[go: up one dir, main page]

Academia.eduAcademia.edu
An Architecture for Inquiry : Building Computer Platforms for Discovery Jon L. Awbrey and Susan M. Awbrey University of Texas Medical Branch at Galveston Abstract More and more we hear the complaint that the gap between research and instruction is widening and a vital sense of motivation is falling between the cracks. It is our vision that intelligent computing systems will become a partner in the reintegration of discovery and learning within the inquiry process. We will address certain issues that must be faced if computer media are to have the characteristics necessary to support this integration. The development of the computer to date has required a careful attention to the syntax and semantics of the rather limited symbol systems we have induced computers to use. A capacity for communicating in multiple modalities with non-uniform communities of symbol users — for sharing in the discovery of a pluralistic universe — will demand a quantum leap in our understanding of the pragmatic dimensions of symbol use. In the future the capacity for inquiry must permeate the living architecture of the computer system. A computer program that begins to embody these ideas will be discussed. Introduction Today, we are recognizing that students must become active learners and problem-solvers to cope with the increasing complexity of their current and future worlds. At the same time faculty are disheartened by the growing gap between their research and instructional roles. What is called for is a reintegration of the discovery and learning processes to rekindle essential motivation. For the student this implies the development of an inquiry approach to real world problems. For the instructor, it is the opportunity to merge two currently disparate functions. It is our thesis that computers, intelligent systems in particular, can play a vital role in this reintegration. The Inquiry Process Inquiry is focused exploration. We may define it as the search for reasoned explanation or as an attempt to find laws that govern and predict outcomes. According to philosopher Charles Sanders Peirce (ref 1), inquiry is a process involving three forms of reasoning. First, a phenomenon catches our attention. It may surprise or annoy us but it does not fit with our expectations. We guess at principles that might explain it. Peirce refers to this process of positing a possible explanation as abductive reasoning. We can also term it hypothesis generation. Next, the results and consequences of the proposed explanation are considered. This is the process of deductive reasoning. Finally, actual consequences are compared to those projected. It is inductive reasoning that determines their fit. The information that anyone (interpreter) has about a phenomenon (object system) is expressed in symbols (signs). This relationship Peirce referred to as the sign relation. For purposes of our discussion, there are two important aspects of this relation. First, the roles within it may change. For example, a person may act as an interpreter or he may be a sign to someone else, as when he smiles or frowns. Secondly, it points out the importance of the interpreter to any discovery since abductive reasoning is done by the interpreter and, therefore, the initial hypothesis generation is subject to all of the constraints placed on it by the interpreter’s knowledge base and assumptions. Architecture for Discovery How then can computers assist with learning about and performing this process of inquiry? What advantages do they offer us? As noted earlier, the problems facing us today in all facets of life from socioeconomics to ecology are characterized by complexity. Computers provide a way of handling data about complex phenomena. For some time computers have been used to store and retrieve large databases of information. Models of quantitative data have also been developed to make predictions about complex systems. However, the difficult task of developing computer programs that can facilitate inquiry has not yet been fully addressed. When developed, computer systems that facilitate inquiry can become a valuable resource for both forming deductions from complex theories and for handling qualitative and sequential data from complex phenomena. A computer program capable of forming logical models based on its environment could provide the following advantages to the faculty researcher. First, by modeling the knowledge base of the researcher it would (1) make the researcher’s expressed knowledge base visible and (2) identify implicit knowledge that the researcher has and is using but which has not been incorporated into the proposed theory. Secondly, such a program might assist in hypothesis generation by identifying constraints and assumptions the researcher brings to the problem. This intelligent computer program could assist the faculty member as an instructor by making the student's knowledge base related to a specific inquiry visible. It could provide an environment for the students to perform inquiry based on the real world data gathered by the faculty researcher using the program. It would be possible to begin with the faculty performing the abductive process and pointing out a phenomenon of interest which the student would then pursue through the deductive and inductive stages. Later, the student might perform all stages of the inquiry process. Modeling the Inquiry Process The following diagram presents a dynamic model of the inquiry process. Figure 1. Dynamics of Inquiry When a phenomenon presents itself, our task is to explain it. We observe the features of the phenomenon (1) and make a guess about its explanation (abduction). We form a “theory” which we represent in terms of observed features and events. This “expressed