The inductive category formation framework, an influential set of theories of learning in psychology and artificial intelligence, is deeply flawed. In this framework a set of necessary and sufficient features is taken to define a category. Such definitions are not functionally justified, are not used by people, and are not inducible by a learning system. Inductive theories depend on having access to all and only relevant features, which is not only impossible but begs a key question in learning. The crucial (...) roles of other cognitive processes (such as explanation and credit assignment) are ignored or oversimplified. Learning necessarily involves pragmatic considerations that can only be handled by complex cognitive processes.We provide an alternative framework for learning according to which category definitions must be based on category function. The learning system invokes other cognitive processes to accomplish difficult tasks, makes inferences, analyses and decides among potential features, and specifies how and when categories are to be generated and modified. We also examine the methodological underpinnings of the two approaches and compare their motivations. (shrink)

Most models of computer programming explain the programmer's behaviour by a single design strategy. This article presents a cognitive architecture that uses cue‐based search to model multiple design strategies including procedural, functional, means‐end or focal, and opportunistic design. The model has been implemented in an artificial intelligence (AI) system that generates Pascal programs from English specifications.Knowledge is represented as nodes that reside in internal or external memory, where a node encodes an action that may range from a line of code (...) to a routine in size. A program is built by linking nodes through a search cue of the form. The cue is broadcast to memory, and any matching node is returned; the cue provides a question to ask, and the return provides the answer. A cue on the newly linked node is then selected as a new focus, and the search process repeated. Each design strategy defines a specific node visiting order that traverses the program structure through its links. (shrink)