integration of data regarding various properties of cortical cell types, connectivity at the level of brain areas, cell types and individual cells, and functional neural activity in vivo. The increasingly crucial contributions of computation and theory in analyzing and systematically modeling data are also highlighted. Together with open sharing of data, tools and models, integrative approaches are essential tools in modern neuroscience for improving our understanding of the brain architecture, mechanisms and function.

Compensatory Regulation of Excitation/Inhibition Balance in the Ventral Hippocampus: Insights from Fragile X Syndrome

The excitation/inhibition (E/I) balance is a critical feature of neural circuits, which is crucial for maintaining optimal brain function by ensuring network stability and preventing neural hyperexcitability. The hippocampus exhibits the particularly interesting characteristics of having different functions and E/I profiles between its dorsal and ventral segments. Furthermore, the hippocampus is particularly vulnerable to epilepsy and implicated in Fragile X Syndrome (FXS), disorders associated with heightened E/I balance and possible deficits in GABA-mediated inhibition. In epilepsy, the ventral hippocampus shows heightened susceptibility to seizures, while in FXS, recent evidence suggests differential alterations in excitability and inhibition between dorsal and ventral regions. This article explores the mechanisms underlying E/I balance regulation, focusing on the hippocampus in epilepsy and FXS, and emphasizing the possible mechanisms that may confer homeostatic flexibility to the ventral hippocampus in maintaining E/I balance. Notably, the ventral hippocampus in adult FXS models shows enhanced GABAergic inhibition, resistance to epileptiform activity, and physiological network pattern (sharp wave-ripples, SWRs), potentially representing a homeostatic adaptation. In contrast, the dorsal hippocampus in these FXS models is more vulnerable to aberrant discharges and displays altered SWRs. These findings highlight the complex, region-specific nature of E/I balance disruptions in neurological disorders and suggest that the ventral hippocampus may possess unique compensatory mechanisms. Specifically, it is proposed that the ventral hippocampus, the brain region most prone to hyperexcitability, may have unique adaptive capabilities at the cellular and network levels that maintain the E/I balance within a normal range to prevent the transition to hyperexcitability and preserve normal function. Investigating the mechanisms underlying these compensatory responses in the ventral hippocampus and their developmental trajectories may offer novel insights into strategies for mitigating E/I imbalances in epilepsy, FXS, and potentially other neuropsychiatric and neurodevelopmental disorders.

Timing is everything: The effect of early-life seizures on developing neuronal circuits subserving spatial memory

Spatial memory, the aspect of memory involving encoding and retrieval of information regarding one's environment and spatial orientation, is a complex biological function incorporating multiple neuronal networks. Hippocampus-dependent spatial memory is not innate and emerges during development in both humans and rodents. For spatial memory to occur, the hippocampus forms highly associative networks integrating external inputs conveying multi-sensory, proprioceptive, contextual, and emotional information onto internally generated dynamics. Hippocampal cognitive maps are produced by sequences of transient ordered neuronal activations that represent not only spatial information but also the temporal order of events in a memory episode. This patterned activity fine-tunes synaptic connectivity of the network and drives the emergence of specific firing necessary for spatial memory. In the rodent hippocampus, there is a sequence of spontaneous activities that are precisely timed, starting with early sharp waves progressing to theta and gamma oscillations, place and grid cell firing, and sharp wave-ripples that must occur for spatial memory to develop. Whereas normal activity patterns are required for circuit maturation, aberrant neuronal activity during development can have major adverse consequences, disrupting the development of spatial memory. Seizures during infancy, involving massive bursts of synchronized network activity, result in impaired spatial memory when animals are tested as adolescents or adults. This impaired spatial memory is accompanied by alterations in theta and gamma oscillations and spatial and temporal coding of place cells. Conversely, enhancement of oscillatory activity following early-life seizures can improve cognitive impairment. The plasticity of developing oscillatory activity in the immature brain provides exciting opportunities for therapeutic intervention in childhood epilepsy. PLAIN LANGUAGE SUMMARY: Children with epilepsy often struggle with memory and learning challenges. Research has shown that seizures can interfere with the brain's natural rhythms, which are crucial for these processes. Seizures in children are particularly harmful because they disrupt the development of brain connections, which are still growing and maturing during this critical time. Exciting new studies in both animals and humans suggest that using electrical or magnetic stimulation to adjust these brain rhythms can help restore memory and learning abilities. This breakthrough offers hope for improving the lives of children with epilepsy.

