This article describes an integrated theory of analogical access and mapping, instantiated in a

Analogy and similarity are often assumed to be distinct psychological processes. In contrast to this position, the authors suggest that both similarity and analogy involve a process of structural alignment and mapping, that is, that similarity is like analogy. In this article, the authors first describe the structure-mapping process as it has been worked out for analogy. Then, this view is extended to similarity, where it is used to generate new predictions. Finally, the authors explore broader implications of structural alignment for psychological processing.

We describe a computational model of how analogs are retrieved from memory using simultaneous satisfaction of a set of semantic, structural, and pragmatic constraints. The model is based on psychological evidence suggesting that human memory retrieval tends to favor analogs that have several kinds of correspondences with the structure that prompts retrieval: semantic similarity, isomorphism, and pragmatic relevance. We describe ARCS, a program that demonstrates how these constraints can be used to select relevant analogs by forming a network of hypotheses and attempting to satisfy the constraints simultaneously. ARCS has been tested on several data bases that display both its psychological plausibility and computational power.

This paper presents a view of planning as a task supported by a dynamic memory. This view attempts to integrate models of memory, learning and planning into a single system that learns about planning by creating new plans and analyzing how they interact with the world. We call this view of planning Case-Based Planning. A case-based planner makes use of its own past experience in developing new plans. It relies on its memory of observed effects, rather than a set of causal rules, to create and modify new plans. Memories of past successes are accessed and modified to create new plans. Memories of past failures are used to warn the planner of impending problems, and memories of past repairs are called upon to tell the planner how to how to deal with them. This view of planning from experience supports and is supported by a learning system that incorporates new experiences into the planner's episodic memory. This learning algorithm gains from the planner's failures as well as its successe...

The use of analogy in human thinking is examined from the perspective of a multiconstraint theory, which postu-lates 3 basic types of constraints: similarity, structure, and purpose. The operation of these constraints is appar-ent in laboratory experiments on analogy and in natural-istic settings, including politics, psychotherapy, and sci-entific research. The multiconstraint theory has been im-plemented in detailed computational simulations of the analogical human mind. ~ ~ any parents know that young children take comfort in getting a kiss on an injury to "make it better. " Little Aaron, aged 24 months, would routinely come to his mother saying things like, "I bump my head. Kiss it. " But one morning, for the first time ever, the tables turned. While his mother was dressing

: Holographic Reduced Representations (HRRs) are a method for encoding nested relational structures in fixed width vector representations. HRRs encode relational structures as vector representations in such a way that the superficial similarity of the vectors reflects both superficial and structural similarity of the relational structures. HRRs also support a number of operations that could be very useful in psychological models of human analogy processing: fast estimation of superficial and structural similarity via a vector dot-product; finding corresponding objects in two structures; and chunking of vector representations. Although similarity assessment and discovery of corresponding objects both theoretically take exponential time to perform fully and accurately, with HRRs one can obtain approximate solutions in constant time. The accuracy of these operations with HRRs mirrors patterns of human performance on analog retrieval and processing tasks. Keywords: neural networks, distributed representations, binding, analogy, analog retrieval, structure, chunking, systematicity 1

Most AI simulations have modeled memory retrieval separately from comprehension, even though both activities seem to use many of the same processes. We have developed REMIND, a model that performs both episodic memory retrieval and language understanding with a single spreading-activation mechanism. This approach has a number of advantages over retrieval-only models. First, because the comprehension process makes inferences about actors * plans and goals, REMIND is able to get abstract remindings that would not be possible without an integrated model. It also allows a more psychologically-plausible model of reminding than previous approaches, since all aspects of a text's interpretation affect what is retrieved through the spreading-activation process, as in human reminding. An inferencing-based retrieval model such as REMIND also has several computational advantages over pure retrieval models. The effects of the understanding process eliminate the need for the separate, purely structural comparisons used in most analogical retrieval models. Further, it potentially explains how the explicit indexing of case-based reasoning models can be eliminated, while retaining its benefits as an emergent property of the comprehension process. 1

While most approaches to similarity assessment are oblivious of knowledge and goals, there is ample evidence that these elements of problem solving play an important role in similarity judgements. This paper is concerned with an approach for integrating assessment of similarity into a framework of problem solving that embodies central notions of problem solving like goals, knowledge and learning. We review empirical findings that unravel characteristics of similarity assessment most of which have not been covered by purely syntactic models of similarity. A formal account of similarity assessment that allows for the integration of central ideas of problem solving is developed. Given a goal and a domain theory, an appropriate perspective is taken that brings into focus only goal-relevant features of a problem description as input to similarity assessment. Key words: problem solving, case-based reasoning, similarity, EBG 1 Introduction In recent years, there has been an upsurge of inter...

The cognitive process of analogical reasoning has generally been thought to occur in at least three stages: retrieval of a source description from memory, mapping of that source description to the target description, and transfer of relationships from source to target. Here we are concerned with the first stage, the retrieval of relevant sources from long-term memory for their use in analogical reasoning. Specifically we ask: what can people retrieve from long-term memory when stimulated by a description of some situation, and how do they do it? Psychological experiments suggest that subjects display two sorts of retrieval patterns: a novice pattern and an expert pattern. Those subjects that show the novice pattern are much more likely to retrieve so-called merely-apparently-similar sources to a target description rather than analogically-related sources. Merelyapparently-similar sources for a target are characterized by the two descriptions sharing superficial features, such as the types or identities of objects or actors involved, whereas in an analogical relationship the source and target share description structure that is useful for making analogical inferences. In contrast to those who show the novice pattern, those subjects who show the expert pattern are much more likely to retrieve analogically-related sources to a target stimulus, as long as the target and sources fall into a domain in which

This research focused on the role that higher order structural properties of stereotypic knowledge play in the processing of social information. It is argued that stereotypic assumptions about cause-effect relations provide important constraints for the causal structure underlying the perceiver's subjective representation of social information. Experiment 1 shows how, within the context of a jury decision experiment, the causal structure underlying stereotypic knowledge about African Americans influences the construal of causality in a situation involving a member of that group. Results from 2 additional experiments indicate that this construal effect is based in part on stereotypic knowledge affecting the encoding of the trial evidence instead of on biasing responses at the output stage. The implications of these findings are discussed, and a theoretical framework is offered according to which the application of category knowledge involves not only the matching of stereotypic attributes but also the alignment of structural relations in the environment. The notion that subjective experience goes beyond the bare sensation of stimuli, that we actively construe reality instead of passively registering our environment, has long guided psycho-