I'm a novice here..but in virtual neural networks, does adding more virtual neurons improve the performance of the network in terms of recognizing patterns? Has this been experimentally verified anywhere? Can you then make an analogy that the same would happen in a biological neural network? Or is the virtual neural network too different to a biological one? For that matter, has anybody added appropriate neuron types in appropriate regions of an adult mice brain and observed how the mice now handled complex tasks such as navigation throught a maze or something? For that matter, has anybody done an experiment where they deliberately added extra neurons to various regions in a mice's brain and observed whether this improved the mice's ability to navigate through a maze or something like that?