Cyto-, gene, and multireceptor architecture of the early postnatal mouse hippocampal complex

Neurotransmitter receptors are key molecules in signal transmission in the adult brain, and their precise spatial and temporal balance expressions also play a critical role in normal brain development. However, the specific balance expression of multiple receptors during hippocampal development is not well characterized. In this study, we used quantitative in vivo receptor autoradiography to measure the distributions and densities of 18 neurotransmitter receptor types in the mouse hippocampal complex at postnatal day 7, and compared them with the expressions of their corresponding encoding genes. We provide a novel and comprehensive characterization of the cyto-, gene, and multireceptor architecture of the developing mouse hippocampal and subicular regions during the developmental period, which typically differs from that in the adult brain. High-density receptor expressions with distinct regional and laminar distributions were observed for AMPA, Kainate, mGluR2/3, GABAA, GABAA/BZ, α2, and A1 receptors during this specific period, whereas NMDA, GABAB, α1, M1, M2, M3, nicotinic α4β2, 5-HT1A, 5-HT2, D1 and D2/D3 receptors exhibited relatively low and homogeneous expressions. This specific balance of multiple receptors aligns with regional cytoarchitecture, neurotransmitter distributions, and gene expressions. Moreover, contrasting with previous findings, we detected a high α2 receptor density, with distinct regional and laminar distribution patterns. A non-covariation differentiation phenomenon between α2 receptor distributions and corresponding gene expressions is also demonstrated in this early developmental period. The multimodal data provides new insights into understanding the hippocampal development from the perspective of cell, gene, and multireceptor levels, and contributes important resources for further interdisciplinary analyses.

The time course and organization of hippocampal replay

The mechanisms by which the brain replays neural activity sequences remain unknown. Recording from large ensembles of hippocampal place cells in freely behaving rats, we observed that replay content is strictly organized over multiple timescales and governed by self-avoidance. After movement cessation, replays avoided the animal's previous path for 3 seconds. Chains of replays avoided self-repetition over a shorter timescale. We used a continuous attractor model of neural activity to demonstrate that neuronal fatigue both generates replay sequences and produces self-avoidance over the observed timescales. In addition, replay of past experience became predominant later into the stopping period, in a manner requiring cortical input. These results indicate a mechanism for replay generation that unexpectedly constrains which sequences can be produced across time.

Evaluation and comparison of methods for neuronal parameter optimization using the Neuroptimus software framework

Finding optimal parameters for detailed neuronal models is a ubiquitous challenge in neuroscientific research. In recent years, manual model tuning has been gradually replaced by automated parameter search using a variety of different tools and methods. However, using most of these software tools and choosing the most appropriate algorithm for a given optimization task require substantial technical expertise, which prevents the majority of researchers from using these methods effectively. To address these issues, we developed a generic platform (called Neuroptimus) that allows users to set up neural parameter optimization tasks via a graphical interface, and to solve these tasks using a wide selection of state-of-the-art parameter search methods implemented by five different Python packages. Neuroptimus also offers several features to support more advanced usage, including the ability to run most algorithms in parallel, which allows it to take advantage of high-performance computing architectures. We used the common interface provided by Neuroptimus to conduct a detailed comparison of more than twenty different algorithms (and implementations) on six distinct benchmarks that represent typical scenarios in neuronal parameter search. We quantified the performance of the algorithms in terms of the best solutions found and in terms of convergence speed. We identified several algorithms, including covariance matrix adaptation evolution strategy and particle swarm optimization, that consistently, without any fine-tuning, found good solutions in all of our use cases. By contrast, some other algorithms including all local search methods provided good solutions only for the simplest use cases, and failed completely on more complex problems. We also demonstrate the versatility of Neuroptimus by applying it to an additional use case that involves tuning the parameters of a subcellular model of biochemical pathways. Finally, we created an online database that allows uploading, querying and analyzing the results of optimization runs performed by Neuroptimus, which enables all researchers to update and extend the current benchmarking study. The tools and analysis we provide should aid members of the neuroscience community to apply parameter search methods more effectively in their research.

