top of page
Search

WETWARE - AT THE EDGE OF CHAOS


The wetware in your head, aka your brain, is slow. So very slow. The information transfer between neurons has a typical latency of 10-20ms. Scientists are baffled at how we’ve been able to solve time-critical computational problems to ensure our survival. We are surprisingly good at solving problems. Many times better than much faster silicon-based computers. So how does the brain overcome the limitations of speed for information transfer across space-time?

A recent research paper with the cryptic title, Complex harmonics reveal low-dimensional manifolds of critical brain dynamics, suggests that this is because the brain operates like a quantum system. That the efficiency stems from the brain’s ability to operate at a ‘critical state,’ where activity balances between order and chaos, maximizing adaptability and information flow.

The study builds on previous research suggesting that the brain’s structure is optimized for distributed computation. Unlike conventional computing, which follows sequential steps, the brain processes information through a network of interactions, making use of rare long-range connections. These connections allow different parts of the brain to work together efficiently, overcoming the limitations of slower neural signals.

This means that the brain is functioning as a non-local system, where different regions synchronise in a way reminiscent of quantum coherence — where particles remain connected regardless of distance.

This research is in early stages, but it does go some way to explain how the brain can make such complex calculations, even though the latency is so high. It is a parallel processing unit where the different processes communicate with each other unrestricted from the direct connectivity between neurons.

This could also be an interesting path to explore for AI research. Could this lead to, for example, more adaptable and efficient algorithms for processing large-scale data?


References:



 
 
 

Comments


© 2024 by Mikael Svanström
bottom of page