This paper introduces the book ``Intrinsically Motivated Learning in Natural and Aritificial Systems", edited by the authors. The book is first in giving an overall coverage of the emerging congitive- science research field of intrinsic motivations. The papers collected in the book have been written by the most important authors who have pioneered the field and are still actively contributing to it. The con- tributions are mainly from modelists and roboticists, but these have wide interdisciplinary interests in understanding brain and behaviour of real organisms, but also from neuroscientists and psychologists. The paper first expands the concept of intrinsic motivations, then introduces a major taxonomy of three classes of intrinsic-motivation mechanisms (knowledge-based intrinsic motivations based on predictors, knowledge- based intrinsic motivations based on novelty detection, and competence- based intrinsic motivations), and on this basis introduces and reviews the various contributions of the book. The contributions have been organised in five parts. The contributions of the first part further expand the con- cept of intrinsic motivations, in particular clarifying their psychological and evolutionary status and the functions that they can play in cogni- tion. The contributions of the second, third, and fourth part focus on the three different classes of intrinsic-motivation mechanisms. The fifth part is on mechanisms complementary to intrinsic motivations, such as bottom-up attention and constraint release. The last part is on new tools that can be used to investigate intrinsic motivations. For the quality of its contributors, the novelty of the topic, and its interdisciplinary per- spective, the book aims to become an important milestone and to foster further research in this intriguing new field of cognitive science.
