PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 11032032

Spike-driven synaptic plasticity: theory, simulation, VLSI implementation.

S Fusi·M Annunziato·D Badoni·A Salamon·D J Amit
View original source ↗Export metadata ↓

Abstract

We present a model for spike-driven dynamics of a plastic synapse, suited for aVLSI implementation. The synaptic device behaves as a capacitor on short timescales and preserves the memory of two stable states (efficacies) on long timescales. The transitions (LTP/LTD) are stochastic because both the number and the distribution of neural spikes in any finite (stimulation) interval fluctuate, even at fixed pre- and postsynaptic spike rates. The dynamics of the single synapse is studied analytically by extending the solution to a classic problem in queuing theory (Takacs process). The model of the synapse is implemented in aVLSI and consists of only 18 transistors. It is also directly simulated. The simulations indicate that LTP/LTD probabilities versus rates are robust to fluctuations of the electronic parameters in a wide range of rates. The solutions for these probabilities are in very good agreement with both the simulations and measurements. Moreover, the probabilities are readily manipulable by variations of the chip's parameters, even in ranges where they are very small. The tests of the electronic device cover the range from spontaneous activity (3-4 Hz) to stimulus-driven rates (50 Hz). Low transition probabilities can be maintained in all ranges, even though the intrinsic time constants of the device are short (approximately 100 ms). Synaptic transitions are triggered by elevated presynaptic rates: for low presynaptic rates, there are essentially no transitions. The synaptic device can preserve its memory for years in the absence of stimulation. Stochasticity of learning is a result of the variability of interspike intervals; noise is a feature of the distributed dynamics of the network. The fact that the synapse is binary on long timescales solves the stability problem of synaptic efficacies in the absence of stimulation. Yet stochastic learning theory ensures that it does not affect the collective behavior of the network, if the transition probabilities are low and LTP is balanced against LTD.

Explore related subjects

Action PotentialsArtificial IntelligenceComputer SimulationComputersElectric ConductivityLong-Term PotentiationModels, NeurologicalNeural InhibitionNeural Networks, ComputerNeuronal PlasticityProbabilityStochastic ProcessesSynapsesJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

S Fusi, M Annunziato, D Badoni, A Salamon, D J Amit. 2000. Spike-driven synaptic plasticity: theory, simulation, VLSI implementation.. https://doi.org/10.1162/089976600300014917

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2001

Thalamo-cortical connections of areas 3a and M1 in marmoset monkeys.

The present investigation is part of a broader effort to examine cortical areas that contribute to manual dexterity, reaching, and grasping. In this study we examine the thalamic connections of electrophysiologically defined regions in area 3a and architectonically defined primary motor cortex (M1). Our studies demonstrate that area 3a receives input from nuclei associated with the somatosensory system: the superior, inferior, and lateral divisions of the ventral posterior complex (VPs, VPi, and VPl, respectively). Surprisingly, area 3a receives the majority of its input from thalamic nuclei associated with the motor system, posterior division of the ventral lateral nucleus of the thalamus (VL), the mediodorsal nucleus (MD), and intralaminar nuclei including the central lateral nucleus (CL) and the centre median nucleus (CM). In addition, sparse but consistent projections to area 3a are from the anterior pulvinar (Pla). Projections from the thalamus to the cortex immediately rostral to area 3a, in the architectonically defined M1, are predominantly from VL, VA, CL, and MD. There is a conspicuous absence of inputs from the nuclei associated with processing somatic inputs (VP complex). Our results indicate that area 3a is much like a motor area, in part because of its substantial connections with motor nuclei of the thalamus and motor areas of the neocortex (Huffman et al. [2000] Soc. Neurosci. Abstr. 25:1116). The indirect input from the cerebellum and basal ganglia via the ventral lateral nucleus of the thalamus supports its role in proprioception. Furthermore, the presence of input from somatosensory thalamic nuclei suggests that it plays an important role in somatosensory and motor integration.

