Showing papers in "International Journal of Human-computer Studies \/ International Journal of Man-machine Studies in 1973"

Towards more intelligent teaching systems

Abstract: This paper suggests criteria against which the “intelligence” of a teaching machine can be judged. With the electronic computer in mind, distinctions are made between pre-structured, generative, adaptive, and self-improving teaching systems. From the work which has been carried out at Leeds, examples are taken which illustrate the characteristics and intelligence of these systems and the requirements for their implementation.

Caste: A system for exhibiting learning strategies and regulating uncertainties

Abstract: CASTE (Course Assembly System and Tutorial Environment) is a facility for observing and controlling human learning. This paper describes the system and presents data on its use as a conversational system for teaching elementary concepts of probability theory to Technical College Students. A summary is given of the work on adaptive teaching systems that led to the development of CASTE for subject matters where students can and do adapt a variety of learning strategies. For such cases adaptive teaching is not sufficient; a system is called for which learns about a student's preferred learning strategy by engaging him in a dialogue about his learning; CASTE is such a system.

A theory of conversations and individuals (Exemplified by the Learning Process on CASTE)

Abstract: The main tenet of the theory is that the minimal experimental situation for making psychological observations is a conversation. The logical and structural requirements for making such observations are presented in a series of icons which dynamically represent formalisms in the abstract theory of self-reproducing automata. Two sorts of stable, self-reproducing systems are distinguished: mechanically characterized individuals (M-Individuals) and psychologically characterized individuals. (P-Individuals). A conversation is a P-Individual (a selfreproducing class of procedures) that is executed in one or more of a restricted class of M-Individuals (processors). The theory is exemplified by work on learning and teaching using CASTE (Course Assembly System and Tutorial Environment) which is itself a physical embodiment of the theory in the form of a vehicle for observing conversations. Other exemplifications are given as interpretations, within the current theory, of the paradigms extant in conventional experimental psychology.

The design and evaluation of an adaptive teaching system

Abstract: The value in using the computer as a teaching machine is likely to be a function of how well it can adapt to individual differences in performance. The term “adaptive” is used to imply that the student is routed to material which follows different teaching strategies and that tasks are put together, on-line, to fit a particular student's competence. The components of an adaptive instructional system are, representations of tasks and student performances, and a set of teaching operations which are controlled by means-ends guidance rules. Two examples which illustrate these components are discussed. The first, set in junior mathematics, uses generative material and adjusts task difficulty, type of feedback and remedial teaching to suit individual performances. Experiments which were necessary to establish the means-ends guidance rules are outlined. The second example, the planning of investigations in applied statistics and in physical chemistry, concerns the teaching of undergraduates. The material is pre-stored and the difficulties in designing adaptive decision rules in these contexts are indicated. However, experiments show the need for such rules. Finally, the integration of this teaching with more conventional methods is briefly outlined.

Human operators and automatic adaptive controllers: A comparative study on a particular control task

Abstract: Many schemes have been put forward for general-purpose learning machines capable of performing a variety of tasks, but little work has been done on direct comparison of the performance of these with the most adaptable of all controllers, the human operator. A suitable environment for such a comparison is described here, and experiments with several different adaptive controllers are reported. The most successful of the automatic adaptive controllers consists of a series of stochastic estimators, the outputs from which determine the probability that the controller will take a certain action for each region of the environment's statespace. The results show that human operators can achieve good control by estimating derivative information from a continuous display of the state of the environment, but are badly affected by random external disturbances. Even the best automatic controller considered can use only coarsely quantized input information, and so has poorer control, but it is relatively immune to external disturbances.

Experience with a mini-computer-based hospital administration system

Abstract: As part of the design of a hospital administration system based on a mini-computer, a pilot study has been conducted of an admission system for one speciality within the hospital. This paper describes the pilot study, with particular emphasis on the conversational data entry procedures used with non-computer oriented personnel and the documents produced to collect and disseminate information appropriately. The overall study demonstrates the feasibility of setting up flexible and useful computer-based systems integrated into the present hospital administration procedures, given ziuitable computer resources and a thorough analysis of the hospital requirements.

