Representation, search, and the evolution of routines in problem solving

Abstract

This article discusses the emergence and adaptive dynamics of routines in problem solving by formalizing and analyzing into details a simple and widely studied problem, the Tower of Hanoi. In particular, we show how different cognitive representations and different measures of performance modify the landscape on which the search for routines takes place and therefore the set and properties of the possible outcomes of this search. Then we discuss the case of hierarchical representations and hierarchies of routines and show their advantages in terms of flexibility, evolvability, and scalability.