High-level perception—the process of making sense of complex data at an abstract, conceptual level—is fundamental to human cognition. Through high-level perception, chaotic environmen-tal stimuli are organized into the mental representations that are used throughout cognitive pro-cessing. Much work in traditional artificial intelligence has ignored the process of high-level perception, by starting with hand-coded representations. In this paper, we argue that this dis-missal of perceptual processes leads to distorted models of human cognition. We examine some existing artificial-intelligence models—notably BACON, a model of scientific discovery, and the Structure-Mapping Engine, a model of analogical thought—and argue that these are flawed pre-cisely because they downplay the role of high-level perception. Further, we argue that perceptu-al processes cannot be separated from other cognitive processes even in principle, and therefore that traditional artificial-intelligence models cannot be defended by supposing the existence of a “representation module ” that supplies representations ready-made. Finally, we describe a model of high-level perception and analogical thought in which perceptual processing is integrated with analogical mapping, leading to the flexible build-up of representations appropriate to a given context. 1 The Problem of Perception One of the deepest problems in cognitive science is that of understanding how people make sense of the vast amount of raw data constantly bombarding them from their environment. The essence of human perception lies in the ability of the mind to hew order from this chaos, whether this means simply detecting movement in the visual field, recognizing sadness in a tone of voice, perceiving a threat on a chessboard, or coming to understand the Iran–Contra affair in terms of

There are conflicting findings regarding the level at which recognition is initiated. To address the issue, we propose a model for recognition based on analogical mapping and transfer. The preattentive visual information maps to the structural representations in memory, thus allowing the formation of anticipations about what is present in the environment. Because the representations of all parts, objects, and events are interconnected, the anticipations can be formed at any level, and attention is biased to the most rigorously anticipated aspect of the environment. The mechanisms of the model, as well as some simulation results, are presented

The one-to-one mapping constraint has been implemented in almost all of models for analogy making, but has not yet been studied systematically. A psychological experiment was performed that tested whether the strength of one-to-one mapping pressure can be manipulated by a prior task unrelated to analogymaking. The results demonstrate that people’s answers to match-to-sample tasks can be systematically shifted toward one or another direction by a relatively short manipulation.

We propose an object recognition model based on analogical mapping and transfer. The objective of our model is to be able to generate and bind structural representations; and to recognize objects from a small set of primitives. The input is mapped to the associative memory and activation is spread upwards. Anticipations are generated through local mappings and transferred to be tested serially in the order of their relevance. Due to these mechanisms, our model is able to simulate phenomena such as object priming and global precedence effect. Additionally, it provides a framework for integrating visual perception and other higher-order cognitive processes.

Analogical reasoning is a highly sophisticated cognitive process and it could be the missing link for the understanding of cognitive abilities in natural complex systems. In many current approaches for modeling analogies and analogical reasoning, cognitive abilities are examined in isolation from related issues in order to control the environment and the underlying context. Although these research endeavors are successful in various aspects and applications, it seems as if each result of modeling a particular ability minimizes the chances to reach overall goals like modeling human-level intelligence. With this workshop we aim to bring together researchers who are working in the field of analogical reasoning and are relating their models to other cognitive abilities. This workshop focuses on analogy as an integrating basis for human cognition. These workshop proceedings provide evidence that the field of analogies is both maturing and still of high relevance to many research areas in cognitive science. After a highly-competitive review process 9 submissions were selected for long presentations at the workshop. The acceptance rate of submissions for long presentations was 45%. Further 5 submissions were accepted for short presentations. All 14 extended abstracts are included in the proceedings and assigned to one of the following four categories: