Dynamics and computation in mixed networks containing neurons that accelerate towards spiking

Manz, Paul; Goedeke, Sven; Memmesheimer, Raoul-Martin

doi:10.1103/physreve.100.042404

Cited by 3 publications

(5 citation statements)

References 70 publications

(155 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We explicitly model excitatory neurons. Since the irregular activity in biological neural networks is likely due to a balanced state, where excitatory and inhibitory inputs to each neuron balance [28,[32][33][34][35][36], our model implicitly incorporates inhibitory neurons, see Fig 1.…”

Section: Poisson Neuronsmentioning

confidence: 99%

Purely STDP-based assembly dynamics: Stability, learning, overlaps, drift and aging

Manz

Memmesheimer

2023

PLoS Comput Biol

Self Cite

View full text Add to dashboard Cite

Memories may be encoded in the brain via strongly interconnected groups of neurons, called assemblies. The concept of Hebbian plasticity suggests that these assemblies are generated through synaptic plasticity, strengthening the recurrent connections within select groups of neurons that receive correlated stimulation. To remain stable in absence of such stimulation, the assemblies need to be self-reinforcing under the plasticity rule. Previous models of such assembly maintenance require additional mechanisms of fast homeostatic plasticity often with biologically implausible timescales. Here we provide a model of neuronal assembly generation and maintenance purely based on spike-timing-dependent plasticity (STDP) between excitatory neurons. It uses irregularly and stochastically spiking neurons and STDP that depresses connections of uncorrelated neurons. We find that assemblies do not grow beyond a certain size, because temporally imprecisely correlated spikes dominate the plasticity in large assemblies. Assemblies in the model can be learned or spontaneously emerge. The model allows for prominent, stable overlap structures between static assemblies. Further, assemblies can drift, particularly according to a novel, transient overlap-based mechanism. Finally the model indicates that assemblies grow in the aging brain, where connectivity decreases.

show abstract

Section: Poisson Neuronsmentioning

confidence: 99%

Purely STDP-based assembly dynamics: Stability, learning, overlaps, drift and aging

Manz

Memmesheimer

2023

PLoS Comput Biol

Self Cite

View full text Add to dashboard Cite

show abstract

“…[38]) or include discontinuities, delays or having the shape of a δ-function (e.g. [5,36,51]). Even though pulse-coupling is very commonly used in network models in theoretical neuroscience a systematic understanding from a dynamical systems perspective is missing [5,38].…”

Section: Pulse-coupling With and Without Delay As A Neural Featurementioning

confidence: 99%

“…(e.g. [36,51,59]. However, Jahnke et al [36] as well as Manz et al [51] demonstrated that the irregular dynamics arising in large balanced state networks may be chaotic or not, depending on the dynamical properties of the network units, i.e.…”

Section: The Role Of Chaos In Neurosciencementioning

confidence: 99%

“…[36,51,59]. However, Jahnke et al [36] as well as Manz et al [51] demonstrated that the irregular dynamics arising in large balanced state networks may be chaotic or not, depending on the dynamical properties of the network units, i.e. whether the coupling is excitatory or inhibitory [36] or whether the neuron models used are leaky integrators (with positive dissipation) or anti-leaky integrators (with negative dissipation) [51].…”

Section: The Role Of Chaos In Neurosciencementioning

confidence: 99%

“…through neuro-imagine techniques such as fMRI or EEG, the mesoscopic scale concerning the neural network level is still relatively difficult to access [69,78]. In theoretical neuroscience researchers often create complex models consisting of larger networks which produce a broad range of complex dynamical behavior [12,36,51,59,76,88,89]. The neural units of these networks typically involve features that are specific to biological and especially neuronal systems and not common to systems of classical physics.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Complex Dynamics Enabled by Basic Neural Features

Regel¹

View full text Add to dashboard Cite

What is a Good Model?A good model is simple. A good model, just as in every science, is a model that is complex enough to produce the targeted empirical observations while being as simple as possible. Already Occam in the fourteenth century [9] proposed the idea of choosing the simplest of all models which can explain the behavior. His idea was that the simplest model with the least amount of assumptions necessary to explain something is more likely to be true. The more assumptions and complexity is added, the more likely it is that one adds a wrong assumption, which renders the model wrong.There is another reason for choosing the simplest model: In a trivial sense the best model to explain something is the object itself -the brain is the best model for the brain. However, even if somebody manages to read out every detail of a brain, already a daunting task, and produces a gigantic computer simulation which produces Thesis StructureIn Chapter 2 the theoretical background is introduced and important concepts are defined. In Chapter 3, 4 and 5 we present three systems that show complex behavior arising from basic neural features. Chapter 3 concerns a two dimensional adaptive system captured by two ordinary differential equations. A proof for the numerically observed phase space structure of nested limit cycles is presented for a simplified version of the system. The system in chapter 4 consists of symmetrically all-to-all pulse-coupled phase oscillators. The transition to order conservation by introducing self-loops and the reordering process of the oscillators is studied analytically. Chapter

show abstract

Purely STDP-based assembly dynamics: stability, learning, overlaps, drift and aging

Manz

Memmesheimer

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Memories may be encoded in the brain via strongly interconnected groups of neurons, called assemblies. The concept of Hebbian plasticity suggests that these assemblies are generated through synaptic plasticity, strengthening the recurrent connections within select groups of neurons that receive correlated stimulation. To remain stable in absence of such stimulation, the assemblies need to be self-reinforcing under the plasticity rule. Previous models of such assembly maintenance require additional mechanisms of fast homeostatic plasticity often with biologically implausible timescales. Here we provide a model of neuronal assembly generation and maintenance purely based on spike-timing-dependent plasticity (STDP) between excitatory neurons. It uses irregularly and stochastically spiking neurons and STDP that depresses connections of uncorrelated neurons. We find that assemblies do not grow beyond a certain size, because temporally imprecise spike correlations dominate the plasticity in large assemblies. Assemblies in the model can be learned or spontaneously emerge. The model allows for prominent, stable overlap structures between static assemblies. Further, assemblies can drift, particularly according to a novel, transient overlap-based mechanism. Finally the model indicates that assemblies grow in the aging brain, where connectivity decreases.

show abstract

Dynamics and computation in mixed networks containing neurons that accelerate towards spiking

Cited by 3 publications

References 70 publications

Purely STDP-based assembly dynamics: Stability, learning, overlaps, drift and aging

Purely STDP-based assembly dynamics: Stability, learning, overlaps, drift and aging

Complex Dynamics Enabled by Basic Neural Features

Purely STDP-based assembly dynamics: stability, learning, overlaps, drift and aging

Contact Info

Product

Resources

About