Avalanche Patterns Research Articles

Characterization of the neuronal and network dynamics of liquid state machines

Reservoir computing (RC) is a relatively new machine-learning framework that uses an abstract neural network model, called reservoir. The reservoir forms a complex system with high dimensionality, nonlinearity, and intrinsic memory effect due to recurrent connections among individual neurons. RC manifests a best-in-class performance in processing information generated by complex dynamical systems, yet little is known about its microscopic/macroscopic dynamics underlying the computational capability. Here, we characterize the neuronal and network dynamics of liquid state machines (LSMs) using numerical simulations and Modified National Institute of Standards and Technology (MNIST) database classification tasks. The computational performance of LSMs largely depends on a dynamic range of neuronal avalanches whereby the avalanche patterns are determined by the neuron and network models. A larger dynamic range leads to higher performance—the MNIST classification accuracy is highest when the avalanche sizes follow a slowly decaying power-law distribution with an exponent of ∼1.5, followed by the power-law statistics with a larger exponent and the mixture of power-law/log-normal distributions. Network-theoretic analysis suggests that the formation of large functional clusters and the promotion of dynamic transitions between large and small clusters may contribute to the scale-invariant nature. This study provides new insight into our understanding of the computational principles of RC concerning the actions of the individual neurons and the system-level collective behavior.

(034) Are All Orgasms Created Equally Regardless of Age?

Abstract Introduction Female orgasm has been described as “a sense of throbbing once a second, due to rhythmic contraction and relaxation of the pelvic floor muscles…”1. Physiologic measurement of pelvic floor muscle (PFM) activity has been reported utilized electromyography2 as well as rectal and vaginal pressure biofeedback. 3 None have reported on the specificity of PFM variability during arousal and orgasm or variability across the life cycle. Likewise, there’s been no correlation between these physiologic responses to the subjective nature of orgasm. While knowledge of neuroendocrinology and neuroimaging contribute to some understanding of the physiology of the sexual cycle, research on variability and correlation of physiologic responses of PFMs to the subjectivity of orgasm across is lacking. Objective The purpose is to objectively measure, compare, and correlate the following physiologic responses: a) time to orgasm; b) peak orgasmic force; and c) length of orgasm and correlate those findings with subjective orgasm characteristics utilizing the Orgasmic Rating Scale (ORS) (report to follow). Methods Objective data recorded by in-home users of Lioness (SmartBod Incorporated), an intravaginal, pressure biofeedback device capable of recording PFM activity in response to clitoral vibration. Arousal and orgasmic activity were recorded to users’ phones and a secure SmartBod server. Users were identified by a unique device number with no personal identifiers. Additional demographic data and ORS responses from 26 consenting users were collected and a secondary analysis of the data was performed by the first author. One-hundred-sixty-two biofeedback tracings were included in the review. Data was grouped according to age range, orgasmic type (volcano, avalanche, wave), and by length of time to orgasm regardless of age. Results Ages ranged from 21-60. 17/26 were nulliparous; 5 were peri- or post-menopausal. Review of age grouped data averages showed those 31-40 took the longest to orgasm (598sec) but produced the longest orgasmic response (30sec). Those 51-60 took the least time, 219sec, however created the greatest orgasmic force of 1425gm/f. When comparing averages of the 3 orgasmic types, volcano and avalanche patterns were equally dispersed among all groups with the 51-60s patterning only seen as wave. Those identified as having an avalanche pattern took the longest achieve an orgasm (30sec) but produced the longest orgasmic response (30sec). Those with the wave pattern produced the greatest peak force at 940.8gm/f. The median time to orgasm was 483 sec (8min). Those below the median in time (less time to orgasm) took 227sec (4min) on average to orgasm with peak forces of 783gm/f. Those above the median, on average took 789sec (13min) to reach orgasm when a peak orgasmic force of 722gm/ was produced. Conclusions Orgasmic patterns from this small sample of women suggest that those in the 31-40 age range have the longest orgasms and need the most arousal to get there; however, those in the 51-60 year range produced the great peak orgasmic force. Those with the avalanche patterned orgasm take the longest to get there as but have the longest orgasm. Taking less arousal time overall appears to create the strongest orgasmic response. This likely indicates that women in this condition are more sexually aroused to begin with. Disclosure Any of the authors act as a consultant, employee or shareholder of an industry for: SmartBod.

