Early lock-in of structured and specialised information flows during neural development

Area of Science:

• Neuroscience
• Computational Neuroscience

Background:

• Mature brains exhibit complex information processing, but its developmental emergence is poorly understood.
• Lack of effective estimators for spiking data has hindered developmental studies of information flow.

Purpose of the Study:

• To quantify changes in information transfer entropy during neural development.
• To investigate the emergence and spatial structure of information flow in developing neural networks.

Main Methods:

• Utilized advanced estimators to quantify transfer entropy in developing dissociated neural cell cultures.
• Analyzed spontaneous neural activity and population bursts.
• Compared findings with a model network using spike-timing-dependent plasticity.

Main Results:

• Information flow across developing networks significantly increases.
• Spatial structure of information flow tends to stabilize as it emerges.
• During population bursts, neural nodes adopt specialized roles (transmitter, mediator, receiver) aligned with spike ordering.
• These roles lock-in as information flows are established.

Conclusions:

• Neural development involves a substantial increase in information processing capacity.
• Emerging information flow patterns exhibit structural stabilization and functional specialization.
• Observed developmental patterns in information flow are consistent with plasticity-based network models.