Action Potentials↗
PubMed2001

Spontaneous remission of paroxysmal dystonia coincides with normalization of entopeduncular activity in dt(SZ) mutants.

Recent studies have shown a dramatically decreased spontaneous discharge rate of entopeduncular neurons in a unique animal model of idiopathic paroxysmal dystonia, the dt(sz) mutant hamster. These changes were found in animals at the age at which the most marked expression of dystonia is usually observed. In this rodent model, the age-dependent disappearance of stress-inducible dystonic attacks at an age of approximately 10 weeks allows investigations of the relevance of pathophysiological changes for the occurrence of dystonia by ontogenetic studies. Therefore, we examined the entopeduncular activity by extracellular single unit recordings in groups of dt(sz) mutants and nondystonic control hamsters at 17-22 weeks of age. In contrast to recent findings, after the complete remission of dystonia, the mean discharge rate of entopeduncular neurons in dt(sz) mutants (28.1 +/- 1.2 spikes/sec) was similar to that of age-matched nondystonic control hamsters (30.8 +/- 0.9 spikes/sec). Thus, the disappearance of paroxysmal dystonia is accompanied by a normalization of the entopeduncular activity in dt(sz) mutants. The present data clearly demonstrate the fundamental importance of a decreased basal ganglia output for the expression of paroxysmal dystonia.

Action Potentials↗
PubMed2001

The effects of lead on transient outward currents of acutely dissociated rat dorsal root ganglia.

The effects of Pb2+ on transient outward currents (TOCs) were investigated on rat dorsal root ganglia (DRG) neurons at postnatal days of 15 approximately 21, using the conventional whole-cell patch-clamp technique. In media-sized (35 approximately 40 microm) neurons and in the presence of 50 mM TEA, TOCs that preliminarly included an A-current (IA) and a D-current (ID), were clearly present and dominant. Application of Pb2+ lengthened the initial delay of TOCs and increased the onset-peak time in a concentration-dependent manner. The amplitudes of initial outward current peak were reduced with increasing Pb2+ concentrations. The inhibitory effects of Pb2+ on TOCs were reversible with 80 approximately 90% of current reversed in 2 approximately 10 min at 1 approximately 400 microM Pb2+. For the normalized activation curves fitted by a single Boltzmann equation under each condition, there was a shift to more depolarized voltages with increasing concentrations of Pb2+. The V1/2 and the slope factor (k) increased from 12.76+/-1.49 mV and 15.31+/-1.66 mV (n=10) under control condition to 39.91+/-5.44 mV (n=10, P<0.01) and 21.39+/-3.13 mV (n=10, P<0.05) at 400 microM Pb2+, respectively, indicating that Pb2+ decreased the activation of TOCs. For the normalized steady-state inactivation curves, the V1/2 and the k increased from -92.31+/-2.72 and 8.59+/-1.36 mV (n=10) to -55.65+/-3.67 (n=10, P<0.01) and 23.02+/-2.98 mV (n=10, P<0.01) at 400 microM Pb2+, respectively. The curves were shifted to more depolarized voltages by Pb2+, indicating that channels were less likely to be inactivated at higher concentrations of Pb2+ at any given potential. The fast (tf) and slow (ts) decay time-constants were both significantly increased by increasing concentrations of Pb2+ (n=10, P<0.05), indicating that Pb2+ increased the decay time-course of TOCs. These effects were concentration-dependent and partly reversible following washing. Ca2+ modulated the TOCs gating and might share same binding site with Pb2+, for which Ca2+ had very low affinity. In summary, the results demonstrated that Pb2+ was a dose- and voltage-dependent, and reversible blocker of TOCs in rat DRG neurons. After Pb2+ application, normal sensory physiology of DRG neurons was affected, and these neurons might display aberrant firing properties that resulted in abnormal sensations. This variation caused by Pb2+ could underlie the toxical modulation of sensory input to the central nervous system.

Action Potentials↗