Probably the book was doomed at the very beginning, because the goal of writing a cognitive science textbook is just not a good idea. The purpose of a textbook is to allow beginners to rapidly assimilate the intellectual framework of a scientific field. The textbook is a repackaging, in an easier-to-digest form, of the most important research papers of the field. So the textbook is a ladder, provided by current experts, that allows students to climb rapidly and reach the experts’ level, so that they can start working on the important current problems of the field. A good textbook will be coherent, because it reflects the underlying intellectual coherence of the field; and it will be directed, leading the student in a progressive manner from basic ideas and simple problems to more advanced theory, applications, and research problems.
If you accept this description, then you can see why writing a cognitive science textbook is probably a bad idea. For a textbook to make sense, the field it describes must be mature and coherent. The practitioners must have a shared set of ideas (a Kuhnian paradigm), upon whose significance they all agree, for the textbook to present as established introductory material. It must be possible to identify the key research findings, and associated papers or experiments, so that those findings can be repackaged into the textbook. A mature field is one that has produced a cumulative series of progressively more sophisticated intellectual achievements, and the textbook must describe this history, in a streamlined form, to the student.
Cognitive science, as a field, probably fails to meet these requirements. It is a strange hybrid of ideas from computer science, statistics, psychology, neuroscience, philosophy, and linguistics. There is no clear progression from one idea to the next—what is the connection between SRHDLU and Wason selection test? Obviously, understanding SHRDLU is not a prerequisite for doing research in cognitive science in the same way that understanding the Schrodinger equation is a prerequisite for doing research in physics.
My advice for people interested in cognitive science would be to read the interesting or relevant research papers directly, and to read the textbooks in the allied fields of computer science, neuroscience, linguistics, psychology, and so on.
Probably the book was doomed at the very beginning, because the goal of writing a cognitive science textbook is just not a good idea. The purpose of a textbook is to allow beginners to rapidly assimilate the intellectual framework of a scientific field. The textbook is a repackaging, in an easier-to-digest form, of the most important research papers of the field. So the textbook is a ladder, provided by current experts, that allows students to climb rapidly and reach the experts’ level, so that they can start working on the important current problems of the field. A good textbook will be coherent, because it reflects the underlying intellectual coherence of the field; and it will be directed, leading the student in a progressive manner from basic ideas and simple problems to more advanced theory, applications, and research problems.
If you accept this description, then you can see why writing a cognitive science textbook is probably a bad idea. For a textbook to make sense, the field it describes must be mature and coherent. The practitioners must have a shared set of ideas (a Kuhnian paradigm), upon whose significance they all agree, for the textbook to present as established introductory material. It must be possible to identify the key research findings, and associated papers or experiments, so that those findings can be repackaged into the textbook. A mature field is one that has produced a cumulative series of progressively more sophisticated intellectual achievements, and the textbook must describe this history, in a streamlined form, to the student.
Cognitive science, as a field, probably fails to meet these requirements. It is a strange hybrid of ideas from computer science, statistics, psychology, neuroscience, philosophy, and linguistics. There is no clear progression from one idea to the next—what is the connection between SRHDLU and Wason selection test? Obviously, understanding SHRDLU is not a prerequisite for doing research in cognitive science in the same way that understanding the Schrodinger equation is a prerequisite for doing research in physics.
My advice for people interested in cognitive science would be to read the interesting or relevant research papers directly, and to read the textbooks in the allied fields of computer science, neuroscience, linguistics, psychology, and so on.