Structured stabilization in recurrent neural circuits through inhibitory synaptic plasticity
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Inhibitory interneurons play a dual role in recurrently connected biological circuits: they regulate global neural activity to prevent runaway excitation, and contribute to diverse neural computations. While the first role can be achieved through unstructured connections tuned for homeostatic rate stabilization, computational tasks often require structured excitatory-inhibitory (E/I) connectivity. Here, we consider a broad class of pairwise inhibitory spike-timing dependent plasticity (iSTDP) rules, demonstrating how inhibitory synapses can self-organize to both stabilize excitation and generate functionally relevant connectivity structures—a process we call “structured stabilization”. We show that in both E/I circuit motifs and large spiking recurrent neural networks the choice of iSTDP rule can lead to either mutually connected E/I pairs, or to lateral inhibition, where an inhibitory neuron connects to an excitatory neuron that does not directly connect back to it. In a one-dimensional ring network with two inhibitory subpopulations following these distinct iSTDP rules, the effective connectivity among the excitatory units self-organizes into a Mexican-hat-like profile, with excitatory influence in the center and inhibitory influence away from the center. This leads to emergent dynamical properties such as contextual modulation effects and modular spontaneous activity. Our theoretical work introduces a family of rules that retains the broad applicability and simplicity of spike-timing-based plasticity, while promoting structured, self-organized stabilization. These findings highlight the rich interplay between iSTDP rules, circuit structure, and neuronal dynamics, offering a framework for understanding how inhibitory plasticity shapes network function.
Neural circuits in the cortex must achieve two things at once: keep activity stable and support flexible computations during development and learning. Most theory has treated changes at inhibitory synapses mainly as a homeostatic brake that keeps excitation in check. Here we show that timing-based rules at inhibitory synapses in recurrent circuits can do much more. In our models, these rules both stabilize excitatory activity and naturally give rise to distinct patterns of excitatory–inhibitory connections, a process we call “structured stabilization”. In recurrent networks with local excitatory connectivity, the resulting structure supports context-dependent responses similar to those observed in adult sensory cortex and patterned activity with long-range correlations as observed in developing cortex.