Magnitude‐Frequency Distributions and Slip‐History Predictions for Earthquakes Using Cellular Automata and Absorbing Markov Chains

AbstractCellular automata have proven effective in obtaining statistical insights into expected time series, magnitude‐frequency distributions, and average slip histories of earthquakes by confirming, for instance, the Gutenberg‐Richter magnitude‐frequency distribution and the existence of scaling functions for slip histories. Yet, exhaustive modeling is often required to obtain such insights since the model behavior is generally difficult to predict from fixed input parameters, such as the dissipation and long‐range stress interaction distance. We demonstrate that the temporal dynamics of a cellular automaton (CA), representing discretized equations of motion, can be simplified and modeled as an absorbing Markov chain with transition matrices that are fully determined by CA parameters. Time series, frequency‐size distributions, and slip histories of the Markov chain Monte Carlo (MCMC) and CA models are stochastically equivalent. The proposed method is a mean‐field approximation that replicates temporal CA statistics by ignoring spatial components. Fundamentally, the temporal portion of CA can be represented as a memoryless process in which the current outcome only depends on the immediate past. We believe the transparency of the statistical model may provide pertinent insights into the mean‐field behavior of a variety of physical applications near a critical state, including earthquake and avalanche patterns. For instance, the average slip histories display a typical but asymmetric shape due to a preferred path through probability space with initial acceleration of slip rate to peak size followed by slower deceleration toward rupture arrest.

A regional spatiotemporal analysis of large magnitude snow avalanches using tree rings

Abstract. Snow avalanches affect transportation corridors and settlements worldwide. In many mountainous regions, robust records of avalanche frequency and magnitude are sparse or non-existent. However, dendrochronological methods can be used to fill this gap and infer historical avalanche patterns. In this study, we developed a tree-ring-based avalanche chronology for large magnitude avalanche events (size ≥∼D3) using dendrochronological techniques for a portion of the US northern Rocky Mountains. We used a strategic sampling design to examine avalanche activity through time and across nested spatial scales (i.e., from individual paths, four distinct subregions, and the region). We analyzed 673 samples in total from 647 suitable trees collected from 12 avalanche paths from which 2134 growth disturbances were identified over the years 1636 to 2017 CE. Using existing indexing approaches, we developed a regional avalanche activity index to discriminate avalanche events from noise in the tree-ring record. Large magnitude avalanches, common across the region, occurred in 30 individual years and exhibited a median return interval of approximately 3 years (mean = 5.21 years). The median large magnitude avalanche return interval (3–8 years) and the total number of avalanche years (12–18) varies throughout the four subregions, suggesting the important influence of local terrain and weather factors. We tested subsampling routines for regional representation, finding that sampling 8 random paths out of a total of 12 avalanche paths in the region captures up to 83 % of the regional chronology, whereas four paths capture only 43 % to 73 %. The greatest value probability of detection for any given path in our dataset is 40 %, suggesting that sampling a single path would capture no more than 40 % of the regional avalanche activity. Results emphasize the importance of sample size, scale, and spatial extent when attempting to derive a regional large magnitude avalanche event chronology from tree-ring records.

Selective Participation of Single Cortical Neurons in Neuronal Avalanches.

