Designed to determine whether abstract and concrete concepts are classified similarly, this study used three separate experiments to explore the vertical and horizontal dimensions used by computer scientists to categorize the common concepts of their field. It found that concrete concept categorization operates over a wider range than previously…

Examines processes by which general but weak search methods are transformed into powerful, domain-specific search strategies by classifying types of heuristics learning that can occur and components that contribute to such learning. A learning system--SAGE.2--and its structure, behavior in different domains, and future directions are explored. (36…

Summarizes an initial foray in tackling artificial intelligence problems using a connectionist approach. The task chosen is visual recognition of Origami objects, and the questions answered are how to construct a connectionist network to represent and recognize projected Origami line drawings and the advantages such an approach would have. (30…

Reports results of studies with an unsupervised learning paradigm called competitive learning which is examined using computer simulation and formal analysis. When competitive learning is applied to parallel networks of neuron-like elements, many potentially useful learning tasks can be accomplished. (Author)

Presents MAC/FAC, a model of similarity-based retrieval that attempts to capture psychological phenomena; discusses its limitations and extensions, its relationship with other retrieval models, and its placement in the context of other work on the nature of similarity. Examines the utility of the model through psychological experiments and…

Presents four computational interpretations of the Gricean conversational maxims in the generation of referring expressions: (1) full brevity interpretation; (2) greedy heuristic interpretation; (3) local brevity interpretation; and (4) incremental interpretation. Examines problems involved in generating definite noun phrases that are appropriate…

Challenges Smolensky's theory that human intuitive/nonconscious cognitive processes can only be accurately explained in terms of subsymbolic computations in artificial neural networks. Symbolic learning models of two cognitive tasks involving nonconscious acquisition of information are presented: learning production rules and artificial finite…

The study of consciousness requires a language for exact and precise communication. This research project investigated how workers from different cultural contexts value information that is relevant to their job performance. Thirty-five randomly selected mechanics, from Sweden, England, West Germany, Italy, and the United States, were selected as…

Describes the construction of a knowledge representation from the identification of the problem (nephrotic syndrome) to a running computer simulation of causal reasoning to provide a vertical slice of the construction of a cognitive model. Interactions between textbook knowledge, observations of human experts, and computational requirements are…

Describes a study which analyzed problem solvers' error detection processes by instructing subjects to think aloud when solving statistical problems. Effects of evaluative episodes on error detection, detection of different error types, error detection processes per se, and relationship of error detection behavior to problem-solving proficiency…

Describes three studies concerned with learning to operate a control panel device and how this learning is affected by understanding a device model that describes its internal mechanism. Results indicate benefits of a device model depend on whether it supports direct inference of exact steps required to operate the device. (Author/MBR)

Used experimental tasks--spontaneous telling of a story, reading, and parsing--to determine whether empirical data reflect narrative structure of stories. It was concluded that spontaneous pausing reflects the narrative structure and can be used as a guide to constructing theories of narrative structure and deciding between competing theories.…

This theoretical analysis articulates verbal rule properties and use implications, describes an empirical study examining characteristics of problem solving performance when verbal rules are instructed as a problem solving method, and discusses role of verbal rules in problem solving and implications for learning procedural skills. (MBR)

Describes an instructional experiment aimed at helping sixth-graders sustain reflective thought processes in composition independently. Instruction included modeling of thinking aloud, clue use stimulating self-questioning, and direct strategy instruction emphasizing dialectical synthesis of conflicting ideas. Results indicate gains were made at…