I continue to try to understand how a human brain works and how we can use it to set up AI which would copy the behavior of a actual person. A few days ago I uploaded a question where I described a process of function of the brain of such a AI. However I have changed my mind on some things.What if let's say we have a correlation function and a neural network. My new idea is that first the correlation function will look for patterns between the input and some stored data or memories or logic. Then if the correlation coefficient is greater than some number let's say 0.7 then the neural network will process those data and do what a neural network does. This will ensure that we would have a dynamic pattern recognition process and that the AI could still learn from new experiences, although it would be able to understand the relevance of a input with a understood pattern(memory or logic). Is there such a mode in Machine Learning?