Neuronal avalanches are scale-invariant neuronal population activity patterns in the cortex that emerge in vivo in the awake state and in vitro during balanced excitation and inhibition. Theory and experiments suggest that avalanches indicate a state of cortex that improves numerous aspects of information processing by allowing for the transient and selective formation of local as well as system-wide spanning neuronal groups. If avalanches are indeed involved with information processing, one might expect that single neurons would participate in avalanche patterns selectively. Alternatively, all neurons could participate proportionally to their own activity in each avalanche as would be expected for a population rate code. Distinguishing these hypotheses, however, has been difficult as robust avalanche analysis requires technically challenging measures of their intricate organization in space and time at the population level, while also recording sub- or suprathreshold activity from individual neurons with high temporal resolution. Here, we identify repeated avalanches in the ongoing local field potential (LFP) measured with high-density microelectrode arrays in the cortex of awake nonhuman primates and in acute cortex slices from young and adult rats. We studied extracellular unit firing in vivo and intracellular responses of pyramidal neurons in vitro. We found that single neurons participate selectively in specific LFP-based avalanche patterns. Furthermore, we show in vitro that manipulating the balance of excitation and inhibition abolishes this selectivity. Our results support the view that avalanches represent the selective, scale-invariant formation of neuronal groups in line with the idea of Hebbian cell assemblies underlying cortical information processing.

The investigation of avalanche patterns on railroad tracks with steep slopes

This study aims to investigate the pattern of landslides on the cliffs forming the railway with steep slopes. This research was conducted at Rejosari Natar Station, South Lampung Regency, Lampung Province. The point of observation is at STA 30 + 250. The data in this study were divided into primary and secondary data. Primary data is the data that is directly obtained in the field by taking soil samples which are then processed in the laboratory. The secondary data is the soil characteristics data of the railroad structure starting from the rail load, rail bearings, ballast layers, sub-ballast layers, and sub-grade. These will be used as input in conducting railroad analysis using Plaxis modelling. Plaxis modeling requires these variables and parameters to be analyzed based on the amount of vertical settlement, total stress, and safety factor values that occur due to rail loads. Reserach found that the value of the soil safety factor was <1.25. This means that the soil condition is in a critical condition for collapse and the ground will experience landslides which can endanger the safety of the passing trains. For this reason, improvements are needed in the soil that will experience landslides

Numerical simulation of flux avalanches in type-II superconducting thin films under transient AC magnetic fields

Flux avalanches induced from thermomagnetic instability are crucial challenges for the application of superconducting thin film devices. In this paper, flux avalanches in a type-II superconducting thin film exposed to a transient AC magnetic field are numerically simulated by solving the coupled nonlinear Maxwell’s equations and the heat diffusion equation based on the fast Fourier transform (FFT) method. The dependence of the threshold magnetic field on the ambient temperature, film thickness and magnetic field ramp rate are obtained through these numerical simulations, which show good agreement with experimental results. A linear increase in the threshold field as the film thickness increases and a nonmonotonic increase in the threshold field as the ambient temperature increases have also been found. Our numerical results demonstrate that the threshold field decreases exponentially as the ramp rate increases. Flux avalanche patterns observed in magneto-optical imaging (MOI) measurements for a film exposed to an AC magnetic field are reproduced. The hysteresis magnetization and maximum temperature jump curves are also illustrated. We find that fingering instability plays an important role in the thermomagnetic response of superconducting thin films under transient AC magnetic fields, especially for high magnetic field ramp rates.

Avalanching patterns of irregular sand particles in continual discrete flow**Project supported by the National Natural Science Foundation of China (Grant Nos. 11572201, 91634202, and 11902190).

We investigate the flow patterns of irregular sand particles under avalanching mode in a rotating drum by using the spatial filtering velocimetry technique. By exploring the variations of velocity distribution of granular flow, we find a type of avalanching pattern of irregular sand particles which is similar to that of spherical particles flow. Due to the fact that the initial position of avalanche in this pattern locates at the middle of the drum and the avalanche propagates toward the edge area gradually, we named it as mid-to-edge avalanching pattern. Furthermore, we find another avalanching pattern which slumps from the edge and propagates toward the opposite edge of the flow surface, named as edge-to-edge pattern. By analyzing the temporal and spatial characteristics of these two types of avalanching patterns, we discover that these two types of avalanche patterns are caused by that the avalanching particles constantly perturb the axial adjacent particles. Thus, the particles on the flow surface are involved in avalanching sequentially in order of the axial distance from the initial position.