Digital image processing for information extraction

Abstract: In recent years the modern digital computer has been used to process images, to emphasize details, to sharpen pictures, to modify the tonal range, to aid picture interpretation, to remove anomalies, and to extract quantitative information A price to be paid for this extreme flexibility in handling linear and non-linear operations is that a number of anomalies caused by the camera, such as geometric distortion, MTF roll-off, vignetting, and non-uniform intensity response must be taken into account or removed to avoid their interference with the information extraction process Once this is done, computer techniques may be used to emphasize details, perform analyses, classify materials by multi-variate analysis (usually multi-spectral), detect temporal differences, etc Digital processing may also be used to modify various aspects of pictures to enhance the ability of the human photo interpreter in extracting information A number of these processes are illustrated in this paper

Opinion-distribution model for subjective rating studies

Abstract: A newly-developed opinion-distribution model employs the logistic function as a near-Gaussian cumulative distribution on a notional T scale. The logistic is also used to transform the T scale into a finite t -scale, where distributions may be obtained in a wide variety of shapes through simple manipulation of the T -scale distribution. The binomial-type distributions encountered in category-scaling experiments can be accurately described by quantizing the appropriate t -scale distributions using the same number of equal intervals as there are categories. Apart from the value of the model as an analytical device, there is strong evidence that opinion categories are often formed by equal division of a continuum bearing a close relation to the T -scale. Where opinions recorded on a continuous scale were investigated, the t -scale distributions were found to apply directly. Because of the use of the logistic for both T -scale distributions and the transform, the distributions on the T scale are described by very simple algebraic expressions. The t scale facilitates the comparison of results obtained with different category scales. On the other hand, manipulation of distributions, including combining or separating sources of variance, is easy on the T scale.

Computer-assisted mathematics instruction for community College students

Abstract: The Individualization Project of the Ontario Institute for Studies in Education (OISE) and several Colleges of Applied Arts and Technology (CAATS) jointly are developing a curriculum for upgrading mathematics skills. The project uses the computer facilities of the OISE and the National Research Council of Canada for on-line diagnosis and instruction of mathematics skills. The Computer-Assisted Learning system software was designed at the OISE and implemented with a view toward independence from any specific computer; therefore all system programs are coded in widely standardized languages. The authoring language, CAN -4 which is interpreted by the system programs, allows the curriculum designer and student a large degree of flexibility in the instructional process. The mathematics curriculum material is designed as a set of generative modules keyed to behavioral objectives prerequisite for first year CAATS students. The project's goal is to provide the student with the most efficient means of instruction suited to his individual needs. Advanced assessment techniques requiring an online computer system provide valid decisions which can reduce the student's time required for mastery of the curriculum. Reports on the performance of each student and course are used by the project staff to improve the system and the teaching/learning process.

Language Standards for Computer-Aided Learning.

Abstract: Instructional applications of the computer have resulted in the development of many specialized high-level programming languages; each one aimed at providing effective access to the computer by student, teacher, course author, and/or researcher. At any given installation, the benefits to be gained from such a language are quite evident. However, the variety of languages which have evolved has resulted in great difficulty in exchanging Computer-Aided Learning (CAL) program materials among different institutions. Thus much effort is expended on development of software which exists elsewhere, but not in a form suitable for the particular computer installation at the institution wishing to make use of it. A common programming language for CAL applications would represent a major step towards permitting exchange of CAL software between different computer installations. To be effective, such a language must take into account the variety of approaches used in instructional applications of the computer. It must consider the requirements of users with a variety of skills. It must include provision for the utilization of a wide range of specialized input-output devices required to permit effective communication between the user and the computer. As with all standards, provision must be made for the CAL language standard to be updated in order to respond to changes in user requirements. The development and maintenance of standards for programming languages requires the co-operation of the user, the equipment supplier, and the researcher. As a first step towards arriving at definition of user requirements for a CAL programming language, the National Research Council of Canada's Associate Committee on Instructional Technology convened a working panel whose members have participated actively in a variety of CAL applications, with a variety of programming languages involved. The task assigned to the working panel was the definition of a functional specification of a programming language to meet user requirements. To arrive at a standard from the functional specification, requires the production of a detailed specification, followed by implementation at a number of centres. This implementation stage must be completed before one can say a standard truly exists, since a standard has very little meaning to the user if he cannot apply it.

An automaton framework for neural nets that learn

Abstract: Brindley (1967, 1969, 1972) has discussed nets of several types of formal neurons, many of whose functions are modifiable by their own input stimuli. Because Brindley's results are widely referred to, for example Marr (1970, 1971) and include some of the scarce non-trivial theorems on learning nets, it is important that serious side-conditions be made explicit. The language of finite automata is used to mathematicize the problem of adaptation sufficiently to remove some ambiguities of Brindley's approach. We close the paper by relating our framework to other formal studies of adaptation.

