Simple spontaneously active Hebbian learning model: homeostasis of activity and connectivity, and consequences for learning and epileptogenesis

Putative roles for homeostatic plasticity in epileptogenesis

Homeostatic plasticity allows neural circuits to maintain an average activity level while preserving the ability to learn new associations and efficiently transmit information. This dynamic process usually protects the brain from excessive activity, like seizures. However, in certain contexts, homeostatic plasticity might produce seizures, either in response to an acute provocation or more chronically as a driver of epileptogenesis. Here, we review three seizure conditions in which homeostatic plasticity likely plays an important role: acute drug withdrawal seizures, posttraumatic or disconnection epilepsy, and cyclic seizures. Identifying the homeostatic mechanisms active at different stages of development and in different circuits could allow better targeting of therapies, including determining when neuromodulation might be most effective, proposing ways to prevent epileptogenesis, and determining how to disrupt the cycle of recurring seizure clusters.

Synaptic and intrinsic homeostasis cooperate to optimize single neuron response properties and tune integrator circuits

Homeostatic processes that provide negative feedback to regulate neuronal firing rate are essential for normal brain function, and observations suggest that multiple such processes may operate simultaneously in the same network. We pose two questions: why might a diversity of homeostatic pathways be necessary, and how can they operate in concert without opposing and undermining each other? To address these questions, we perform a computational and analytical study of cell-intrinsic homeostasis and synaptic homeostasis in single-neuron and recurrent circuit models. We demonstrate analytically and in simulation that when two such mechanisms are controlled on a long time scale by firing rate via simple and general feedback rules, they can robustly operate in tandem to tune the mean and variance of single neuron's firing rate to desired goals. This property allows the system to recover desired behavior after chronic changes in input statistics. We illustrate the power of this homeostatic tuning scheme by using it to regain high mutual information between neuronal input and output after major changes in input statistics. We then show that such dual homeostasis can be applied to tune the behavior of a neural integrator, a system that is notoriously sensitive to variation in parameters. These results are robust to variation in goals and model parameters. We argue that a set of homeostatic processes that appear to redundantly regulate mean firing rate may work together to control firing rate mean and variance and thus maintain performance in a parameter-sensitive task such as integration.

Using network dynamic fMRI for detection of epileptogenic foci

Background

Epilepsy is one of the most prevalent neurological disorders. It remains medically intractable for about one-third of patients with focal epilepsy, for whom precise localization of the epileptogenic zone responsible for seizure initiation may be critical for successful surgery. Existing fMRI literature points to widespread network disturbances in functional connectivity. Per previous scalp and intracranial EEG studies and consistent with excessive local synchronization during interictal discharges, we hypothesized that, relative to same regions in healthy controls, epileptogenic foci would exhibit less chaotic dynamics, identifiable via entropic analyses of resting state fMRI time series.

Methods

In order to first validate this hypothesis on a cohort of patients with known ground truth, here we test individuals with well-defined epileptogenic foci (left mesial temporal lobe epilepsy). We analyzed voxel-wise resting-state fMRI time-series using the autocorrelation function (ACF), an entropic measure of regulation and feedback, and performed follow-up seed-to-voxel functional connectivity analysis. Disruptions in connectivity of the region exhibiting abnormal dynamics were examined in relation to duration of epilepsy and patients’ cognitive performance using a delayed verbal memory recall task.

Results

ACF analysis revealed constrained (less chaotic) functional dynamics in left temporal lobe epilepsy patients, primarily localized to ipsilateral temporal pole, proximal to presumed focal points. Autocorrelation decay rates differentiated, with 100 % accuracy, between patients and healthy controls on a subject-by-subject basis within a leave-one-subject out classification framework. Regions identified via ACF analysis formed a less efficient network in patients, as compared to controls. Constrained dynamics were linked with locally increased and long-range decreased connectivity that, in turn, correlated significantly with impaired memory (local left temporal connectivity) and epilepsy duration (left temporal – posterior cingulate cortex connectivity).