Self-Organization of Genome Expression from Embryo to Terminal Cell Fate: Single-Cell Statistical Mechanics of Biological Regulation.

A statistical mechanical mean-field approach to the temporal development of biological regulation provides a phenomenological, but basic description of the dynamical behavior of genome expression in terms of autonomous self-organization with a critical transition (Self-Organized Criticality: SOC). This approach reveals the basis of self-regulation/organization of genome expression, where the extreme complexity of living matter precludes any strict mechanistic approach. The self-organization in SOC involves two critical behaviors: scaling-divergent behavior (genome avalanche) and sandpile-type critical behavior. Genome avalanche patterns—competition between order (scaling) and disorder (divergence) reflect the opposite sequence of events characterizing the self-organization process in embryo development and helper T17 terminal cell differentiation, respectively. On the other hand, the temporal development of sandpile-type criticality (the degree of SOC control) in mouse embryo suggests the existence of an SOC control landscape with a critical transition state (i.e., the erasure of zygote-state criticality). This indicates that a phase transition of the mouse genome before and after reprogramming (immediately after the late 2-cell state) occurs through a dynamical change in a control parameter. This result provides a quantitative open-thermodynamic appreciation of the still largely qualitative notion of the epigenetic landscape. Our results suggest: (i) the existence of coherent waves of condensation/de-condensation in chromatin, which are transmitted across regions of different gene-expression levels along the genome; and (ii) essentially the same critical dynamics we observed for cell-differentiation processes exist in overall RNA expression during embryo development, which is particularly relevant because it gives further proof of SOC control of overall expression as a universal feature.

Numerical simulation on the flux avalanche behaviors of microstructured superconducting thin films

Controlling and suppressing the propagation of magnetic flux avalanches is an important issue for the application of type-II superconductors. The effects of engineered pinning centers (antidots) on the guidance of flux avalanche propagation paths in type-II superconducting thin films are numerically investigated by solving the coupled nonlinear Maxwell's equations and the thermal diffusion equations. The field dependence of critical current density is considered in the simulation in this paper. Dynamic propagations of the thermomagnetic avalanches within the superconducting films patterned with different arrangements of antidots (like random, periodic square, and conformal mapping arrays) are presented. We reveal that presence of the antidots significantly modifies the propagation paths of the avalanches. The flux avalanche patterns of the superconducting films change with the variation of the arrangements of antidots. The patterned antidots in the form of conformal mapping arrays within the superconducting film exhibit strong guidance to the thermomagnetic avalanches. In addition, introducing the antidots in the form of conformal mapping arrays into the superconducting film can effectively lower the magnetic flux jump sizes.

Influences of non-uniformities and anisotropies on the flux avalanche behaviors of type-II superconducting films

In this paper, the anisotropic flux avalanche processes in thin square-shaped type-II superconducting films are numerically investigated by solving the coupled nonlinear Maxwell’s equations and the thermal diffusion equations. Influences of the non-uniformities and intrinsic critical current density anisotropies originate from the manufacturing process are considered in the simulation. In addition, we also studied the effect of the extrinsic anisotropy induced by the in-plane magnetic field. The results demonstrate that the non-uniformities and anisotropies of the critical current density play significant roles in the flux avalanche process of the thin film superconductors. Slight anisotropy (either intrinsic or extrinsic) can dramatically change the propagation direction of avalanches in the superconducting film, which is consistent with the experimental results. Simulations on the thin square-shaped isotropic superconducting films show that the threshold magnetic field for the flux avalanches increases with the angle between the applied field and the superconducting film-plane. In addition, the flux avalanche patterns change with the angular variation of the in-plane component of external magnetic field. When the in-plane magnetic field component is along the diagonal lines of the superconducting square, symmetric flux avalanche penetration patterns occur to the film.

Synaptic plasticity and neuronal refractory time cause scaling behaviour of neuronal avalanches.

