Growth strategy determines the memory and structural properties of brain networks

Area of Science:

• Computational neuroscience
• Complex systems

Background:

• The relationship between structure and function influences emergent properties in natural systems.
• Brain synaptic density exhibits specific temporal profiles during development.

Purpose of the Study:

• To investigate the role of synaptic connectivity dynamics in neural network development and function.
• To model the temporal profiles of synaptic density in the brain using an adaptive neural network.
• To understand how network structure influences memory recovery and energy efficiency.

Main Methods:

• Developed an adaptive neural network model coupling activity and topological dynamics.
• Simulated network evolution under noisy conditions.
• Analyzed the impact of transient synaptic connectivity on memory recovery.
• Evaluated energy consumption across different synaptic densities.

Main Results:

• A transient phase of high synaptic connectivity is essential for memory recovery in networks exposed to noise.
• Intermediate synaptic densities lead to optimal developmental paths with reduced energy consumption.
• Transient heterogeneity within the network dictates its developmental trajectory.
• The model reproduces experimentally observed temporal profiles of synaptic density in the brain.

Conclusions:

• Transient synaptic heterogeneity is a key factor in neural network development and function.
• These findings explain characteristic pruning curves in brain areas.
• The results offer insights for designing biologically inspired neural networks with enhanced information processing capabilities.