Learning to search: From weak methods to domain-specific heuristics

Learning from experience involves three distinct components—generating behavior, assigning credit, and modifying behavior. We discuss these components in the context of learning search heuristics, along with the types of learning that can occur. We then focus on SAGE, a system that improves its search strategies with practice. The program is implemented as a production system, and learns by creating and strengthening rules for proposing moves. SAGE incorporates five different heuristics for assigning credit and blame, and employs a discrimination process to direct its search through the space of rules. The system has shown its generality by learning heuristics for directing search in six different task domains. In addition to improving its search behavior on practice problems, SAGE is able to transfer its expertise to scaled-up versions of a task, and in one case, transfers its acquired search strategy to problems with different initial and goal states.     

Feature-processing deficits following brain injury. II. Classification learning, categorical decision making, and feature production

The claim that overselectivity in feature processing underlies the disorders that aphasics display in processing both visual and verbal material was directly tested by exploring the relationships between the behavior of brain-injured subjects on three experimental tasks: classification learning, categorical decision making, and feature production. From each of these tests a score selected as being indicative of overselective responding was entered into a principal components analysis, together with measures of visual recognition and memory, visual reasoning, naming skills, and severity of aphasia. This analysis supported the assumption that feature-processing disability is a specific and separable deficit, although related both to naming ability and to severity of aphasia. The relevance of the overselectivity hypothesis to naming difficulties following brain injury is discussed.     

An attempt to understand students' understanding of basic algebra

This paper reports the results obtained with a group of 24 14-year-old students when presented with a set of algebra tasks by the Leeds Modelling System, LMS. These same students were given a comparable paper-and-pencil test and detailed interviews some four months later. The latter studies uncovered several kinds of student misunderstandings that LMS had not detected. Some students had profound misunderstandings of algebraic notation: Others used strategies such as substituting numbers for variables until the equation balanced. Additionally, it appears that the student errors fall into several distinct classes: namely, manipulative, parsing, clerical, and “random.” LMS and its rule database have been enhanced as the result of this experiment, and LMS is now able to diagnose the majority of the errors encountered in this experiment. Finally, the paper gives a process-oriented explanation for student errors, and re-examines related work in cognitive modelling in the light of the types of student errors reported in this experiment. Misgeneralization is a mechanism suggested to explain some of the mal-rules noted in this study.     

The cognitive structure of social categories

Support for the prototype theory of categorization was found in a study of the structure of social categories. Though occupational terms such as DOCTOR are socially defined, they do not have the classical structure their clear definitional origins would predict. Conceptions of social categories are richer and more complex than those of physical object categories and subjects agree upon them. Comparison of various instructions for eliciting attributes of categories showed that whether subjects are asked to define a term, give characteristics, or describe ways they recognize members of categories, the attributes they list contribute to a prototype structure. These data provide evidence against the view that prototype structure is relevant only to an identification procedure and not to the core of concepts, as has been suggested.     

Languages and designs for probability judgment

Theories of subjective probability are viewed as formal languages for analyzing evidence and expressing degrees of belief. This article focuses on two probability langauges, the Bayesian language and the language of belief functions (Shafer, 1976). We describe and compare the semantics (i.e., the meaning of the scale) and the syntax (i.e., the formal calculus) of these languages. We also investigate some of the designs for probability judgment afforded by the two languages.