This volume contains the proceedings of AISB91, which was held from 16-19 April 1991 at the University of Leeds. This was the 8th biennial conference organized by the Society for the Study of Artificial Intelligence and Simulation of Behaviour. The conference focused on research as opposed to applications, with an emphasis on four special themes: emergent functionality in autonomous agents; neural networks and self-organization; constraint logic programming; and knowledge level expert systems research. There were invited talks by Andy Clark (University of Sussex) on the philosophical foundations of the field, Rolf Pfeifer (University of Zurich) on emotion, and Tony Cohn (University of Leeds) on common sense modelling using logic. There was a closing address by the Conference Chair, Luc Steels, who speculated on the role of consciousness in artificial intelligence. The first four papers discuss research in distributed AI, which is concerned with the problem of how multiple agents and societies of agents can be organized to co-operate and collectively solve a problem. The papers in the second section testify to a renewed interest in the area of robotics and autonomous agents. Recently, there has been a shift of emphasis in research in these areas towards the construction of complete agents. This has led to a review of some of the traditional concepts. Reasoning remains a core topics of AI, with the the papers in the new modes of reasoning section explore some of the current work to find new forms of reasoning. The papers on knowledge systems highlight the recent shift in emphasis from symbol-level considerations (which focus on the formalism in which a system is implemented) to knowledge-level considerations. The final sets of papers focus on recent work in theorem proving and machine learning. This book of proceedings on computer science, artificial intelligence and simulation of behaviour is intended for researchers and postgraduate students.