Interictal spikes during spatial working memory carry helpful or distracting representations of space and have opposing impacts on performance

In temporal lobe epilepsy, interictal spikes (IS) - hypersynchronous bursts of network activity - occur at high rates in between seizures. We sought to understand the influence of IS on working memory by recording hippocampal local field potentials from epileptic mice while they performed a delayed alternation task. We found that IS disrupted performance when they were spatially non-restricted and occurred during running. In contrast, when IS were clustered at reward locations, animals performed well. A machine learning decoding approach revealed that IS at reward sites were larger than IS elsewhere on the maze, and could be classified as occurring at specific reward locations - suggesting they carry informative content for the memory task. Finally, a spiking model revealed that spatially clustered IS preserved hippocampal replay, while spatially dispersed IS disrupted replay by causing over-generalization. Together, these results show that IS can have opposing outcomes on memory.

Distinct mechanisms and functions of episodic memory

The concept of episodic memory (EM) faces significant challenges by two claims: EM might not be a distinct memory system, and EM might be an epiphenomenon of a more general capacity for mental time travel (MTT). Nevertheless, the observations leading to these arguments do not preclude the existence of a mechanically and functionally distinct EM system. First, modular systems, like cognition, can have distinct subsystems that may not be distinguishable in the system's final output. EM could be such a subsystem, even though its effects may be difficult to distinguish from those of other subsystems. Second, EM could have a distinct and consistent low-level function, which is used in diverse high-level functions such as MTT. This article introduces the scenario construction framework, proposing that EM crucially rests on memory traces containing the gist of an episodic experience. During retrieval, EM traces trigger the reconstruction of semantic representations, which were active during the remembered episode, and are further enriched with semantic information, to generate a scenario of the past experience. This conceptualization of EM is consistent with studies on the neural basis of EM and resolves the two challenges while retaining the key properties associated with EM. This article is part of the theme issue 'Elements of episodic memory: lessons from 40 years of research'.

Inhibitory plasticity supports replay generalization in the hippocampus

Memory consolidation assimilates recent experiences into long-term memory. This process requires the replay of learned sequences, although the content of these sequences remains controversial. Recent work has shown that the statistics of replay deviate from those of experience: stimuli that are experientially salient may be either recruited or suppressed from sharp-wave ripples. In this study, we found that this phenomenon can be explained parsimoniously and biologically plausibly by a Hebbian spike-time-dependent plasticity rule at inhibitory synapses. Using models at three levels of abstraction-leaky integrate-and-fire, biophysically detailed and abstract binary-we show that this rule enables efficient generalization, and we make specific predictions about the consequences of intact and perturbed inhibitory dynamics for network dynamics and cognition. Finally, we use optogenetics to artificially implant non-generalizable representations into the network in awake behaving mice, and we find that these representations also accumulate inhibition during sharp-wave ripples, experimentally validating a major prediction of our model. Our work outlines a potential direct link between the synaptic and cognitive levels of memory consolidation, with implications for both normal learning and neurological disease.

Learning, Fast and Slow: Single- and Many-Shot Learning in the Hippocampus

The hippocampus is critical for memory and spatial navigation. The ability to map novel environments, as well as more abstract conceptual relationships, is fundamental to the cognitive flexibility that humans and other animals require to survive in a dynamic world. In this review, we survey recent advances in our understanding of how this flexibility is implemented anatomically and functionally by hippocampal circuitry, during both active exploration (online) and rest (offline). We discuss the advantages and limitations of spike timing-dependent plasticity and the more recently discovered behavioral timescale synaptic plasticity in supporting distinct learning modes in the hippocampus. Finally, we suggest complementary roles for these plasticity types in explaining many-shot and single-shot learning in the hippocampus and discuss how these rules could work together to support the learning of cognitive maps.