Conclusions

Our current results suggest that data driven functional MRI methods that target network dynamics hold promise in providing clinically valuable tools for identification of epileptic regions.

Post and pre-compensatory Hebbian learning for categorisation

A system with some degree of biological plausibility is developed to categorise items from a widely used machine learning benchmark. The system uses fatiguing leaky integrate and fire neurons, a relatively coarse point model that roughly duplicates biological spiking properties; this allows spontaneous firing based on hypo-fatigue so that neurons not directly stimulated by the environment may be included in the circuit. A novel compensatory Hebbian learning algorithm is used that considers the total synaptic weight coming into a neuron. The network is unsupervised and entirely self-organising. This is relatively effective as a machine learning algorithm, categorising with just neurons, and the performance is comparable with a Kohonen map. However the learning algorithm is not stable, and behaviour decays as length of training increases. Variables including learning rate, inhibition and topology are explored leading to stable systems driven by the environment. The model is thus a reasonable next step toward a full neural memory model.

Neuronal avalanches, epileptic quakes and other transient forms of neurodynamics

Power-law behaviors in brain activity in healthy animals, in the form of neuronal avalanches, potentially benefit the computational activities of the brain, including information storage, transmission and processing. In contrast, power-law behaviors associated with seizures, in the form of epileptic quakes, potentially interfere with the brain's computational activities. This review draws attention to the potential roles played by homeostatic mechanisms and multistable time-delayed recurrent inhibitory loops in the generation of power-law phenomena. Moreover, it is suggested that distinctions between health and disease are scale-dependent. In other words, what is abnormal and defines disease it is not the propagation of neural activity but the propagation of activity in a neural population that is large enough to interfere with the normal activities of the brain. From this point of view, epilepsy is a disease that results from a failure of mechanisms, possibly located in part in the cortex itself or in the deep brain nuclei and brainstem, which truncate or otherwise confine the spatiotemporal scales of these power-law phenomena.

Reduction of seizure occurrence from exposure to auditory stimulation in individuals with neurological handicaps: a randomized controlled trial

BACKGROUND

The purpose of this work was to determine in a clinical trial the efficacy of reducing or preventing seizures in patients with neurological handicaps through sustained cortical activation evoked by passive exposure to a specific auditory stimulus (particular music). The specific type of stimulation had been determined in previous studies to evoke anti-epileptiform/anti-seizure brain activity.

METHODS

The study was conducted at the Thad E. Saleeby Center in Harstville, South Carolina, which is a permanent residence for individuals with heterogeneous neurological impairments, many with epilepsy. We investigated the ability to reduce or prevent seizures in subjects through cortical stimulation from sustained passive nightly exposure to a specific auditory stimulus (music) in a three-year randomized controlled study. In year 1, baseline seizure rates were established. In year 2, subjects were randomly assigned to treatment and control groups. Treatment group subjects were exposed during sleeping hours to specific music at regular intervals. Control subjects received no music exposure and were maintained on regular anti-seizure medication. In year 3, music treatment was terminated and seizure rates followed. We found a significant treatment effect (p = 0.024) during the treatment phase persisting through the follow-up phase (p = 0.002). Subjects exposed to treatment exhibited a significant 24% decrease in seizures during the treatment phase, and a 33% decrease persisting through the follow-up phase. Twenty-four percent of treatment subjects exhibited a complete absence of seizures during treatment.

CONCLUSION/SIGNIFICANCE

Exposure to specific auditory stimuli (i.e. music) can significantly reduce seizures in subjects with a range of epilepsy and seizure types, in some cases achieving a complete cessation of seizures. These results are consistent with previous work showing reductions in epileptiform activity from particular music exposure and offers potential for achieving a non-invasive, non-pharmacologic treatment of epilepsy.

TRIAL REGISTRATION

Clinicaltrials.gov NCT01459692.

