Neuroscience

Modelling chronic stress as an excitatory-inhibitory perturbation in recurrent working-memory networks

Mauricio A Diaz, Manuela A. Beyer, Janina Hesse

Featured July 1, 2026

This analysis was generated by SciGrove. Upload your own PDFs or enter a DOI — and get the same AI breakdown on any paper.

Get started

AI-generated analysis — This is SciGrove's AI interpretation of the paper, not peer-reviewed content. Always refer to the original paper.

Simply

By simulating stress in brain-like computer networks, researchers found that stronger inhibitory connections cause stress effects, but training under stress makes networks resilient yet less adaptable to new situations.

In depth

The paper models chronic stress in recurrent neural networks, identifying that increased inhibitory-to-excitatory synaptic strength is the primary mechanism reproducing experimental signatures of stress-induced prefrontal dysfunction. They demonstrate that resilience training preserves task performance under stress by maintaining the network's dynamical and energetic state, but this comes at a cost of reduced generalization performance to novel task conditions.

Key Takeaways

1
Chronic stress can be computationally modeled as a selective increase in inhibitory-to-excitatory synaptic strength within recurrent neural networks, accurately reproducing observed biological stress signatures.
2
Networks trained under stress (resilient networks) maintain working memory performance and preserve their baseline dynamical and energetic states, effectively absorbing the perturbation.
3
This stress resilience introduces a delay-generalization trade-off, where resilient networks perform worse on novel, out-of-distribution memory delays compared to naive networks, reflecting a specialization to trained conditions.

Conceptual Flow

HIGH LEVEL

Methodology: How was it done?

Scientists built a simple brain model to test how different types of 'stress' affect its ability to remember things, then trained it to be 'stronger' against stress.

Brain Model

Memory Task

Stress Types

Test & Train

Stress Effects

Resilience

Results: What did they find?

They discovered that one specific type of 'stress' best matched real brain changes, and while training made the model resilient, it also made it less good at new, unfamiliar memory tasks.

Stress Type: Strong Inhibition

Resilience Training

Model Behavior

Memory Preserved

Less Flexible