Hebbian learning reconsidered: Representation of static and dynamic objects in associative neural nets