Understanding computers and cognition: A new foundation for design by Terry Winograd and Fernando Flores, 207 pages, Ablex 1986, $24.95

ledge representation that appears to be missing in the chapter on Prolog. I was disappointed at the two page index, which omitted such terms as "deep knowledge", "LISP", "evaluation", "induction" or "machine learning"—these last two particularly surprising as this is the subject of three of the chapters. The references are located at the end of each chapter and total 182: 30 refer to papers written in foreign languages, and less than a quarter are since 1984. To whom will this book appeal? Those most likely to benefit include the computer scientist with experience in conventional medical informatics, any AI researcher wishing to find out about work behind the Iron Curtain, and the doctor with an interest in medical decision aids. It may not be of great benefit to the student or the professional AI worker or researcher, unless they have a specific interest in medical applications, as it fails to cover some of the recent developments in the field; neither will it be very useful to those who do not already have some idea both of the methods used in this area, and of the domain itself. In summary, though one may not expect from the title that the book contains such a wide variety of material, most of it is of interest to those working in the medical informatics field. The professional knowledge engineer or computer scientist will certainly find two or three chapters of value to them, but will perhaps not wish to add it to their own personal library.