Modulation of working memory duration by synaptic and astrocytic mechanisms

Area of Science:

• Neuroscience
• Computational Neuroscience
• Astrogliology

Background:

• Short-term synaptic plasticity and presynaptic vesicle release are crucial for working memory (WM) models.
• Astrocytes are increasingly implicated in higher cognitive functions, including WM, via synaptic transmission modulation.

Purpose of the Study:

• To investigate the impact of astrocytic signaling on the stability and duration of WM representations.
• To model astrocytic regulation of presynaptic release probability in synaptic attractor models of WM.

Main Methods:

• Simulated WM protocols in firing rate and spiking neural networks with and without astrocytic regulation.
• Analyzed phase space dynamics in rate networks to understand astrocytic influence.
• Investigated the interplay between astrocytic signaling, noise, and WM representation stability.

Main Results:

• Astrocytic signaling significantly alters the duration and stability of WM representations.
• A 'window of vulnerability' emerges with strong astrocytic regulation and slow presynaptic timescales, making WM susceptible to noise.
• Identified mechanisms by which network noise can either stabilize or destabilize WM representations.

Conclusions:

• Astrocytic regulation is a key determinant of WM representation duration in synaptic attractor models.
• Astrocytic signaling may enable top-down control mechanisms for volitional modulation of WM representations and their duration.