This article proposes a cognitive framework describing the software development process in object-oriented design (OOD) as building internal representations and developing rules. Rule development (method construction) is performed in two problem spaces: a rule space and an instance space. Rules are generated, refined, and evaluated in the rule space by using three main cognitive operations: Infer, Derive, and Evoke. Cognitive activities in the instance space are called mental simulations and are used in conjunction with the Infer operation in the rule space. In an empirical study with college students, we induced different representations to the same problem by using problem isomorphs. Initially, subjects built a representation based on the problem description. As rule development proceeded, the initial internal representation and designed objects were refined, or changed if necessary, to correspond to knowledge gained during rule development. Differences in rule development processes among groups created final designs that are radically different in terms of their level of abstraction and potential reusability. The article concludes by discussing the implications of these results for object-oriented design.

Complex systems such as computers, aerospace systems, etc., are often tested by using a sequence of tests to exercise the functionality of the system. If the system fails a test, an error message is generated, initiating the test selection (TS) phase. The troubleshooter must decide whether or not to run more tests. Should the troubleshooter decide to conduct more tests, a test must be chosen as it may no longer be useful to conform to the predefined sequence. While in the TS phase, the troubleshooter will repeatedly make these decisions until he is done. The authors present a domain-independent framework to automate TS that is based on two computational models, TI and TP. Both models are needed for the authors show there are applications for which one model performs well, while the other model performs poorly. The use of the appropriate model for automating TS is indicated by certain characteristics of the testing sequence and the system under test.>

Sacks & Doyle provide an excellent overview of the fundamental limitations of the SPQR representations for reasoning about the qualitative properties of dynamic systems. We take this opportunity to outline some new directions for qualitative reasoning. In this paper, we provide a rigorous mathematical characterization for the term “qualitative property” in the context of static and dynamic systems. Based on these characterizations, we show that interval representations are well suited for reasoning about the qualitative properties of static systems such as qualitative comparative statics and qualitative stability. Moreover, we also show that symbolic computations help in the derivation of useful global properties of dynamic systems which can be used to guide numerical sampling of differential equations. The integration of symbolic and numeric methods provides a powerful approach for automating the qualitative analysis of differential equations.

The directions for near-future development of artificial intelligence (AI) can be described in terms of four dichotomies: the use of reasoning versus the use of knowledge; the roles of parallel and of serial systems; systems that perform and systems that learn to perform; and programming languages derived from the search metaphor versus languages derived from the logical reasoning metaphor. Although the author believes that there are reasons for emphasizing knowledge systems (production systems) that are serial, capable of expert performance, and designed in terms of the search metaphor, the other pathways are also important and should not be ignored. In particular, empirical work is needed in the construction and empirical testing of the performance of large systems to explore all of these branching pathways.>

Probe selection (PS) in machine diagnosis is viewed as a collection of models that apply under specific conditions. This makes it possible for three polynomial-time optimal algorithms to be developed for simplified PS models that allow different probes to have different costs. The work is compared with previous research, wherein H.A. Simon and J.B. Kadane (1975) review and develop a collection of models for optimal problem-solving search. The relationship between these models and the three newly developed algorithms for PS is explored. Two of the algorithms are unlike the ones discussed by Simon and Kadane. The third cannot be related to the problem-solving models.>

The author discusses the roles that humans now play versus the roles that could be taken over by artificial intelligence in developing computer systems. Also discussed is how the intelligent part of the automatic system can communicate effectively with humans. Topics covered include an artificial intelligence overview; weak methods; the heuristic search; the problem space; the knowledge base; expert systems; and conclusions drawn from a description of the general artificial intelligence paradigm.

Abstract This article reviews recent progress in modeling human cognitive processes. Particular attention is paid to the use of computer programming languages as a formalism for modeling, and to computer simulation of the behavior of the systems modeled. Theories of human cognitive processes can be attempted at several levels: at the level of neural processes, at the level of elementary information processes (e.g., retrieval from memory, scanning down lists in memory, comparing simple symbols, etc.), or at the level of higher mental processes (e.g., problem solving, concept attainment). This article will not deal at all with neural models; it focuses mainly upon higher mental processes, but not without some attention to modeling the elementary processes and especially to the relationships between elementary and complex processes.

article Free Access Share on Experiments with a Heuristic Compiler Author: Herbert A. Simon Carnegie Institute of Technology, Pittsburgh, Pennsylvania and The RAND Corporation, Santa Monica, California Carnegie Institute of Technology, Pittsburgh, Pennsylvania and The RAND Corporation, Santa Monica, CaliforniaView Profile Authors Info & Claims Journal of the ACMVolume 10Issue 4Oct. 1963 pp 493–506https://doi.org/10.1145/321186.321192Published:01 October 1963Publication History 48citation616DownloadsMetricsTotal Citations48Total Downloads616Last 12 Months57Last 6 weeks1 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my Alerts New Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

In this paper we describe a complex information processing system, which we call the logic theory machine, that is capable of discovering proofs for theorems in symbolic logic. This system, in contrast to the systematic algorithms that are ordinarily employed in computation, relies heavily on heuristic methods similar to those that have been observed in . human problem solving activity. The specification is written in a formal language, of the nature of a pseudo-code, that is suitable for coding for digital computers. However, the present paper is concerned exclusively with specification of the system, and not with its realization in a computer. The logic theory machine is part of a program of research to understand complex information processing systems by specifying and synthesizing a substantial variety of such systems for empirical study.