Self-avoidance dominates the selection of hippocampal replay

Spontaneous neural activity sequences are generated by the brain in the absence of external input 1-12 , yet how they are produced remains unknown. During immobility, hippocampal replay sequences depict spatial paths related to the animal's past experience or predicted future 13 . By recording from large ensembles of hippocampal place cells 14 in combination with optogenetic manipulation of cortical input in freely behaving rats, we show here that the selection of hippocampal replay is governed by a novel self-avoidance principle. Following movement cessation, replay of the animal's past path is strongly avoided, while replay of the future path predominates. Moreover, when the past and future paths overlap, early replays avoid both and depict entirely different trajectories. Further, replays avoid self-repetition, on a shorter timescale compared to the avoidance of previous behavioral trajectories. Eventually, several seconds into the stopping period, replay of the past trajectory dominates. This temporal organization contrasts with established and recent predictions 9,10,15,16 but is well-recapitulated by a symmetry-breaking attractor model of sequence generation in which individual neurons adapt their firing rates over time 26-35 . However, while the model is sufficient to produce avoidance of recently traversed or reactivated paths, it requires an additional excitatory input into recently activated cells to produce the later window of past-dominance. We performed optogenetic perturbations to demonstrate that this input is provided by medial entorhinal cortex, revealing its role in maintaining a memory of past experience that biases hippocampal replay. Together, these data provide specific evidence for how hippocampal replays are generated.

CA3 Circuit Model Compressing Sequential Information in Theta Oscillation and Replay

The hippocampus plays a critical role in the compression and retrieval of sequential information. During wakefulness, it achieves this through theta phase precession and theta sequences. Subsequently, during periods of sleep or rest, the compressed information reactivates through sharp-wave ripple events, manifesting as memory replay. However, how these sequential neuronal activities are generated and how they store information about the external environment remain unknown. We developed a hippocampal cornu ammonis 3 (CA3) computational model based on anatomical and electrophysiological evidence from the biological CA3 circuit to address these questions. The model comprises theta rhythm inhibition, place input, and CA3-CA3 plastic recurrent connection. The model can compress the sequence of the external inputs, reproduce theta phase precession and replay, learn additional sequences, and reorganize previously learned sequences. A gradual increase in synaptic inputs, controlled by interactions between theta-paced inhibition and place inputs, explained the mechanism of sequence acquisition. This model highlights the crucial role of plasticity in the CA3 recurrent connection and theta oscillational dynamics and hypothesizes how the CA3 circuit acquires, compresses, and replays sequential information.

Intra-ripple frequency accommodation in an inhibitory network model for hippocampal ripple oscillations

Hippocampal ripple oscillations have been implicated in important cognitive functions such as memory consolidation and planning. Multiple computational models have been proposed to explain the emergence of ripple oscillations, relying either on excitation or inhibition as the main pacemaker. Nevertheless, the generating mechanism of ripples remains unclear. An interesting dynamical feature of experimentally measured ripples, which may advance model selection, is intra-ripple frequency accommodation (IFA): a decay of the instantaneous ripple frequency over the course of a ripple event. So far, only a feedback-based inhibition-first model, which relies on delayed inhibitory synaptic coupling, has been shown to reproduce IFA. Here we use an analytical mean-field approach and numerical simulations of a leaky integrate-and-fire spiking network to explain the mechanism of IFA. We develop a drift-based approximation for the oscillation dynamics of the population rate and the mean membrane potential of interneurons under strong excitatory drive and strong inhibitory coupling. For IFA, the speed at which the excitatory drive changes is critical. We demonstrate that IFA arises due to a speed-dependent hysteresis effect in the dynamics of the mean membrane potential, when the interneurons receive transient, sharp wave-associated excitation. We thus predict that the IFA asymmetry vanishes in the limit of slowly changing drive, but is otherwise a robust feature of the feedback-based inhibition-first ripple model.

