arXiv subjects

Miguel A. Muñoz

Publications and source records attributed to Miguel A. Muñoz.

At least 37 records · Page 2Linked to original sources

arXiv2022

Autocorrelations from emergent bistability in homeostatic spiking neural networks on neuromorphic hardware

A unique feature of neuromorphic computing is that memory is an implicit part of processing through traces of past information in the system's collective dynamics. The extent of memory about past inputs is commonly quantified by the autocorrelation time of collective dynamics. Based on past experimental evidence, a potential explanation for the underlying autocorrelations are close-to-critical fluctuations. Here, we show for self-organized networks of excitatory and inhibitory leaky integrate-and-fire neurons that autocorrelations can originate from emergent bistability upon reducing external input strength. We identify the bistability as a fluctuation-induced stochastic switching between metastable active and quiescent states in the vicinity of a non-equilibrium phase transition. This bistability occurs for networks with fixed heterogeneous weights as a consequence of homeostatic self-organization during development. Specifically, in our experiments on neuromorphic hardware and in computer simulations, the emergent bistability gives rise to autocorrelation times exceeding 500 ms despite single-neuron timescales of only 20 ms. Our results provide the first verification of biologically compatible autocorrelation times in networks of leaky integrate-and-fire neurons, which here are not generated by close-to-critical fluctuations but by emergent bistability in homeostatically regulated networks. Our results thereby constitute a new, complementary mechanism for emergent autocorrelations in networks of spiking neurons, with implications for biological and artificial networks, and introduces the general paradigm of fluctuation-induced bistability for driven systems with absorbing states.

q-bio.NC↗

arXiv2022

The excitatory-inhibitory branching process: a parsimonious view of cortical asynchronous states, excitability, and criticality

The branching process is the minimal model for propagation dynamics, avalanches and criticality, broadly used in neuroscience. A simple extension of it, adding inhibitory nodes, induces a much-richer phenomenology, including, an intermediate phase, between quiescence and saturation, that exhibits the key features of "asynchronous states" in cortical networks. Remarkably, in the inhibition-dominated case, it exhibits an extremely-rich phase diagram, that captures a wealth of non-trivial features of spontaneous brain activity, such as collective excitability, hysteresis, tilted avalanche shapes, and partial synchronization, allowing us to rationalize striking empirical findings within a common and parsimonious framework.

cond-mat.stat-mech↗

arXiv2022

Modular architecture facilitates noise-driven control of synchrony in neuronal networks

Brain functions require both segregated processing of information in specialized circuits, as well as integration across circuits to perform high-level information processing. One possible way to implement these seemingly opposing demands is by flexibly switching between synchronous and less synchronous states. Understanding how complex synchronization patterns are controlled by the interaction of network architecture and external perturbations is thus a central challenge in neuroscience, but the mechanisms behind such interactions remain elusive. Here, we utilise precision neuroengineering to manipulate cultured neuronal networks and show that a modular architecture facilitates desynchronization upon asynchronous stimulation, making external noise a control parameter of synchrony. Using spiking neuron models, we then demonstrate that external noise can reduce the level of available synaptic resources, which make intermodular interactions more stochastic and thereby facilitates the breakdown of synchrony. Finally, the phenomenology of stochastic intermodular interactions is formulated into a mesoscopic model that incorporates a state-dependent gating mechanism for signal propagation. Taken together, our results demonstrate a network mechanism by which asynchronous inputs tune the inherent dynamical state in structured networks of excitable units.

q-bio.NC↗

arXiv2021

Phenotypic-dependent variability and the emergence of tolerance in bacterial populations

Ecological and evolutionary dynamics have been historically regarded as unfolding at broadly separated timescales. However, these two types of processes are nowadays well documented to much more tightly than traditionally assumed, especially in communities of microorganisms. With this motivation in mind, here we scrutinize recent experimental results showing evidence of rapid evolution of tolerance by lag in bacterial populations that are periodically exposed to antibiotic stress in laboratory conditions. In particular, the distribution of single-cell lag times evolves its average value to approximately fit the antibiotic-exposure time. Moreover, the distribution develops right skewed heavy tails, revealing the presence of individuals with anomalously large lag times. Here, we develop a parsimonious individual-based model mimicking the actual demographic processes of the experimental setup. Individuals are characterized by a single phenotypic trait: their intrinsic lag time, which is transmitted with variation to the progeny. The model (in a version in which the amplitude of phenotypic variations grows with the parentś lag time) is able to reproduce quite well the key empirical observations. Furthermore, we develop a general mathematical framework allowing us to describe with good accuracy the properties of the stochastic model by means of a macroscopic equation, which generalizes the Crow Kimura equation in population genetics. From a broader perspective, this work represents a benchmark for the mathematical framework designed to tackle much more general eco-evolutionary problems, thus paving the road to further research avenues.