What do standard transformational grammars produce? — A computational study

Abstract: The base rules and transformations of a recently published elementary grammar of English are transcribed into a computer format and tested. Problems encountered in transcribing the grammar are explored and it is suggested that the emphasis on formal argumentation in linguistics is not justified by the current level of syntactic theory.

Response routing in selcuk networks and lashley's dilemma

Abstract: Given a large net of modules of limited logic and memory in which connectivity is primarily to near neighbors with one-way channels, the question is asked of how responses to a generally broadcast request can be routed to the site of the request. A routing procedure which does not require site information is proposed. The procedure is based on the Selcuk Principle which maintains that in properly constructed networks routing can be achieved with each module remembering only on which input channels the request had first arrived. The procedure is successfully applied to a neural network model proposed by Eccles and to one advanced by Burns. The strongest requirement of the proposed procedure is that there be switching among output channels and this appears consistent with the findings of Chun, Raymond & Lettvin (1970) that there is selective invasion of axonal arborization. Since the Selcuk Network scheme achieves response routing on the basis of strictly local information, pursuant to the broadcast of a general message, it has applicability to content-based addressing and to computer-to-computer communication nets. It also constitutes an approach to pattern recognition whereby the image on the “retina” is transformed into a time spectrum of responses, the analysis of which yields information on angle, size, curvature, and position of the edges.

An interactive graphics system for computer-assisted musical composition

Abstract: The essential components of an ideal facility for computer-assisted musical composition and research are briefly described. More detailed consideration is given to the central role of a graphical display unit in such a system. The display acts as a control unit, provides a visual representation of the music score for editing, and allows the user to initiate music transformations and specify arrangements of submelodies. Some transformations and arrangements of the sort commonly used in composition are described, including transposition, inversion, retrograde, augmentation, and diminution. Design of an efficient, compact, and flexible data structure for a computer-aided music facility is undertaken. The proposed structure is a combination of pointer vectors and linked lists. A portion of the proposed system has been implemented on the CDC GRID display computer interfaced to an IBM 360/67 at the University of Alberta. The implementation focuses on the seldom met needs of the composer with respect to the symbolic medium and hardware through which he interacts with the computer during the composition process. The command language implemented to assist the composer is described and illustrated. The implementation seeks to demonstrate the feasibility and value of providing traditional music symbols on a graphical display unit as the central component of a computer-aided music facility.

Computer-Assisted Instruction on a Small Computer.

Abstract: The Computer-Assisted Instruction (CAD Project in the Faculty of Education at the University of Calgary has been primarily concerned with the use of a small time-shared computer. Seven terminals are in use in the University, and nine are currently located in schools and a research institute for the developmentally handicapped. The paper describes the development of simulations, computer-assisted instruction and computer-managed instruction programs. A CAI language and an information system have been written for the time-shared machine, and these are discussed. Most programs have been developed for use at school level. They relate, for example, to elementary arithmetic English, social studies, mathematics, physics and counselling. Particular attention is now being given to programs for the handicapped. Experience with this machine indicates, certain advantages and also certain limitations in the use of a small computer. It is comparatively inexpensive to bring such a system into operation and to develop a variety of programs. However, the difficulty of maintaning a number of programs in constant use indicates the need for access to a larger machine and a more versatile CAI language. Particular attention is drawn to the need for terminals specifically designed for CAI, and the need to obtain some compatibility of both hardware and software within Canada in order to facilitate co-operation in the development and use of instructional programs.

The Computer-Aided Learning Program at the National Research Council of Canada

Abstract: This paper, delivered at a Conference on “Transmission Technology for Education” sponsored by the Council of Ministers of Education, Canada, in May 1972, deals with the activities of the National Research Council of Canada (NRC) in the field of computer-aided learning. The first portion deals with the reasons for undertaking the work. The second part describes the work itself, and the facilities available. A third part describes the means of co-ordination between educational users, the various educational authorities, and the laboratories. The final portion is a series of questions which arise out of the problems associated with the division of responsibilities in the Canadian context. Other questions are of a more general kind, as they relate to the utility of CAL and the direction which further technology ought to take.

