Limits...
Synaptic plasticity enables adaptive self-tuning critical networks.

Stepp N, Plenz D, Srinivasa N - PLoS Comput. Biol. (2015)

Bottom Line: Here, we simulate a 10,000 neuron, deterministic, plastic network of spiking neurons.Brief external perturbations lead to adaptive, long-term modification of intrinsic network connectivity through long-term excitatory plasticity, whereas long-term inhibitory plasticity enables rapid self-tuning of the network back to a critical state.The critical state is characterized by a branching parameter oscillating around unity, a critical exponent close to -3/2 and a long tail distribution of a self-similarity parameter between 0.5 and 1.

View Article: PubMed Central - PubMed

Affiliation: Center for Neural and Emergent Systems, Information and System Sciences Lab, HRL Laboratories LLC, Malibu, California, United States of America.

ABSTRACT
During rest, the mammalian cortex displays spontaneous neural activity. Spiking of single neurons during rest has been described as irregular and asynchronous. In contrast, recent in vivo and in vitro population measures of spontaneous activity, using the LFP, EEG, MEG or fMRI suggest that the default state of the cortex is critical, manifested by spontaneous, scale-invariant, cascades of activity known as neuronal avalanches. Criticality keeps a network poised for optimal information processing, but this view seems to be difficult to reconcile with apparently irregular single neuron spiking. Here, we simulate a 10,000 neuron, deterministic, plastic network of spiking neurons. We show that a combination of short- and long-term synaptic plasticity enables these networks to exhibit criticality in the face of intrinsic, i.e. self-sustained, asynchronous spiking. Brief external perturbations lead to adaptive, long-term modification of intrinsic network connectivity through long-term excitatory plasticity, whereas long-term inhibitory plasticity enables rapid self-tuning of the network back to a critical state. The critical state is characterized by a branching parameter oscillating around unity, a critical exponent close to -3/2 and a long tail distribution of a self-similarity parameter between 0.5 and 1.

Show MeSH

Related in: MedlinePlus

Plot of graph matrices for a subset of neurons at A) 0 s simulation time and B) 300 s simulation time.Each dot represents a synaptic weight on a linear scale such that white and black correspond to 0 and 1, respectively. At 0 s connections are distributed randomly with all weights set to 0.5. After 300 s, excitatory weights have mostly dropped towards 0, with just a few weights increasing towards 1. Inhibitory weights increased towards 1 with only a few connections being reduced. C) The time-series of mean in-degree for excitatory and inhibitory synapses in both pulsed and non-pulsed cases. Note that excitatory in-degree is approximately 80 at time 0 s, but drops dramatically during the first second.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4295840&req=5

pcbi.1004043.g007: Plot of graph matrices for a subset of neurons at A) 0 s simulation time and B) 300 s simulation time.Each dot represents a synaptic weight on a linear scale such that white and black correspond to 0 and 1, respectively. At 0 s connections are distributed randomly with all weights set to 0.5. After 300 s, excitatory weights have mostly dropped towards 0, with just a few weights increasing towards 1. Inhibitory weights increased towards 1 with only a few connections being reduced. C) The time-series of mean in-degree for excitatory and inhibitory synapses in both pulsed and non-pulsed cases. Note that excitatory in-degree is approximately 80 at time 0 s, but drops dramatically during the first second.

Mentions: Speaking more directly to topology, it is possible to define an in-degree, the number of incoming connections to a neuron, using a synaptic weight threshold. In this manner, a connection was considered present if its strength had a value of at least 0.1. Applying this threshold, the unperturbed simulated network began (Fig. 7a) with a mean excitatory in-degree of 80.07 (SD = 8.862) and a mean inhibitory in-degree of 20.06 (SD = 4.471). Since pre and post neurons for all connections were chosen using the same random selection procedure, in- and out-degrees were approximately equal. After 300 s of simulation time (Fig. 7b) the mean in-degree of excitatory synapses dropped significantly to 15.19 (SD = 3.845, t(15998) = 600.7, p < 0.001). Inhibitory in-degree remained largely unchanged at 19.93 (SD = 4.341, t(15998) = 1.901, p = 0.057). Fig. 7c shows the in-degree time-series for both perturbed and unperturbed cases.


