The brain is a rather complex organ, capable of, and responsible for, the processing of information, and organised as a network of interacting cells. One of the properties arising from this complexity is memory, which is the ability to learn from experience and to retrieve stored information, affecting decisions and behaviour. Memory is today understood to arise from the adaptation of synapses (i.e. connections between neurons) to the activity of neighbouring cells [1]: this leads to the formation of patterns of neurons, called engrams, which comprise the physical basis of memory [2]. As such, interactions that shape the brain network, resulting in memory as an emergent phenomenon.

Numerous systems and processes outside neuro-science feature a dense matrix of interactions and interconnections as a defining factor. These systems also exhibit the formation of patterns and feature dynamics akin to those of brain memory. This allows for the development of high-level, general models that capture the very fundamental characteristics of brain memory through the broad perspective of network-based dynamical processes.

In this seminar we will present two of such models: a first model featuring a multi-scale, hierarchical memory dynamics [3], and a second model that combines network memory with the coordination of various agents.

REFERENCES

[1] D. Hebb, “The organization of behavior; a neuropsychological theory” (1949).

[2] S. Josselyn et al., Science 367, 6473 (2020).

[3] G. Zanardi et al., Phys. Rev. E 110, 5 (2024).