q-bio.PE↗

arXiv2021

The broad edge of synchronisation: Griffiths effects and collective phenomena in brain networks

Many of the amazing functional capabilities of the brain are collective properties stemming from the interactions of large sets of individual neurons. In particular, the most salient collective phenomena in brain activity are oscillations, which require the synchronous activation of many neurons. Here, we analyse parsimonious dynamical models of neural synchronisation running on top of synthetic networks that capture essential aspects of the actual brain anatomical connectivity such as a hierarchical-modular and core-periphery structure. These models reveal the emergence of complex collective states with intermediate and flexible levels of synchronisation, halfway in the synchronous-asynchronous spectrum. These states are best described as broad Griffiths-like phases, i.e. an extension of standard critical points that emerge in structurally heterogeneous systems. We analyse different routes (bifurcations) to synchronisation and stress the relevance of 'hybrid-type transitions' to generate rich dynamical patterns. Overall, our results illustrate the complex interplay between structure and dynamics, underlining key aspects leading to rich collective states needed to sustain brain functionality.

q-bio.NC↗

arXiv2021

Optimal input representation in neural systems at the edge of chaos

Shedding light onto how biological systems represent, process and store information in noisy environments is a key and challenging goal. A stimulating, though controversial, hypothesis poses that operating in dynamical regimes near the edge of a phase transition, i.e. at criticality or the "edge of chaos", can provide information-processing living systems with important operational advantages, creating, e.g., an optimal trade-off between robustness and flexibility. Here, we elaborate on a recent theoretical result, which establishes that the spectrum of covariance matrices of neural networks representing complex inputs in a robust way needs to decay as a power-law of the rank, with an exponent close to unity, a result that has been indeed experimentally verified in neurons of the mouse visual cortex. Aimed at understanding and mimicking these results, we construct an artificial neural network and train it to classify images. Remarkably, we find that the best performance in such a task is obtained when the network operates near the critical point, at which the eigenspectrum of the covariance matrix follows the very same statistics as actual neurons do. Thus, we conclude that operating near criticality can also have -- besides the usually alleged virtues -- the advantage of allowing for flexible, robust and efficient input representations.

cond-mat.dis-nn↗

arXiv2021

Hybrid-type synchronization transitions: where marginal coherence, scale-free avalanches, and bistability live together

The human cortex is never at rest but in a state of sparse and noisy neural activity that can be detected at broadly diverse resolution scales. It has been conjectured that such a state is best described as a critical dynamical process -- whose nature is still not fully understood -- where scale-free avalanches of activity emerge at the edge of a phase transition. In particular, some works suggest that this is most likely a synchronization transition, separating synchronous from asynchronous phases. Here, by investigating a simplified model of coupled excitable oscillators describing the cortex dynamics at a mesoscopic level, we investigate the possible nature of such a synchronization phase transition. Within our modeling approach, we conclude that -- in order to reproduce all key empirical observations, such as scale-free avalanches and bistability, on which fundamental functional advantages rely -- the transition to collective oscillatory behavior needs to be of an unconventional hybrid type, with mixed features of type-I and type-II excitability, opening the possibility for a particularly rich dynamical repertoire.

