Education & learning publications

This paper traces a progression of four computer-based methods for studying and fostering both the structure and the on-line development of knowledge. Each empirical technique employs ECHO, a connectionist model that instantiates the theory of explanatory coherence (TEC). First, verbal protocols of subjects’ reasonings were modeled post hoc. Next, ECHOpredicted, a priori, subjects’ textbased believability ratings. Later, the bifurcation/bootstrapping method was developed to elicit and account for individuals’ background knowledge, while assessing intercoder reliability regarding ECHO simulations. Finally, Convince Me, our “reasoner’s workbench,” automated the explication both of subjects’ knowledge bases and of their belief assessments; the Convince Me software permits contrasts between the model’s predictions and subjects’ proposition-wise evaluations. These experimental systems enhance our understanding of the relationships among-and determinant features regarding-hypotheses, evidence, and the arguments that incorporate them.

We propose a template library as a good representation of programming knowledge, and programming case studies as part of an effective context for illustrating design skills and strategies for utilizing this knowledge. In this project, we devised an on-line network of Pascal programming templates called a template library, and tested it with subjects (classified as novice, intermediate, and expert Pascal programmers) both as a stand alone resource and in conjunction with programming case studies. We investigated three questions using these tools: 1) How do subjects organize templates? 2) How well can subjects understand and locate templates in the template library? 3)
Does the template library help subjects reuse templates to solve new problems? Results suggest that the template representations helped subjects remember and reuse information, and that subjects gained deeper understandings if the representation was introduced in the context of a programming case study.

This article describes the design and evaluation of IC-HIP, a multimedia course on integrated circuit manufacturing (Schank & Rowe, 1992). Subjects browsed the course via standard hypermedia links or linear paths. Learning effects were assessed based on navigation
method (hyperlinks vs. path), prior knowledge (low vs. high), and other factors (e.g., subjects’ stated interests in semiconductors, and kinds and number of course nodes viewed). Effects of navigation method, prior knowledge, and pre-instruction interest on nodes viewed
(by media type and topic area), were also assessed. Results suggest that subjects who browsed via hypermedia links tended to more often bridge topic areas rather than explore them in depth, but there were little or no learning differences by knowledge or navigation group, and neither prior interest nor nodes viewed (by number, topic area, or media type) were correlated with learning. These results and future work are discussed.
Keywords:

Abstract The goal of this article is to construct an integrated approach to collaboration and conceptual change. To this end, a case of conceptual change is analyzed from the point of view of conversational interaction. It is proposed that the crux of collaboration is the problem of convergence: How can two (or more) people construct […]

Simpson’s paradox exemplifies a class of problems that can arise when the logic used to reason about the semantics of propositional sentences does not adequately capture certain dependencies between sentences of interest. This paradox has been known as early as 1903 [YUL03], and has been discussed extensively in the statistical literature [SIM51, DAW79, BLY73, CHU42]. The phenomena that typically give rise to Simpson’s paradox can occur in cases such as destructive testing (e.g., determining the breaking strength of materials in orthogonal directions), and identifying the composition of complex alloys. It has also been reported to occur in “real-life” several times since its discovery [KNA85, WAG82]. One such occurrence received wide attention in 1973 over the appearance of a sex bias in the admission policy for graduate students at the University of Berkeley [BIC75]. Given that automated systems will be expected to recognize and cope with the underlying phenomena of this paradox, it is important to develop effective methods for dealing with them, particularly as it impacts the choice of logics that systems must use to reason about real world problems. Only recently, however, has there been any significant indication that Simpson’s paradox merits serious attention by the AI community [PEA88].

A framework for understanding the effects of discovery learning on the transfer of problem-solving skill is presented.A distinction is drawn between applying a learned strategy on a transfer problem versus having to generate a novel strategy to solve a transfer problem. The main premise of the framework is that requiring discovery of a strategy while in training encourages the activation or refinement of procedures that are useful for generating a novel strategy. In general, then, the primary benefit of discovery is that it should facilitate transfer to tasks requiring a novel strategy. Learning by discovery, however, may provide little benefit for tasks that can be completed only by applying the learned strategy. Two experiments provided support and further refinement of this hypothesis. Experiment 1 used a transfer problem that could be solved with the general strategy learned in training but required new move sequences to instantiate the strategy. The results indicated that, when transfer required new move sequences to implement a general strategy learned previously, discovery did not enhance transfer of that strategy. In experiment 2, some transfer problems required using a strategy other than that learned in training. As predicted, in this transfer situation, having to discover a strategy while in training produced better transfer than being provided with a strategy in training. Thus, discovering a strategy provided  benefits when a new strategy had to be generated to solve a transfer problem but not when the learned strategy could be applied, albeit with new moves, to the transfer problem. Educational implications are discussed.