Neuronal avalanches measured in vitro and in vivo in different cortical networks consistently exhibit power law behaviour for the size and duration distributions with exponents typical for a mean field self-organized branching process. These exponents are also recovered in neuronal network simulations implementing various neuronal dynamics on different network topologies. They can therefore be considered a very robust feature of spontaneous neuronal activity. Interestingly, this scaling behaviour is also observed on regular lattices in finite dimensions, which raises the question about the origin of the mean field behavior observed experimentally. In this study we provide an answer to this open question by investigating the effect of activity dependent plasticity in combination with the neuronal refractory time in a neuronal network. Results show that the refractory time hinders backward avalanches forcing a directed propagation. Hebbian plastic adaptation plays the role of sculpting these directed avalanche patterns into the topology of the network slowly changing it into a branched structure where loops are marginal.

Forecasting wet-snow avalanche probability in mountainous terrain

Water percolating through the snow cover can lead to wet-snow instability as well as snowmelt runoff. The accurate prediction of spatial patterns of wet-snow in mountainous terrain therefore has practical applications in both back-country avalanche forecasting and hydrology. Recent research has shown that incident radiation plays a dominant role during the first complete wetting of the snow cover. We therefore investigated if large-scale meteorological forecast data, corrected for subgrid topographic influences on the shortwave radiation balance, together with subgrid mean slopes, can be combined to predict large-scale wet-snow avalanche patterns. Required surface albedo was derived from parameterized snow-covered fraction based on terrain parameters and measured flat field snow depths. We derived avalanche probability density functions for daily mean air temperature and incoming shortwave radiation from detailed observations over six winters using time-lapse photography. Based on probability density functions of these meteorological parameters and of slope angles of previous avalanches, we computed wet-snow probability maps for the entire Swiss Alps at a 2.5km resolution. The probability maps compared well with observed wet-snow avalanche activity patterns. Even though the approach cannot forecast the onset of a cycle, since this would require snow cover related parameters, it provides a new approach toward an automatic spatial avalanche forecast built upon simple terrain parameters and easy to obtain large-scale meteorological surface variables. The advantage of our method is that it does not require running small-scale models with demanding model input parameters.

Viscoelastic effects in avalanche dynamics: a key to earthquake statistics.

In many complex systems a continuous input of energy over time can be suddenly relaxed in the form of avalanches. Conventional avalanche models disregard the possibility of internal dynamical effects in the interavalanche periods, and thus miss basic features observed in some real systems. We address this issue by studying a model with viscoelastic relaxation, showing how coherent oscillations of the stress field can emerge spontaneously. Remarkably, these oscillations generate avalanche patterns that are similar to those observed in seismic phenomena.

The diversity of flux avalanche patterns in superconducting films

The variety of morphologies in flux patterns created by thermomagnetic dendritic avalanches in type-II superconducting films is investigated using numerical simulations. The avalanches are triggered by introducing a hot spot at the edge of a strip-shaped sample, which is initially prepared in a partially penetrated Bean critical state by slowly ramping the transversely applied magnetic field. The simulation scheme is based on a model accounting for the nonlinear and nonlocal electrodynamics of superconductors in the transverse geometry. By systematically varying the parameters representing the Joule heating, heat conduction in the film, and heat transfer to the substrate, a wide variety of avalanche patterns are formed, and quantitative characterizations of the areal extension, branch width etc are made. The results show that branching is suppressed by the lateral heat diffusion, while large Joule heating gives many branches, and heat removal into the substrate limits the areal size. The morphology shows significant dependence also on the initial flux penetration depth.