theory” (2), is comprised of the laws and principles believed to govern the phenomenon. Using this theory we can deduce the possible consequences and outcomes it would predict (deduction) and formulate a model of it (3). We can then compare the model with the properties of the original phenomenon, some of which may need to be elicited by further experiment (induction). When a theory is expressed, the investigator may not have included all of the necessary underlying knowledge in the expressed theory. By representing the theory computationally, this missing implicit knowledge (4) often comes to light and assists in clarifying the theory (explication). A Beginning: Theme I Developing a computer program for inquiry that recognizes events, forms models of its environment, and formulates rules based on experience means careful attention to the fundamentals of the symbol systems used. The authors have developed a prototype, PC-based program designed to integrate inductive and deductive reasoning. Theme I is comprised of two components, called Index and Study. Index is a learning algorithm for sequential data. It acquires a two-level formal language that describes the qualitative features of a given domain. Study builds logical models of this domain using propositional calculus. Theme I has been applied to studies on family interaction, and a study involving its use in clinical reasoning is in process. Conclusion The development of a computer program for inquiry that uses artificial intelligence is underway. The ultimate goal of the project is development of an interactive tool for research that assists students and investigators in inquiries involving qualitative data. In the future such programs could provide an environment for students to participate in and become proficient at abductive, deductive, and inductive reasoning. It is hoped that development of a computer architecture for inquiry will assist in the reintegration of discovery and learning and restore the vitality of exploration to the educational process. References 1. C.S. Peirce, Collected Papers of Charles Sanders Peirce (Harvard University Press, Cambridge, MA, 1931–1960). Slides Slide 1 Welcome Welcome to An Architecture for Inquiry: Building Computer Platforms for Discovery Slide 2 Using Technology Today students must become active learners and problem solvers to cope with the increasing complexity of their current and future worlds. They will need to develop a lifelong affinity for learning and exploration. Educators such as John Dewey have long recognized the value of experience in education. What has changed is the ability to create environments in the classroom that both model the world and allow students to interact with it. Slide 3 Research-Instruction Our purpose is to develop computer systems that allow the reintegration of the discovery or research process with the learning process Our belief is that technology is more than an efficient store-house for facts — it can become an interactive environment for inquiry We recognize that individual students cannot recreate all discoveries or do everything from scratch but what they can experience and learn are the processes of discovery that can be continued throughout life. Slide 4 Make Visible We wish to use technology to make visible the steps that a researcher often carries out in his/her head during the discovery process. Slide 5 What Is Inquiry? Inquiry is the process of moving from an observation to an explanation. Slide 6 Phenomenon A phenomenon presents itself and we attempt to explain it. First, we observe it and collect some data either formally or informally. We guess at a possible explanation and form a theory of the laws and principles that govern the phenomenon. This process has been called abduction by logician and philosopher Charles S. Peirce. Slide 7 Deduction From our theory we propose consequences and outcomes that would result if our theory is true and we form a model of how the phenomenon would act. This is our deductive process. Slide 8 Induction This model is then compared with the original phenomenon through the process of induction. Slide 9 Clarification Often we find that many of our underlying assumptions and knowledge have not been clearly identified as part of our theory. This implicit knowledge can be brought to light to help improve our theory. Slide 10 Total Process These stages are repeated and refined to bring our model closer to the actual phenomenon and our initial theory closer to the ideal. Slide 11 AI Assist Researcher When the computer is used to create a computational model of the logic involved in these processes, a model of the researcher’s knowledge base can be created. This model can be used both to clarify the theory and to identify constraints and assumptions that are being taken for granted so the theory can be made more comprehensive and explicit. Slide 12 AI Assist Students When a computational model of the student's problem-solving processes is created, it can make the student's knowledge base visible to the student and the instructor and help identify gaps and difficulties in understanding. A program which allows students to participate in the inquiry process based on real world data gathered by researchers can also provide the opportunity for students to practice and initially learn deduction and induction and later to begin the abductive process of theory formation. The Method of Inquiry An Empirical Aphorism A Plausible Basis A Logical Theory An Outline of Models An Improved Theory An Improved Outline An Outline of Positive Features Susan M. Awbrey, Assistant Professor, Director of Learning Resource Center; Jon L. Awbrey, Faculty Associate: Computer Specialist, School of Nursing, University of Texas Medical Branch, Galveston, Texas, USA, 77550. PAGE 15