There may be a limit on our capacity to suppress anthropocentric tendencies toward non-human others. Normally, we do not reach this limit in our dealings with animals, the environment, etc. Thus, continued striving to overcome anthropocentrism when confronted with these non-human others may be justified. Anticipation of super artificial intelligence may force us to face this limit, denying us the ability to free ourselves completely of anthropocentrism. This could be for our own good.

A Pyramid Model of the Perception of Partially Visible Figures Michael R. Scheessele (mscheess@iusb.edu) Department of Computer & Information Sciences, Indiana University - South Bend 1700 Mishawaka Ave., South Bend, IN 46634 USA Zygmunt Pizlo (pizlo@psych.purdue.edu) Department of Psychological Sciences, Purdue University 1364 Psychological Sciences Bldg., West Lafayette, IN 47907 USA Introduction Frequently, figures in our visual field are only partially visible. One figure may partially occlude another, for example, or a particular figure may appear fragmented due either to camouflage or to low contrast between it and the background. Despite such challenges, the human visual system routinely perceives figures that may only be partially visible. One prior theory of the perception of partially occluded figures (Nakayama, Shimojo, & Silverman, 1989) states that contours iintrinsici to a figure of interest must be distinguished from those iextrinsici to it and that this classification requires depth cues. Our theory proposes that the human visual system can use a variety of cues, local or global, to perform this classification and that this classification serves as the basis for perception of both partially occluded and fragmented figures. Further, we propose that an exponential pyramid, from the machine vision literature, provides a good model of how the human visual system implements this classification. Exponential Pyramid Model Description The Exponential Pyramid has been proposed as an adequate model of the human visual system (Rosenfeld, 1990; Pizlo, Salach-Golyska, & Rosenfeld, 1997). Our model uses a inon-overlapped quad-pyramidi. Assume that the bottom layer of the pyramid has n processing nodes. The next layer has n/4 nodes, the one above that n/16 nodes, and so on. The top layer has only one node. Each node in a layer connects with four distinct echildi nodes in the immediately lower layer and one eparenti node in the immediately higher layer. Such a pyramid has (log 4 n) + 1 layers. Each node in the pyramid has limited memory and processing capability. An image is input to the bottom layer (Jolion & Rosenfeld, 1994). The image may also be represented at each higher layer (with increasing spatial scale or ereceptive field sizei). Our model features a bottom-up processing stage followed by a top-down stage. In the bottom-up stage, local variance of various contour features (e.g., orientation, length) is computed. When the variance of a contour feature abruptly changes between successively higher layers, the presence and position of a figure in the image is indicated (i.e., the figure ecomes into viewi). In the top-down stage, the statistical information computed in the bottom-up stage is used to classify image contours as either intrinsic or extrinsic to the target figure. The model has only one free parameter: the standard deviation of decisional noise. Model and human performance were compared across 11 experimental conditions. Method In each trial of the human psychophysical experiments, a polygonal figure was partially occluded by simple shapes n diamonds (Exp. 1, two conditions) and squares (Exp. 2, nine conditions). A subjectis task was to respond whether the figure was presented in its upright or rotated (180 o ) position. Contours of occluders differed from those of the figure in terms of orientation (Exp. 1) and length (Exp. 2). Depth cues from occluders were minimal. Model simulations were run for all 11 conditions using the same sets of stimuli as those used by the human subjects. Results Subjects used orientation (Exp. 1) and length (Exp. 2) differences between the contours of a target figure and those of occluders, in detecting the figure. Model simulations accounted well for human performance in the 11 conditions of Experiments 1 and 2. Conclusions The human visual system can detect and use a variety of cues, local or global, to classify contours as either intrinsic or extrinsic to a partially visible figure. Our exponential pyramid-based computer model provides a good account of how the human visual system implements this process. References Jolion, J. M., & Rosenfeld, A. (1994). A pyramidal framework for early vision. Dordrecht, The Netherlands: Kluwer Academic Publishers. Nakayama, K., Shimojo, S., & Silverman, G. H. (1989). Stereoscopic depth: Its relation to image segmentation, grouping, and the recognition of occluded objects. Perception, 18, 55-68. Pizlo, Z., Salach-Golyska, M., & Rosenfeld, A. (1997). Curve detection in a noisy image. Vision Research, 37, Rosenfeld, A. (1990). Pyramid algorithms for efficient vision. In C. Blakemore (Ed.), Vision: Coding and efficiency. Cambridge, Great Britain: Cambridge University Press.

I propose a framework, derived from moral theory, for assessing the moral status of intelligent machines. Using this framework, I claim that some current and foreseeable intelligent machines have approximately as much moral status as plants, trees, and other environmental entities. This claim raises the question: what obligations could a moral agent (e.g., a normal adult human) have toward an intelligent machine? I propose that the threshold for any moral obligation should be the "functional morality of Wallach and Allen [20], while the upper limit of our obligations should not exceed the upper limit of our obligations toward plants, trees, and other environmental entities.

Coverage of discrete structures is necessary for any computer science curriculum. More broadly, teaching skill in critical thinking is essential to providing a well-rounded college education. We describe our experience with incorporating a 'critical thinking module' (CTM) into a discrete structures (DS) course.

The two-stage model of amodal completion or TSM (Sekuler & Palmer, 1992), and the ambiguity theory (Rauschenberger, Peterson, Mosca, & Bruno, 2004) provide conflicting accounts of the phenomenon of amodal completion in 2-D images. TSM claims that an initial mosaic (2-D) representation gives way to a later amodally completed (3-D) representation. Furthermore, the 2-D representation is accessible only prior to formation of the 3-D representation. On the other hand, the ambiguity theory claims that the 2-D and 3-D representations develop in parallel and that preference for one of the coexisting representations over the other may be subject to the influence of spatiotemporal context provided by other elements in the visual display. Our experiments support the claim that, once formed, both representations coexist, with spatiotemporal context potentially determining which representation is perceived.