cond-mat.stat-mech↗

arXiv2021

Measles-induced immune amnesia and its effects in concurrent epidemics

It has been recently discovered that the measles virus can wipe out the adaptive immune system, destroying B lymphocytes and reducing the diversity of non-specific B cells of the infected host. In particular, this implies that previously acquired immunization from vaccination or direct exposition to other pathogens could be erased in a phenomenon named "immune amnesia", whose effects can become particularly worrisome given the actual rise of anti-vaccination movements. Here we present the first attempt to incorporate immune amnesia into standard models of epidemic spreading. In particular, we analyze diverse variants of a model that describes the spreading of two concurrent pathogens causing measles and another generic disease: the SIR-IA model. Analytical and computational studies confirm that immune amnesia can indeed have important consequences for epidemic spreading, significantly altering the vaccination coverage required to reach herd-immunity for concurring infectious diseases. More specifically, we uncover the existence of novel propagating and endemic phases which are induced by immune amnesia, that appear both in fully-connected and more structured networks, such as random networks and power-law degree-distributed ones. In particular, the transitions from a quiescent state into these novel phases can become rather abrupt in some cases that we specifically analyze. Furthermore, we discuss the meaning and consequences of our results and their relation with, e.g., immunization strategies, together with the possibility that explosive types of transitions may emerge, making immune-amnesia effects particularly dramatic. This work opens the door to further developments and analyses of immune amnesia effects, contributing, more generally, to the theory of interacting epidemics on complex networks.

cond-mat.stat-mech↗

arXiv2020

Small variation in dynamic functional connectivity in cerebellar networks

Brain networks can be defined and explored through their connectivity. Here, we analyzed the relationship between structural connectivity (SC) across 2,514 regions that cover the entire brain and brainstem, and their dynamic functional connectivity (DFC). To do so, we focused on a combination of two metrics: the first assesses the degree of SC-DFC similarity and the second is the intrinsic variability of the DFC networks over time. Overall, we found that cerebellar networks have a smaller DFC variability than other networks in the brain. Moreover, the internal structure of the cerebellum could be clearly divided in two distinct posterior and anterior parts, the latter also connected to the brainstem. The mechanism to maintain small variability of the DFC in the posterior part of the cerebellum is consistent with another of our findings, namely, that this structure exhibits the highest SC-DFC similarity relative to the other networks studied. By contrast, the anterior part of the cerebellum also exhibits small DFC variability but it has the lowest SC-DFC similarity, suggesting a different mechanism is at play. Because this structure connects to the brainstem, which regulates sleep cycles, cardiac and respiratory functioning, we suggest that such critical functionality drives the low variability in the DFC. Overall, the low variability detected in DFC expands our current knowledge of cerebellar networks, which are extremely rich and complex, participating in a wide range of cognitive functions, from movement control and coordination to executive function or emotional regulation. Moreover, the association between such low variability and structure suggests that differentiated computational principles can be applied in the cerebellum as opposed to other structures, such as the cerebral cortex.

q-bio.NC↗

arXiv2020

Classes of critical avalanche dynamics in complex networks

Dynamical processes exhibiting absorbing states are essential in the modeling of a large variety of situations from material science to epidemiology and social sciences. Such processes exhibit the possibility of avalanching behavior upon slow driving. Here, we study the distribution of sizes and durations of avalanches for well-known dynamical processes on complex networks. We find that all analyzed models display a similar critical behavior, characterized by the presence of two distinct regimes. At small scales, sizes and durations of avalanches exhibit distributions that are dependent on the network topology and the model dynamics. At asymptotically large scales instead -- irrespective of the type of dynamics and of the topology of the underlying network -- sizes and durations of avalanches are characterized by power-law distributions with the exponents of the standard mean-field critical branching process.

physics.soc-ph↗

arXiv2020

Intermittent percolation and the scale-free distribution of vegetation clusters

Understanding the causes and effects of spatial vegetation patterns is a fundamental problem in ecology, especially because these can be used as early predictors of catastrophic shifts such as desertification processes. Empirical studies of the vegetation cover in some areas such as drylands and semiarid regions have revealed the existence of vegetation patches of broadly diverse sizes. In particular, the probability distribution of patch sizes can be fitted by a power law, i.e. vegetation patches are approximately scale free up to some maximum size. Different explanatory mechanisms, such as plant-plant interactions and plant-water feedback loops have been proposed to rationalize the emergence of such scale-free patterns, yet a full understanding has not been reached. Using a simple model for vegetation dynamics, we show that environmental temporal variability -- a well-recognized feature of semiarid environments -- promotes in a robust way (i.e. for a wide range of parameter values) the emergence of vegetation patches with broadly distributed cluster sizes. Furthermore, this result is related to a percolation phenomenon that occurs in an intermittent or fluctuating way. The model also reveals that the power-law exponents fitting the tails of the probability distributions depend on the overall vegetation-cover density, in agreement with empirical observations. This supports the idea that environmental variability plays a key role in the formation of scale-free vegetation patterns. From a practical viewpoint, this may be of importance to predict the effects that changes in environmental conditions may have in real ecosystems. From a theoretical side, our study sheds new light on a novel type of percolation phenomena occurring under temporally-varying external conditions, that still needs further work to be fully characterized.