Associative and predictive hippocampal codes support memory-guided behaviors

Episodic memory involves learning and recalling associations between items and their spatiotemporal context. Those memories can be further used to generate internal models of the world that enable predictions to be made. The mechanisms that support these associative and predictive aspects of memory are not yet understood. In this study, we used an optogenetic manipulation to perturb the sequential structure, but not global network dynamics, of place cells as rats traversed specific spatial trajectories. This perturbation abolished replay of those trajectories and the development of predictive representations, leading to impaired learning of new optimal trajectories during memory-guided navigation. However, place cell assembly reactivation and reward-context associative learning were unaffected. Our results show a mechanistic dissociation between two complementary hippocampal codes: an associative code (through coactivity) and a predictive code (through sequences).

Phase information is conserved in sparse, synchronous population-rate-codes via phase-to-rate recoding

Neural computation is often traced in terms of either rate- or phase-codes. However, most circuit operations will simultaneously affect information across both coding schemes. It remains unclear how phase and rate coded information is transmitted, in the face of continuous modification at consecutive processing stages. Here, we study this question in the entorhinal cortex (EC)- dentate gyrus (DG)- CA3 system using three distinct computational models. We demonstrate that DG feedback inhibition leverages EC phase information to improve rate-coding, a computation we term phase-to-rate recoding. Our results suggest that it i) supports the conservation of phase information within sparse rate-codes and ii) enhances the efficiency of plasticity in downstream CA3 via increased synchrony. Given the ubiquity of both phase-coding and feedback circuits, our results raise the question whether phase-to-rate recoding is a recurring computational motif, which supports the generation of sparse, synchronous population-rate-codes in areas beyond the DG.

Hippocampal memory reactivation during sleep is correlated with specific cortical states of the retrosplenial and prefrontal cortices

Episodic memories are thought to be stabilized through the coordination of cortico-hippocampal activity during sleep. However, the timing and mechanism of this coordination remain unknown. To investigate this, we studied the relationship between hippocampal reactivation and slow-wave sleep up and down states of the retrosplenial cortex (RTC) and prefrontal cortex (PFC). We found that hippocampal reactivations are strongly correlated with specific cortical states. Reactivation occurred during sustained cortical Up states or during the transition from up to down state. Interestingly, the most prevalent interaction with memory reactivation in the hippocampus occurred during sustained up states of the PFC and RTC, while hippocampal reactivation and cortical up-to-down state transition in the RTC showed the strongest coordination. Reactivation usually occurred within 150-200 msec of a cortical Up state onset, indicating that a buildup of excitation during cortical Up state activity influences the probability of memory reactivation in CA1. Conversely, CA1 reactivation occurred 30-50 msec before the onset of a cortical down state, suggesting that memory reactivation affects down state initiation in the RTC and PFC, but the effect in the RTC was more robust. Our findings provide evidence that supports and highlights the complexity of bidirectional communication between cortical regions and the hippocampus during sleep.

Offline memory replay in recurrent neuronal networks emerges from constraints on online dynamics

During spatial exploration, neural circuits in the hippocampus store memories of sequences of sensory events encountered in the environment. When sensory information is absent during 'offline' resting periods, brief neuronal population bursts can 'replay' sequences of activity that resemble bouts of sensory experience. These sequences can occur in either forward or reverse order, and can even include spatial trajectories that have not been experienced, but are consistent with the topology of the environment. The neural circuit mechanisms underlying this variable and flexible sequence generation are unknown. Here we demonstrate in a recurrent spiking network model of hippocampal area CA3 that experimental constraints on network dynamics such as population sparsity, stimulus selectivity, rhythmicity and spike rate adaptation, as well as associative synaptic connectivity, enable additional emergent properties, including variable offline memory replay. In an online stimulus-driven state, we observed the emergence of neuronal sequences that swept from representations of past to future stimuli on the timescale of the theta rhythm. In an offline state driven only by noise, the network generated both forward and reverse neuronal sequences, and recapitulated the experimental observation that offline memory replay events tend to include salient locations like the site of a reward. These results demonstrate that biological constraints on the dynamics of recurrent neural circuits are sufficient to enable memories of sensory events stored in the strengths of synaptic connections to be flexibly read out during rest and sleep, which is thought to be important for memory consolidation and planning of future behaviour. KEY POINTS: A recurrent spiking network model of hippocampal area CA3 was optimized to recapitulate experimentally observed network dynamics during simulated spatial exploration. During simulated offline rest, the network exhibited the emergent property of generating flexible forward, reverse and mixed direction memory replay events. Network perturbations and analysis of model diversity and degeneracy identified associative synaptic connectivity and key features of network dynamics as important for offline sequence generation. Network simulations demonstrate that population over-representation of salient positions like the site of reward results in biased memory replay.