Being critical of criticality in the brain

Relatively recent work has reported that networks of neurons can produce avalanches of activity whose sizes follow a power law distribution. This suggests that these networks may be operating near a critical point, poised between a phase where activity rapidly dies out and a phase where activity is amplified over time. The hypothesis that the electrical activity of neural networks in the brain is critical is potentially important, as many simulations suggest that information processing functions would be optimized at the critical point. This hypothesis, however, is still controversial. Here we will explain the concept of criticality and review the substantial objections to the criticality hypothesis raised by skeptics. Points and counter points are presented in dialog form.

Failure of adaptive self-organized criticality during epileptic seizure attacks

Critical dynamics are assumed to be an attractive mode for normal brain functioning as information processing and computational capabilities are found to be optimal in the critical state. Recent experimental observations of neuronal activity patterns following power-law distributions, a hallmark of systems at a critical state, have led to the hypothesis that human brain dynamics could be poised at a phase transition between ordered and disordered activity. A so far unresolved question concerns the medical significance of critical brain activity and how it relates to pathological conditions. Using data from invasive electroencephalogram recordings from humans we show that during epileptic seizure attacks neuronal activity patterns deviate from the normally observed power-law distribution characterizing critical dynamics. The comparison of these observations to results from a computational model exhibiting self-organized criticality (SOC) based on adaptive networks allows further insights into the underlying dynamics. Together these results suggest that brain dynamics deviates from criticality during seizures caused by the failure of adaptive SOC.

Aberrant neuronal avalanches in cortical tissue removed from juvenile epilepsy patients

Some forms of epilepsy may arise as a result of pathologic interactions among neurons. Many forms of collective activity have been identified, including waves, spirals, oscillations, synchrony, and neuronal avalanches. All these emergent activity patterns have been hypothesized to show pathologic signatures associated with epilepsy. Here, the authors used 60-channel multielectrode arrays to record neuronal avalanches in cortical tissue removed from juvenile epilepsy patients. For comparison, they also recorded activity in rat cortical slices. The authors found that some human tissue removed from epilepsy patients exhibited prolonged periods of hyperactivity not seen in rat slices. In addition, they found a positive correlation between the branching parameter, a measure of network gain, and firing rate in human slices during periods of hyperactivity. This relationship was not present in rat slices. The authors suggest that this positive correlation between the branching parameter and the firing rate is part of a positive feedback loop and may contribute to some forms of epilepsy. These results also indicate that neuronal avalanches are abnormally regulated in slices removed from pediatric epilepsy patients.

A few strong connections: optimizing information retention in neuronal avalanches

BACKGROUND

How living neural networks retain information is still incompletely understood. Two prominent ideas on this topic have developed in parallel, but have remained somewhat unconnected. The first of these, the "synaptic hypothesis," holds that information can be retained in synaptic connection strengths, or weights, between neurons. Recent work inspired by statistical mechanics has suggested that networks will retain the most information when their weights are distributed in a skewed manner, with many weak weights and only a few strong ones. The second of these ideas is that information can be represented by stable activity patterns. Multineuron recordings have shown that sequences of neural activity distributed over many neurons are repeated above chance levels when animals perform well-learned tasks. Although these two ideas are compelling, no one to our knowledge has yet linked the predicted optimum distribution of weights to stable activity patterns actually observed in living neural networks.

RESULTS

Here, we explore this link by comparing stable activity patterns from cortical slice networks recorded with multielectrode arrays to stable patterns produced by a model with a tunable weight distribution. This model was previously shown to capture central features of the dynamics in these slice networks, including neuronal avalanche cascades. We find that when the model weight distribution is appropriately skewed, it correctly matches the distribution of repeating patterns observed in the data. In addition, this same distribution of weights maximizes the capacity of the network model to retain stable activity patterns. Thus, the distribution that best fits the data is also the distribution that maximizes the number of stable patterns.