The Computer-Aided Instruction Activities of the Division of Educational Research Services at the University of Alberta

Abstract: The Division of Educational Research Services has operated an IBM 1500 Computer-Aided Instruction (CAI) system for the past four years. The activities of the system can be categorized into three major areas: (a) demonstration; (b) research; and (c) production. The demonstration Activities have involved a large number of lay and professional groups, teachers and students from local schools, and university staff and students. The research activity has been initiated from three sources: graduate students in various faculty departments, faculty staff, and the non-academic support staff for the system. Academic researchers have used the system more as a data-collection device, while graduate students have used the system more in the study of computer-assisted instruction. During the past year the system has been used for the teaching of reading to beginning deaf children of age five to six years, APL statistical laboratories, coursewriter programming to university students, and to provide an enrichment program to students in junior and senior high schools. It has been used to study problems in linguistics as well as in the measurement of intelligence through a simulation of the Wechsler Intelligence Scale for Children (WlSC). The largest continuous-production type of course which has been operated now for two years is the medical series in cardiology. This series has been the most successful in terms of student learning and attitudes. A second project has now been funded for the development of programs to simulate medical patients and some work has already been started in this area. More recently a new research area has been opened by the funding through Canada Council of an oculometer project. This project is concerned with developing the hardware interface necessary to have a small mini-computer monitor the video output of a Honeywell video oculometer, and to calculate pupiliary dilation and the target being observed by the eye. The development is directed towards the study of eye movement and pupil dilation while a student interacts with the CAI computer.

Making Computer-Aided Instruction make a difference in college teaching

Abstract: This presentation attempts to discuss some of the factors that enable a computer-based instructional system to function as part of a practical and effective learning system, operating within the academic mainstream of a university environment. It is the author's view that all too many university Computer-Aided Instruction (CAD installations are so far removed from this context that CAI as an instructional tool is already in some danger of following the “programmed instruction” of an earlier era to well-deserved oblivion. A brief summary of CAI activities at Simon Fraser University will be given, with particular mention of the organization of CAI services, the use of our course-ware by other institutions, and our attempts to interface audio-visual devices such as slides, audiotapes, and the Philips PIP unit, to a conventional CAI terminal system. Finally, we will comment on our experience with the Coursewriter III system, and on the steps that must be taken if this system is to become a versatile and adequate instructional tool.

The Role of the Mini-Computer in Computer-Aided Instruction.

Abstract: This paper will deal with the concept of using a specially-designed graphics system that is being developed about a mini-computer to appraise the capability of carrying out computer-aided instruction (CAD at a modest cost in remote locations where it may not be possible to provide a large-scale system with time-shared terminals. At this stage, most of the system hardware has been constructed and the software has commenced. The system is constructed about a 12-bit word mini-computer with a 4K-core memory; the graphics facility consists of a logic-character generator with a high degree of flexibility, a line generator, a 4K-core refresh memory, 8 × 10 in display tube and a light pen. As back-up storage for display files, a magnetic tape cassette drive is included in the system. For interactive working there is a teletype and a set of push buttons, in addition to the light pen. By providing hardware generators, the amount of data storage is minimized and their novel design is such, that they could be produced easily in quantity by LSI techniques thereby keeping the cost to a minimum.

A Co-operative research project in Computer-Aided Learning

Abstract: In 1967, the National Research Council (NRC) began a preliminary study of the application of computers as aids to learning. This initial work led to the establishment of a central research facility which is used by the NRC and a number of educational research organizations in a co-operative program of research into computer-aided learning. This central facility includes a medium-scale time sharing computer which is accessible to the participating organizations by means of remote terminals. A major objective of the project is to provide a facility which will allow the active co-operation of research workers throughout Canada in the development and evaluation of Computer-Aided Learning (CAL) systems to meet Canadian requirements. The NRC efforts are concentrated in the areas of development of terminal equipment, specialized computer facilities, and system programs. Examples of such development include audio tape and disc storage units, a transparent touch-sensitive tablet for computer input, alphanumeric and graphic display devices, line-concentrator systems and supporting-system programs. The co-operating educational organizations provide the facilities for the development and testing of course material and evaluation of terminal equipment. In some cases they assist in the development of system programs. The first participating educational organization to go “on line” was the Ontario Institute for Studies in Education in Toronto which was linked to the computer early in 1970. Since that time, the number of on-line participants has grown to six. The project provides a means for close communication between the educational researcher and the system designer. In this way, effective use can be made of available resources to arrive at a system specification which more closely meets the requirements of the user than might otherwise be possible.