Abstract:Neural assemblies representing different engrams compete for successful retrieval in the CA3 region of the hippocampus, yet the detailed mechanisms underlying their formation remain elusive. Recent research indicates that hippocampal inhibitory neurons respond selectively to stimuli and exhibit diverse plasticity, suggesting their significant role in engram formation. Conventional attractor network models for CA3 typically employ global inhibition, where inhibitory neurons uniformly suppress the activity of excitatory neurons. However, these models may not fully capture the complex dynamics of competition arising from sparse distributed coding and may not accurately reflect the specific roles of inhibitory neurons in the competition between neural assemblies during memory retrieval. We propose a mechanism for engram formation in CA3 using a spiking neural network model, emphasizing the critical role of the association between excitatory and inhibitory neurons through heterosynaptic plasticity. In our model, inhibitory neurons are associated with specific neural assemblies during encoding and selectively inhibit excitatory neurons involved in competing assemblies during retrieval. With a simplified dentate gyrus (DG) in a feed-forward structure, this proposed mechanism results in sparsely distributed engrams in CA3. The sparse distributed coding in the model allows us to investigate the effects of selective inhibition on pattern completion under various configurations, such as partially overlapping competing engrams. Our results demonstrate that selective inhibition provides more stable pattern completion and enhances retrieval performance compared to global inhibition alone. Furthermore, the observed neural activity in the hippocampal subregions of the model aligns with experimental findings on these regions' roles in pattern separation and pattern completion.

Fine Discrimination of Analog Patterns by Nonlinear Dendritic Inhibition

A dendritic disinhibitory circuit mechanism for pathway-specific gating

Dynamics Analysis and Analog Associative Memory of Networks with LT Neurons.

Analysis of Cyclic Dynamics for Networks of Linear Threshold Neurons.

Oscillation-Driven Spike-Timing Dependent Plasticity Allows Multiple Overlapping Pattern Recognition in Inhibitory Interneuron Networks

Approximating nonlinear functions with latent boundaries in low-rank excitatory-inhibitory spiking networks

A probabilistic model for learning in cortical microcircuit motifs with data-based divisive inhibition

Impact of dendritic non-linearities on the computational capabilities of neurons

Activity-dependent Neuron Model for Noise Resistance

Selective inhibition in CA3: A mechanism for stable pattern completion through heterosynaptic plasticity

Nonlinear Multiplicative Dendritic Integration in Neuron and Network Models

Structured Synaptic Inhibition Has a Critical Role in Multiple-Choice Motion-Discrimination Tasks

Discriminating neural ensemble patterns through dendritic computations in randomly connected feedforward networks

How do neurons operate on sparse distributed representations? A mathematical theory of sparsity, neurons and active dendrites

Beyond spiking networks: The computational advantages of dendritic amplification and input segregation

Unsupervised learning by a nonlinear network with Hebbian excitatory and anti-Hebbian inhibitory neurons

Inhibition in Random Neuronal Networks Enhances Response Variability and Disrupts Stimulus Discrimination

Few-shot pattern detection by transient boosting of somato-dendritic coupling

A dendritic substrate for temporal diversity of cortical inhibition

Sparse coding with a somato-dendritic rule