This book is essentially a much extended and modified version of Theory of Problem Solving. An Approach to Artificial Intelligence (New York : Elsevier, 1969. XIV-189 p.). – The purpose of this book is twofold : first, it proposes to develop one specific theory of problem solving and learning on which research on Artificial Intelligence (AI) can be based; second, it wants to describe some actual or proposed computer programs based on this theory that exhibit problem solving behaviour. – Chapter 1, A theory of problem solving; – 2, Problems or one-person games; – 3, Two person games; – 4, Automatization of heuristic discovery : representation and learning; – 5, Learning in a flexible description language. – Appendix : A1, Basic definitions; A2, Problems and graphs; A3, Games; A4, Heuristic discovery; A5, Learning.