Abstract:Brain's connectivity reveals modular and hierarchical structures at various scales. This organization is typically believed to support the coexistence of segregation (specialization) and integration (binding) of information. Motivated by developmental processes, some authors have studied the self-organization of neural networks into modular hierarchies mediated by adaptive mechanism under spontaneous neural activity. Following evidence that the sensory cortices organize into assemblies under selective stimuli, other authors have shown that stable neural assemblies can emerge in random neural networks due to targeted stimulation, embedding various forms of synaptic plasticity in presence of homeostatic and/or control mechanisms. Here, we show that simple spike-timing-dependent plasticity (STDP) rules, based only on pre- and post-synaptic spike times, can also lead to the stable encoding of memories in the absence of any control mechanism. We develop a model of spiking neurons, trained to stimuli targeting different sub-populations. The model is intended to satisfy biologically plausible features: (i) it contains excitatory and inhibitory neurons with Hebbian and anti-Hebbian STDP; (ii) neither the neuronal activity nor the synaptic weights are frozen after the learning phase. Instead, the neurons are allowed to fire spontaneously while synaptic plasticity remains active. We find that only the combination of two inhibitory STDP sub-populations allows for the formation of stable modular organization in the network, with each sub-population playing a distinctive role. The Hebbian sub-population controls for the firing activity, while the anti-Hebbian one promotes pattern selectivity. After the learning phase, the network settles into an asynchronous irregular resting-state. This post-learning activity is associated with spontaneous memory recalls, which turn fundamental for the long-term consolidation of the learned memories. Due to its simplicity, the model, here introduced, can represent a test-bed for further investigations on the role played by STDP on memory storing and maintenance.

Hierarchical Self-Programming in Recurrent Neural Networks

Self-excitation of Neurons Leads to Multiperiodicity of Discrete-Time Neural Networks with Distributed Delays

A Hierarchical Organized Memory Model Using Spiking Neurons

Motor primitive and sequence self-organization in a hierarchical recurrent neural network

Self-organization of multi-layer spiking neural networks

Stability switches, oscillatory multistability, and spatio-temporal patterns of nonlinear oscillations in recurrently delay coupled neural networks

Incorporating structural plasticity into self-organization recurrent networks for sequence learning

Hierarchical Conflict Propagation: Sequence Learning in a Recurrent Deep Neural Network

Self-organization of microcircuits in networks of neurons with plastic synapses

Emergence and maintenance of modularity in neural networks with Hebbian and anti-Hebbian inhibitory STDP

Hierarchical Neural Program Synthesis

Nonlinear Dynamics Analysis of a Self-Organizing Recurrent Neural Network: Chaos Waning

Stimulus-Driven and Spontaneous Dynamics in Excitatory-Inhibitory Recurrent Neural Networks for Sequence Representation

Bistability and Tunable Gating in a Recurrent Neuronal Network with Synaptic Failure and Hidden Neurons

Critical neural networks with short and long term plasticity

Making Neural Programming Architectures Generalize via Recursion

Evolving Self-Assembling Neural Networks: From Spontaneous Activity to Experience-Dependent Learning

Episodic Memory Theory for the Mechanistic Interpretation of Recurrent Neural Networks

Recurrent Network Models Of Sequence Generation And Memory

Self-organized critical neural networks

Identification of Recurrent Dynamics in Distributed Neural Populations