cond-mat.stat-mech↗

arXiv2020

Feedback mechanisms for self-organization to the edge of a phase transition

Scale-free outbursts of activity are commonly observed in physical, geological, and biological systems. The idea of self-organized criticality (SOC), introduced back in 1987 by Bak, Tang and Wiesenfeld suggests that, under certain circumstances, natural systems can seemingly self-tune to a critical state with its concomitant power-laws and scaling. Theoretical progress allowed for a rationalization of how SOC works by relating its critical properties to those of a standard non-equilibrium second-order phase transition that separates an active state in which dynamical activity reverberates indefinitely, from an absorbing or quiescent state where activity eventually ceases. Here, we briefly review these ideas as well as a recent closely-related concept: self-organized bistability (SOB). In SOB, the very same type of feedback operates in a system characterized by a discontinuos phase transition, which has no critical point but instead presents bistability between active and quiescent states. SOB also leads to scale-invariant avalanches of activity but, in this case, with a different type of scaling and coexisting with anomalously large outbursts. Moreover, SOB explains experiments with real sandpiles more closely than SOC. We review similarities and differences between SOC and SOB by presenting and analyzing them under a common theoretical framework, covering recent results as well as possible future developments. We also discuss other related concepts for "imperfect" self-organization such as "self-organized quasi-criticality" and "self-organized collective oscillations", of relevance in e.g. neuroscience, with the aim of providing an overview of feedback mechanisms for self-organization to the edge of a phase transition.

cond-mat.stat-mech↗

arXiv2020

Self-organized bistability and its possible relevance for brain dynamics

Self-organized bistability (SOB) is the counterpart of 'self-organized criticality' (SOC), for systems tuning themselves to the edge of bistability of a discontinuous phase transition, rather than to the critical point of a continuous one. The equations defining the mathematical theory of SOB turn out to bear strong resemblance to a (Landau-Ginzburg) theory recently proposed to analyze the dynamics of the cerebral cortex. This theory describes the neuronal activity of coupled mesoscopic patches of cortex, homeostatically regulated by short-term synaptic plasticity. The theory for cortex dynamics entails, however, some significant differences with respect to SOB, including the lack of a (bulk) conservation law, the absence of a perfect separation of timescales and, the fact that in the former, but not in the second, there is a parameter that controls the overall system state (in blatant contrast with the very idea of self-organization). Here, we scrutinize --by employing a combination of analytical and computational tools-- the analogies and differences between both theories and explore whether in some limit SOB can play an important role to explain the emergence of scale-invariant neuronal avalanches observed empirically in the cortex. We conclude that, actually, in the limit of infinitely slow synaptic-dynamics, the two theories become identical, but the timescales required for the self-organization mechanism to be effective do not seem to be biologically plausible. We discuss the key differences between self-organization mechanisms with/without conservation and with/without infinitely separated timescales. In particular, we introduce the concept of 'self-organized collective oscillations' and scrutinize the implications of our findings in neuroscience, shedding new light into the problems of scale invariance and oscillations in cortical dynamics.

cond-mat.stat-mech↗

arXiv2020

Evolution in the Debian GNU/Linux software network: analogies and differences with gene regulatory networks

Biological networks exhibit intricate architectures deemed to be crucial for their functionality. In particular, gene regulatory networks, which play a key role in information processing in the cell, display non-trivial architectural features such as scale-free degree distributions, high modularity, and low average distance between connected genes. Such networks result from complex evolutionary and adaptive processes difficult to track down empirically. On the other hand, there exists detailed information on the developmental (or evolutionary) stages of open-software networks that result from self-organized growth across versions. Here, we study the evolution of the Debian GNU/Linux software network, focusing on the changes of key structural and statistical features over time. Our results show that evolution has led to a network structure in which the out-degree distribution is scale-free and the in-degree distribution is a stretched exponential. In addition, while modularity, directionality of information flow, and average distance between elements grew, vulnerability decreased over time. These features resemble closely those currently shown by gene regulatory networks, suggesting the existence of common adaptive pathways for the architectural design of information-processing networks. Differences in other hierarchical aspects point to system-specific solutions to similar evolutionary challenges.