CONCLUSIONS

We conclude that local cortical networks are very likely to use a highly skewed weight distribution to optimize information retention, as predicted by theory. Fixed distributions impose constraints on learning, however. The network must have mechanisms for preserving the overall weight distribution while allowing individual connection strengths to change with learning.

Advances in the application of technology to epilepsy: the CIMIT/NIO Epilepsy Innovation Summit

In 2008, a group of clinicians, scientists, engineers, and industry representatives met to discuss advances in the application of engineering technologies to the diagnosis and treatment of patients with epilepsy. The presentations also provided a guide for further technological development, specifically in the evaluation of patients for epilepsy surgery, seizure onset detection and seizure prediction, intracranial treatment systems, and extracranial treatment systems. This article summarizes the discussions and demonstrates that cross-disciplinary interactions can catalyze collaborations between physicians and engineers to address and solve many of the pressing unmet needs in epilepsy.

Zwanzig-Mori projection operators and EEG dynamics: deriving a simple equation of motion

We present a macroscopic theory of electroencephalogram (EEG) dynamics based on the laws of motion that govern atomic and molecular motion. The theory is an application of Zwanzig-Mori projection operators. The result is a simple equation of motion that has the form of a generalized Langevin equation (GLE), which requires knowledge only of macroscopic properties. The macroscopic properties can be extracted from experimental data by one of two possible variational principles. These variational principles are our principal contribution to the formalism. Potential applications are discussed, including applications to the theory of critical phenomena in the brain, Granger causality and Kalman filters.

PACS code: 87.19.lj

An open hypothesis: is epilepsy learned, and can it be unlearned?

Plasticity is central to the ability of a neural system to learn and also to its ability to develop spontaneous seizures. What is the connection between the two? Learning itself is known to be a destabilizing process at the algorithmic level. We have investigated necessary constraints on a spontaneously active Hebbian learning system and find that the ability to learn appears to confer an intrinsic vulnerability to epileptogenesis on that system. We hypothesize that epilepsy arises as an abnormal learned response of such a system to certain repeated provocations. This response is a network-level effect. If epilepsy really is a learned response, then it should be possible to reverse it, that is, to unlearn epilepsy. Unlearning epilepsy may then provide a new approach to its treatment.

Self-organization and neuronal avalanches in networks of dissociated cortical neurons

Dissociated cortical neurons from rat embryos cultured onto micro-electrode arrays exhibit characteristic patterns of electrophysiological activity, ranging from isolated spikes in the first days of development to highly synchronized bursts after 3-4 weeks in vitro. In this work we analyzed these features by considering the approach proposed by the self-organized criticality theory: we found that networks of dissociated cortical neurons also generate spontaneous events of spreading activity, previously observed in cortical slices, in the form of neuronal avalanches. Choosing an appropriate time scale of observation to detect such neuronal avalanches, we studied the dynamics by considering the spontaneous activity during acute recordings in mature cultures and following the development of the network. We observed different behaviors, i.e. sub-critical, critical or super-critical distributions of avalanche sizes and durations, depending on both the age and the development of cultures. In order to clarify this variability, neuronal avalanches were correlated with other statistical parameters describing the global activity of the network. Criticality was found in correspondence to medium synchronization among bursts and high ratio between bursting and spiking activity. Then, the action of specific drugs affecting global bursting dynamics (i.e. acetylcholine and bicuculline) was investigated to confirm the correlation between criticality and regulated balance between synchronization and variability in the bursting activity. Finally, a computational model of neuronal network was developed in order to interpret the experimental results and understand which parameters (e.g. connectivity, excitability) influence the distribution of avalanches. In summary, cortical neurons preserve their capability to self-organize in an effective network even when dissociated and cultured in vitro. The distribution of avalanche features seems to be critical in those cultures displaying medium synchronization among bursts and poor random spiking activity, as confirmed by chemical manipulation experiments and modeling studies.