Molar vs Molecular Approaches to Reward Schedule and Serial Learning Phenomena