physics.soc-ph↗

arXiv2019

Time-series thresholding and the definition of avalanche size

Avalanches whose sizes and durations are distributed as power laws appear in many contexts. Here, we show that there is a hidden peril in thresholding continuous times series --either from empirical or synthetic data-- for the detection of avalanches. In particular, we consider two possible alternative definitions of avalanche size used e.g. in the empirical determination of avalanche exponents in the analysis of neural-activity data. By performing analytical and computational studies of an Ornstein-Uhlenbeck process (taken as a guiding example) we show that if (i) relatively large threshold values are employed to determine the beginning and ending of avalanches and if (ii) --as sometimes done in the literature-- avalanche sizes are defined as the total area (above zero) of the avalanche, then true asymptotic scaling behavior is not seen, instead the observations are dominated by transient effects. These can induce misinterpretations of the resulting scaling regimes as well as to a wrong assignation of universality classes.

cond-mat.dis-nn↗

arXiv2019

Burstiness in activity-driven networks and the epidemic threshold

We study the effect of heterogeneous temporal activations on epidemic spreading in temporal networks. We focus on the susceptible-infected-susceptible (SIS) model on activity-driven networks with burstiness. By using an activity-based mean-field approach, we derive a closed analytical form for the epidemic threshold for arbitrary activity and inter-event time distributions. We show that, as expected, burstiness lowers the epidemic threshold while its effect on prevalence is twofold. In low-infective systems burstiness raises the average infection probability, while it weakens epidemic spreading for high infectivity. Our results can help clarify the conflicting effects of burstiness reported in the literature. We also discuss the scaling properties at the transition, showing that they are not affected by burstiness.

physics.soc-ph↗

arXiv2019

Jensen's force and the statistical mechanics of cortical asynchronous states

The cortex exhibits self-sustained highly-irregular activity even under resting conditions, whose origin and function need to be fully understood. It is believed that this can be described as an "asynchronous state" stemming from the balance between excitation and inhibition, with important consequences for information-processing, though a competing hypothesis claims it stems from critical dynamics. By analyzing a parsimonious neural-network model with excitatory and inhibitory interactions, we elucidate a noise-induced mechanism called "Jensen's force" responsible for the emergence of a novel phase of arbitrarily-low but self-sustained activity, which reproduces all the experimental features of asynchronous states. The simplicity of our framework allows for a deep understanding of asynchronous states from a broad statistical-mechanics perspective and of the phase transitions to other standard phases it exhibits, opening the door to reconcile, asynchronous-state and critical-state hypotheses. We argue that Jensen's forces are measurable experimentally and might be relevant in contexts beyond neuroscience.

cond-mat.dis-nn↗

arXiv2018

Eco-evolutionary Model of Rapid Phenotypic Diversification in Species-Rich Communities

Evolutionary and ecosystem dynamics are often treated as different processes --operating at separate timescales-- even if evidence reveals that rapid evolutionary changes can feed back into ecological interactions. A recent long-term field experiment has explicitly shown that communities of competing plant species can experience very fast phenotypic diversification, and that this gives rise to enhanced complementarity in resource exploitation and to enlarged ecosystem-level productivity. Here, we build on progress made in recent years in the integration of eco-evolutionary dynamics, and present a computational approach aimed at describing these empirical findings in detail. In particular we model a community of organisms of different but similar species evolving in time through mechanisms of birth, competition, sexual reproduction, descent with modification, and death. Based on simple rules, this model provides a rationalization for the emergence of rapid phenotypic diversification in species-rich communities. Furthermore, it also leads to non-trivial predictions about long-term phenotypic change and ecological interactions. Our results illustrate that the presence of highly specialized, non-competing species leads to very stable communities and reveals that phenotypically equivalent species occupying the same niche may emerge and coexist for very long times. Thus, the framework presented here provides a simple approach --complementing existing theories, but specifically devised to account for the specificities of the recent empirical findings for plant communities-- to explain the collective emergence of diversification at a community level, and paves the way to further scrutinize the intimate entanglement of ecological and evolutionary processes, especially in species-rich communities.

q-bio.PE↗