Gravar-mail: Generalized lessons about sequence learning from the study of the serial reaction time task