Synaptic plasticity enables adaptive self-tuning critical networks.

Stepp N, Plenz D, Srinivasa N - PLoS Comput. Biol. (2015)

Plot of graph matrices for a subset of neurons at A) 0 s simulation time and B) 300 s simulation time.Each dot represents a synaptic weight on a linear scale such that white and black correspond to 0 and 1, respectively. At 0 s connections are distributed randomly with all weights set to 0.5. After 300 s, excitatory weights have mostly dropped towards 0, with just a few weights increasing towards 1. Inhibitory weights increased towards 1 with only a few connections being reduced. C) The time-series of mean in-degree for excitatory and inhibitory synapses in both pulsed and non-pulsed cases. Note that excitatory in-degree is approximately 80 at time 0 s, but drops dramatically during the first second.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4295840&req=5

pcbi.1004043.g007: Plot of graph matrices for a subset of neurons at A) 0 s simulation time and B) 300 s simulation time.Each dot represents a synaptic weight on a linear scale such that white and black correspond to 0 and 1, respectively. At 0 s connections are distributed randomly with all weights set to 0.5. After 300 s, excitatory weights have mostly dropped towards 0, with just a few weights increasing towards 1. Inhibitory weights increased towards 1 with only a few connections being reduced. C) The time-series of mean in-degree for excitatory and inhibitory synapses in both pulsed and non-pulsed cases. Note that excitatory in-degree is approximately 80 at time 0 s, but drops dramatically during the first second.
Mentions: Speaking more directly to topology, it is possible to define an in-degree, the number of incoming connections to a neuron, using a synaptic weight threshold. In this manner, a connection was considered present if its strength had a value of at least 0.1. Applying this threshold, the unperturbed simulated network began (Fig. 7a) with a mean excitatory in-degree of 80.07 (SD = 8.862) and a mean inhibitory in-degree of 20.06 (SD = 4.471). Since pre and post neurons for all connections were chosen using the same random selection procedure, in- and out-degrees were approximately equal. After 300 s of simulation time (Fig. 7b) the mean in-degree of excitatory synapses dropped significantly to 15.19 (SD = 3.845, t(15998) = 600.7, p < 0.001). Inhibitory in-degree remained largely unchanged at 19.93 (SD = 4.341, t(15998) = 1.901, p = 0.057). Fig. 7c shows the in-degree time-series for both perturbed and unperturbed cases.

Bottom Line: Here, we simulate a 10,000 neuron, deterministic, plastic network of spiking neurons.Brief external perturbations lead to adaptive, long-term modification of intrinsic network connectivity through long-term excitatory plasticity, whereas long-term inhibitory plasticity enables rapid self-tuning of the network back to a critical state.The critical state is characterized by a branching parameter oscillating around unity, a critical exponent close to -3/2 and a long tail distribution of a self-similarity parameter between 0.5 and 1.

View Article: PubMed Central - PubMed

Affiliation: Center for Neural and Emergent Systems, Information and System Sciences Lab, HRL Laboratories LLC, Malibu, California, United States of America.

ABSTRACT
During rest, the mammalian cortex displays spontaneous neural activity. Spiking of single neurons during rest has been described as irregular and asynchronous. In contrast, recent in vivo and in vitro population measures of spontaneous activity, using the LFP, EEG, MEG or fMRI suggest that the default state of the cortex is critical, manifested by spontaneous, scale-invariant, cascades of activity known as neuronal avalanches. Criticality keeps a network poised for optimal information processing, but this view seems to be difficult to reconcile with apparently irregular single neuron spiking. Here, we simulate a 10,000 neuron, deterministic, plastic network of spiking neurons. We show that a combination of short- and long-term synaptic plasticity enables these networks to exhibit criticality in the face of intrinsic, i.e. self-sustained, asynchronous spiking. Brief external perturbations lead to adaptive, long-term modification of intrinsic network connectivity through long-term excitatory plasticity, whereas long-term inhibitory plasticity enables rapid self-tuning of the network back to a critical state. The critical state is characterized by a branching parameter oscillating around unity, a critical exponent close to -3/2 and a long tail distribution of a self-similarity parameter between 0.5 and 1.

Show MeSH
Related in: MedlinePlus