We established human performance in recognition of handwritten ZIP codes taken from the standard CEDAR database. We expect that the result will serve as a benchmark for machine performance in recognition of handwritten ZIP codes.

In The and the scientific revolution, C.P. Snow (1959) described the chasm between pure and applied science, on the one hand, and the arts and humanities, on the other. Snow was concerned that the complete lack of understanding between these two cultures would hamper the spread of the scientific/industrial revolution from rich nations to poor. Because of his conviction that this revolution had made lives longer and more comfortable for people of developed nations, he forcefully argued that the intellectual must be bridged--the sooner the better. The gap between these cultures, of course, still exists. Meanwhile, the arts are neglected in primary and secondary schools. Further, the science vocabulary of adults in the U.S. appears to be so poor that a scientific theory is considered suspect simply because it is a theory. Such problems may create increased competition between the cultures. A probable result would be short-sighted prescriptive measures that are at best worthless and at worst dangerous to the mission of bridging the cultures. A better approach may be to examine interdisciplinary fields where this gap seems less wide, for clues to a bridge. Introduction In The Two Cultures and the Scientific Revolution, (1) C.P. Snow (1959) described the emergence of broad, yet distinct, intellectual in Western society. The first of these, embodied by the literary intellectual, encompasses the arts and humanities. The other, embodied by the scientist, comprises mathematics and technology, in addition to the natural and social sciences. Snow was disturbed by the deep lack of understanding and communication between members of these cultures: Thirty years ago the had long ceased to speak--to each other: but at least they managed a kind of frozen smile across the gulf. Now the politeness has gone, and they just make faces. (2) His conviction was that science and technology had made life longer and more bearable for those fortunate enough to have been born in industrialized nations and, further, that the scientific revolution could ease suffering for those living in poor nations. However, he believed that this gulf between the was hindering the spread of the scientific revolution from developed to developing nations, in part because the resulting lack of more complete knowledge was acting to constrain the judgment of policy-makers. (3) Thus, Snow proposed more broad education for students with the hope that this gulf between the might begin to be bridged. The basic problem of the persists however. Proposed U.S. funding increases for math and science will likely exacerbate tension between the cultures, due to the perception that the arts and humanities--especially the arts--are already neglected when compared to the sciences. It would be tempting for the scientist to just sit back and enjoy the windfall, while the literary intellectual rails at the prospect of yet more money being diverted toward the sciences (presumably at the expense of the arts and humanities). Increased tension between the would be unfortunate though--at least in the U.S. At best it would achieve nothing; at worst it may distract attention from a critical question, is America in danger of losing its dominance in science and technology? If the answer is yes, then the problem will definitely be addressed. After all, no reasonably informed person today would dispute the role that science plays in a healthy, prosperous society. But what form should a solution take? Increased rivalry between the would likely obscure paths to an answer. U.S. Science And Technology In Peril? The United States has enjoyed dominance for decades in terms of scientific discovery and technical innovation. That America also has one of the highest standards of living in the world is no coincidence, and this echoes the assertion by Snow that: The scientific revolution is the only method by which most people can gain the primal things (years of life, freedom from hunger, survival for children)--the primal things which we take for granted and which have in reality come to us through having had our own scientific revolution not so long ago. …

We found that the human error rate in recognition of individual handwritten digits is 2.37%. This differs somewhat from two prior studies [1], [2].

When a figure is only partially visible and its contours represent a small fraction of total image contours (as when there is much background clutter), a fast contour classification mechanism may filter non-figure contours in order to restrict the size of the input to subsequent contour grouping mechanisms. The results of two psychophysical experiments suggest that the human visual system can classify figure from non-figure contours on the basis of a difference in some contour property (eg length, orientation, curvature, etc). While certain contour properties (eg orientation, curvature) require only local analysis for classification, other contour properties (eg length) may require more global analysis of the retinal image. We constructed a pyramid-based computational model based on these observations and performed two simulations of experiment 1: one simulation with classification enabled and the other simulation with classification disabled. The classification-based simulation gave the superior account of human performance in experiment 1. When a figure is partially visible, with few contours relative to the number of non-figure contours, contour classification followed by contour grouping can be more efficient than contour grouping alone, owing to smaller input to grouping mechanisms.

Assigning the development of a poker-playing agent as a group project allows flexibility with respect to the topics and techniques typically covered in an introductory Artificial Intelligence course. A poker agent project also provides students the experience of 'authentic' AI research, due to the status of poker as an 'unsolved' problem in AI. Despite this status, a poker agent project is feasible for a semester, half-semester, or quarter-long group project. Problems in assigning group projects are also considered, as are suggestions for mitigating these problems.

In a retinal image, contours belonging to a figure of interest may be intermixed with other contours (caused by occlusion, camouflage, low contrast, etc.), making difficult the identification of figure contours and thus the figure itself. In psychophysical experiments, we found that identification of a figure is facilitated by: differences in relative contour orientation, relative contour curvature, and relative contour length between contours that belong to the figure and those that do not. In short, a difference in any contour property seems to facilitate correct identification of a figure. A computational model, based on the exponential pyramid architecture, was constructed and model simulations of several conditions from the psychophysical experiments were performed. A critical aspect of the model is that it performs contour classification by using statistics computed from the entire image. Model simulations accounted well for the results of 11 experimental conditions, using just one free parameter. These results suggest that the human observer uses global features to make local contour classification decisions in the image and that the exponential pyramid architecture can adequately model perceptual mechanisms involved in figure-ground segregation.