This paper begins with a brief review of SME and MAC/FAC, our simulations of matching and retrieval. Next I lay out several arguments for exploring analogy in the large, including why it is now very feasible and what we can learn by such explorations. A new constraint on cognitive simulations, the Integration Constraint, is proposed: A cognitive simulation of some aspect of analogical processing should be usable as a component in larger-scale cognitive simulations. I believe that the implications of this new constraint for cognitive simulation of analogy are far-reaching. After that, two explorations of larger-scale phenomena are described. First, I describe a theoretical framework in which we model common sense reasoning as an interplay of analogical and first-principles reasoning. Second, I describe how SME and MAC/FAC have been used in a case-based coach that is accessible to engineering thermodynamics students worldwide via electronic mail. These examples show that exploring analogy in the large can provide new insights and new challenges to our simulations. Finally, the broader implications of this approach are discussed.

Hofstadter and his colleagues have criticized current accounts of analogy, claiming that such accounts do not accurately capture interactions between processes of representation construction and processes of mapping. They suggest instead that analogy should be viewed as a form of high level perception that encompasses both representation building and mapping as indivisible operations within a single model. They argue specifically against SME, our model of analogical matching, on the grounds that it is modular, and offer instead programs such as Mitchell and Hofstadter's Copycat as examples of the high level perception approach. In this paper we argue against this position on two grounds. First, we demonstrate that most of their specific arguments invo lving SME and Copycat are incorrect. Second, we argue that the claim that analogy is high-level perception, while in some ways an attractive metaphor, is too vague to be useful as a technical proposal. We focus on five issues: (1) how perception relates to analogy, (2) how flexibility arises in analogical processing, (3) whether analogy is a domain-general process, (4) how micro-worlds should be used in the study of analogy, and (5) how best to assess the psychological plausibility of a model of analogy. We illustrate our discussion with examples taken from computer models embodying both views.

Diagrams often use repetition to convey points and establish contrasts. This paper shows how MAGI, our model of repetition and symmetry detection, can model the cognitive processes humans use when reading repetition-based diagrams. MAGI, which is based on the Structure Mapping Engine, detects repetition by aligning both visual and conceptual relational structure. This lets visual regularity of form support an understanding of the conceptual regularity such forms often depict. We describe JUXTA, which uses this insight to critique a class of diagrams that juxtapose similar scenes to demonstrate physical laws. Introduction In explanatory diagrams, repeated structures often have special significance. To underscore a point or emphasize a difference, diagrams often juxtapose events, scenes, or objects. Examples include a "before and after" display of shirts in a laundry detergent ad and a point-by-point comparison of pumps in a physics text. In such cases, visual repetition heightens cont...

Sketching is often used when working out ideas. This combination of drawing and conceptual labeling is a very natural and effective form of communication and problemsolving. Creating software that can participate in sketching provides many challenges. We outline the nuSketch approach to sketch understanding, which focuses on visual and conceptual understanding instead of recognition. We summarize three experiments in progress with the sketching Knowledge Entry Associate (sKEA), the first open-domain sketch understanding system. sKEA exploits a variety of visual and qualitative spatial reasoning capabilities and human-like analogical matching to tackle a variety of tasks. We present experimental results, and outline future plans. The nuSketch approach

Symmetry detection is a key part of human perception. One incompletely understood aspect of symmetry detection concerns orientation effects. The best-known orientation effect is the preference for vertical symmetry, where symmetry around a vertical axis is detected more quickly and accurately than symmetry at other orientations. Current symmetry detection models have difficulty explaining this effect. Using MAGI (Ferguson, 1994), we show how orientation effects may be caused by interactions between the perceived visual relations and the current reference frame. As evidence for this explanation, we simulate several orientation characteristics, including the preference for vertical symmetry and Wiser's (1981) theory of "intrinsic axes". Finally, we successfully simulate the results of a classic study by Palmer and Hemenway (1978) which explores the relationship between the preference for vertical symmetry, multiple symmetries, and inexact symmetry. Collectively, these results show that orientation effects may be due to characteristics of detected visual relations rather than either exact point-to-point equivalencies or the bilateral symmetry of the visual system.

Evan's 1968 ANALOGY system was the first computer model of analogy. This paper demonstrates that structure-mapping, when combined with high-level visual processing and qualitative representations, can do the same kinds of problems with hand-drawn sketched inputs.

Abstract. This paper describes an architecture for dynamically handling spatial relations in an incremental, nonmonotonic diagrammatic reasoning system. The architecture represents jointly exhaustive and pairwise disjoint (JEPD) spatial relation sets as nodes in a dependency network. These spatial relation sets include interval relations, relative orientation relations, and connectivity relations, but in theory could include any JEPD spatial relation sets. This network then caches dependencies between low-level spatial relations, allowing those relations to be easily assumed or retracted as visual elements are added or removed from a diagram. For example, in the architecture’s Undo mechanism, the dependency network can quickly reactivate cached spatial relations when a previously-deleted element is restored. As part of this work, we describe how the system supports higher-level reasoning, including support for creating default assumptions. We also describe how this system was integrated with an existing drawing program and discuss its possible use in diagrammatic and geographic reasoning. 1

Visual representations seem to play a significant role in many creative analogies. In this paper, we describe a specific role of visual representations: two situations that appear dissimilar non-visually may appear similar when re-represented visually.

A central problem in qualitative reasoning is understanding how people reason about space and shape with diagrams.

In order for neuroevolutionary techniques to produce increasingly complex and sophisticated topologies, new methods need to be developed which effectively exploit reuse and modularity. Bilateral symmetry is an important form of reuse and a key feature of complex biological central nervous systems. We present a method for encoding bilateral symmetry within the context of an existing neuroevolutionary framework, NEAT (NeuroEvolution of Augmenting Topologies). We then present a model of symmetry detection that relies on the symmetry in the structure of the neural system to make symmetry judgments. We demonstrate that this model performs better than an asymmetrical representation on a symmetry discrimination task in which the axis of symmetry is given. On a second task, the networks must first find the axis of symmetry before making the symmetry judgment. In this task the