>So what do the quantum superpositions add over mapping neural networks and simulating them? https://www.damninteresting.com/on-the-origin-of-circuits/ https:/

>So what do the quantum superpositions add over mapping neural networks and simulating them? https://www.damninteresting.com/on-the-origin-of-circuits/ https://www.damninteresting.com/on-the-origin-of-circuits/ When I read your comment I immediately recalled this article about an experiment that designed tone discrimination circuits on FPGAs by stochastic processes (genetic algorithm). The highlight is that at the end of thousands of rounds of controlled evolution, the researchers obtained a circuit that achieved unexpected efficiency through mechanisms they initially failed to explain. There were closed circuits within the overall structure which only interacted with the main structure passively. Its design was the product of its unique electro-magnetic environment to the extent that results could not be replicated on another board of the same type. In other words, a circuit appeared to originate with properties that would not have been arrived at by applying the conventional theory of circuit design. Considering that organic brains are made up of many trillions of unit neurons, which grow into networks that will be trained by stochastic processes, it makes sense to expect them to exhibit a similar level of exploitation of their environment. If that environment is one in which quantum phenomena can result in real effects on the behavior of the system, why shouldn't we expect them to be exploited by real brains?