Spike timing-dependent plasticity and memory

Spike timing-dependent plasticity (STDP) is a bidirectional form of synaptic plasticity discovered about 30 years ago and based on the relative timing of pre- and post-synaptic spiking activity with a millisecond precision. STDP is thought to be involved in the formation of memory but the millisecond-precision spike-timing required for STDP is difficult to reconcile with the much slower timescales of behavioral learning. This review therefore aims to expose and discuss recent findings about i) the multiple STDP learning rules at both excitatory and inhibitory synapses in vitro, ii) the contribution of STDP-like synaptic plasticity in the formation of memory in vivo and iii) the implementation of STDP rules in artificial neural networks and memristive devices.

How our understanding of memory replay evolves

Memory reactivations and replay, widely reported in the hippocampus and cortex across species, have been implicated in memory consolidation, planning, and spatial and skill learning. Technological advances in electrophysiology, calcium imaging, and human neuroimaging techniques have enabled neuroscientists to measure large-scale neural activity with increasing spatiotemporal resolution and have provided opportunities for developing robust analytic methods to identify memory replay. In this article, we first review a large body of historically important and representative memory replay studies from the animal and human literature. We then discuss our current understanding of memory replay functions in learning, planning, and memory consolidation and further discuss the progress in computational modeling that has contributed to these improvements. Next, we review past and present analytic methods for replay analyses and discuss their limitations and challenges. Finally, looking ahead, we discuss some promising analytic methods for detecting nonstereotypical, behaviorally nondecodable structures from large-scale neural recordings. We argue that seamless integration of multisite recordings, real-time replay decoding, and closed-loop manipulation experiments will be essential for delineating the role of memory replay in a wide range of cognitive and motor functions.

Mesoscopic description of hippocampal replay and metastability in spiking neural networks with short-term plasticity

Bottom-up models of functionally relevant patterns of neural activity provide an explicit link between neuronal dynamics and computation. A prime example of functional activity patterns are propagating bursts of place-cell activities called hippocampal replay, which is critical for memory consolidation. The sudden and repeated occurrences of these burst states during ongoing neural activity suggest metastable neural circuit dynamics. As metastability has been attributed to noise and/or slow fatigue mechanisms, we propose a concise mesoscopic model which accounts for both. Crucially, our model is bottom-up: it is analytically derived from the dynamics of finite-size networks of Linear-Nonlinear Poisson neurons with short-term synaptic depression. As such, noise is explicitly linked to stochastic spiking and network size, and fatigue is explicitly linked to synaptic dynamics. To derive the mesoscopic model, we first consider a homogeneous spiking neural network and follow the temporal coarse-graining approach of Gillespie to obtain a "chemical Langevin equation", which can be naturally interpreted as a stochastic neural mass model. The Langevin equation is computationally inexpensive to simulate and enables a thorough study of metastable dynamics in classical setups (population spikes and Up-Down-states dynamics) by means of phase-plane analysis. An extension of the Langevin equation for small network sizes is also presented. The stochastic neural mass model constitutes the basic component of our mesoscopic model for replay. We show that the mesoscopic model faithfully captures the statistical structure of individual replayed trajectories in microscopic simulations and in previously reported experimental data. Moreover, compared to the deterministic Romani-Tsodyks model of place-cell dynamics, it exhibits a higher level of variability regarding order, direction and timing of replayed trajectories, which seems biologically more plausible and could be functionally desirable. This variability is the product of a new dynamical regime where metastability emerges from a complex interplay between finite-size fluctuations and local fatigue.