The effects of El Niño and La Niña on snow and avalanche patterns in British Columbia, Canada, and central Chile

AbstractEl Niño and La Niña affect global climate and atmospheric circulation to determine winter temperature and precipitation patterns. Both winter temperatures and the associated precipitation patterns have effects on mountain snow deposition and snow avalanche occurrences. Approximately 25 000 slab avalanches from 30 winters were analyzed in relation to snowfall patterns contrasted for El Niño and La Niña winters for two avalanche areas with different snow climates in British Columbia (BC), Canada. La Niña winters were shown to produce more snow, more avalanches and a higher percentage of dry avalanches than wet avalanches. The data and analysis show that the avalanche patterns depend on the altitude and snow climate. Analysis of snowfall and accident data from the Andes of Chile suggests behavior opposite to BC. El Niño winters in central Chile produced the most snow and, by inference, the most avalanches. This paper is the first to show the links of El Niño and La Niña to snow avalanche activity.

Reconstructing temporal patterns of snow avalanches at Lago del Desierto,southern Patagonian Andes

Snow avalanches recurrently cause substantial damage to infrastructure and losses in human lives in mountainous environments around the world. Densely populated regions have suffered more intensely the destructive forces of avalanches; however, settlers of these regions have developed tools to mitigate avalanche risk, including avalanche hazard maps. To design detailed hazard maps, accurate information on avalanche runout and frequency is necessary. In areas where avalanches are mostly undocumented, like in the Andes, dendrochronological methods become a valuable tool for reconstructing spatio-temporal avalanche patterns. At Lago del Desierto (southern Patagonian Andes, Argentina), trees from nine avalanche tracks were sampled. Nothofagus pumilio, a winter deciduous broad-leaved tree, was sampled in all cases. Avalanche activity is recognized based on the presence of scars, changes in stem eccentricity, reaction wood and abrupt growth changes. A rating system was used to define the relative importance of each tree-ring indicator in the reconstructed chronology of events. According to this chronology, widespread avalanche activity on the slope was found for the years 1918, 1930, 1931, 1971, 1995 and 1998. Superposed Epoch Analysis indicated that total monthly precipitation during the three snowiest months from May to October of 1971, 1995 and 1998 (years for which climatic data exists for validation) was significantly greater than for years without large avalanche activity. In contrast, no significant correlations were found between monthly temperature variations and years with large avalanche activity on the slope. Atmospheric circulation patterns associated with years of major avalanche activity show features typically observed during the cold phase of the El Niño–Southern Oscillation cycle, i.e. La Niña. For those cases, results also show an increase in the westerlies across the South Pacific and southern South America, resulting in both higher precipitation and stronger winds over the study area that favor snow avalanche episodes. We point out the need of a more systematic sampling strategy in order to reconstruct spatial avalanche patterns in the southern Andes, which are a complement of temporal patterns to design hazard maps.

Neuronal avalanches are diverse and precise activity patterns that are stable for many hours in cortical slice cultures.

A major goal of neuroscience is to elucidate mechanisms of cortical information processing and storage. Previous work from our laboratory (Beggs and Plenz, 2003) revealed that propagation of local field potentials (LFPs) in cortical circuits could be described by the same equations that govern avalanches. Whereas modeling studies suggested that these "neuronal avalanches" were optimal for information transmission, it was not clear what role they could play in information storage. Work from numerous other laboratories has shown that cortical structures can generate reproducible spatiotemporal patterns of activity that could be used as a substrate for memory. Here, we show that although neuronal avalanches lasted only a few milliseconds, their spatiotemporal patterns were also stable and significantly repeatable even many hours later. To investigate these issues, we cultured coronal slices of rat cortex for 4 weeks on 60-channel microelectrode arrays and recorded spontaneous extracellular LFPs continuously for 10 hr. Using correlation-based clustering and a global contrast function, we found that each cortical culture spontaneously produced 4736 +/- 2769 (mean +/- SD) neuronal avalanches per hour that clustered into 30 +/- 14 statistically significant families of spatiotemporal patterns. In 10 hr of recording, over 98% of the mutual information shared by these avalanche patterns were retained. Additionally, jittering analysis revealed that the correlations between avalanches were temporally precise to within +/-4 msec. The long-term stability, diversity, and temporal precision of these avalanches indicate that they fulfill many of the requirements expected of a substrate for memory and suggest that they play a central role in both information transmission and storage within cortical networks.