Mary A. Peterson - US grants

When people are asked to do a number of cognitive tasks involving vision, such as imaging an object or searching for a particular object in a large visual array, it is well known that certain physically measurable, "objective" aspects of the display, such as its complexity, influence how people do the tasks. In addition, subjective aspects, ones that cannot be measured physically, such as familiarity or meaning, also influence people's performance. Few experiments have separated objective from subjective variables sufficiently that one can tell conclusively which ones underly particular experimental effects. The experiments in this research project will isolate subjective variables from objective variables and will examine the role of subjective variables in a series of visually-based cognitive tasks, including imagery tasks, search tasks, and perceptual reversal tasks. These tasks are all potentially influenced by the viewer's knowledge regarding the structure of the shape to be imagined or manipulated. The research flows from the notion that subjective effects in visual cognition are mediated by the viewer's structural knowledge regarding the particular components from which a shape is constructed and the relative locations of these components; the greater the structural knowledge, the faster images can be manipulated, searched, and scanned, and hence, the more efficient the viewer's performance. The experiments will test this notion and, in the process, will attempt to separate the construct of structural knowledge from that of wholistic processing and to examine the interaction between spatial attention and structural knowledge. Finally, the research will develop a functional taxonomy of tasks to identify the types of processing that yield subjective effects. The hypothesis is that these effects will occur when tasks permit the mental manipulation of categorizable, basic- level shapes rather than requiring manipulation of detailed particular shapes. This taxonomy of tasks, together with the construct of structural knowledge, may be useful in identifying the "functional vocabulary" of visual cognition that is employed in real-world tasks requiring speed and accuracy.

Theoretical attempts to understand visual perception often partition perceptual processes into those processes that can be performed simultaneously and those that must be performed successively. Most theorists have assumed that figure-ground organization and shape recognition are performed successively. Figure-ground organization is the determination of which regions in the visual field represent figures and which constitute the backgrounds against which the figures are viewed. Figure and ground regions are separated by a contour; the region to which the contour is assigned appears to be the recognizable figure, whereas the other region appears to be an undifferentiated ground which simply appears to continue behind the figure. Despite the linkage between shape recognition and figure-ground organization, these theorists have argued that figure-ground organization must precede shape recognition, appealing either explicitly or implicitly to the argument that one cannot recognize a figure until there is a figure to be recognized. Consequently, figure-ground relationships are generally conceived to be determined by stimulus variables only (i.e., variables such as the relative areas, symmetries, convexities, and enclosure of the two regions on either side of a contour). Contrary to this argument, however, previous studies in Peterson's laboratory have shown that certain shape recognition analyses are conducted prior to the determination of figure and ground. These experiments can be readily integrated into current theoretical approaches to shape recognition, thereby rendering this argument obsolete. The planned experiments will examine the conditions under which shape recognition processes can contribute to figure-ground computations. One series of experiments will identify which specific types of contours (e.g., luminance contours, binocular disparity contours, and/or color contours) permit shape recognition computations to be performed simultaneously with the analysis of the stimulus variables relevant to figure-ground organization. Another series of experiments will test whether the shape representations corresponding to both interpretations of a reversible figure-ground stimulus are activated simultaneously prior to the assignment of the figure-ground contour. A third series of experiments will attempt to explicate the rules governing the interactions between shape recognition variables and stimulus variables in figure-ground computations. These experiments will provide information about fundamental aspects of visual processing, including the manner in which analyses of physical variables interact with cognitive variables in perceptual organization. These experiments are certain to be relevant to the design of surrogates for physical scenes.

Award Abstract
PI: Mary Peterson

One of the earliest steps in the process of perceiving the visual world is to segregate the visual input into figures and backgrounds. When two adjacent regions in the visual field share a contour, one region is typically seen as the figure, and the other region appears to be the ground. Figures are endowed with a definite shape, whereas grounds are shapeless near the contours they share with figures. It has traditionally been assumed that figure-ground segregation is based on low-level cues, such as depth cues and configural cues (i.e., symmetry, convexity, enclosure or smallness of relative area). These low-level cues are thought to operate before memory representations are accessed. Contrary to the traditional view, recent research indicates that memories of object structure are accessed earlier in the course of perceptual processing than assumed on the traditional model. Furthermore, these object memories exert an influence on image segregation. These findings are consistent with a number of models that allow higher-level processes, such as those accessing memories of object structure, to influence processing at a figure-ground stage. The proposed experiments test the viability of a unique model that does not consider figure-ground segregation to be a stage in the hierarchy of visual processes. Instead, it considers figure-ground segregation to be one possible outcome of the interactions among early image segregation processes operating on both sides of a contour. The model entails facilitation between processes operating on the same side of a contour and inhibition between processes operating on opposite sides of a contour. In addition to accounting for the evidence indicating that memories of object structure can affect image segregation, this model explicitly accounts for three previously unexplored segregation phenomena. Those phenomena are: (1) objects can be recognized consciously only when they are depicted by figures, not when they are depicted by grounds; (2) grounds are shapeless; (3) figure-figure outcomes (in which the regions on both sides of a contour appear to have a definite shape) can be perceived, although figure-ground outcomes tend to be preferred. The experiments in the current proposal constitute a strong test of the inhibition proposed in this new model.

The planned experiments use a priming paradigm to assess whether processes operating on regions ultimately determined to be grounds are first activated and then inhibited in the course of perceptual processing. Other experiments in the proposal explore the relationship between segregation-mediated inhibition and an attentional phenomenon, called negative priming. The proposed experiments promise to provide evidence that will be critical for distinguishing among competing models of how high-level processes affect the earliest visual processes -- those processes that are fundamental to perception and action. In addition, these experiments will integrate two formerly disparate areas of study -- attentional inhibition and image segregation.

Perception seems effortless and yet it remains one of the most challenging scientific topics. For instance, how do we separate beach chairs from the beach in visual perception, or from the people in a crowded beach scene? Indeed it remains a general puzzle how we may see objects in a scene separate from one another and from their backgrounds. Recent computer models succeed in segregating objects in pictures only when past experience is allowed to play a role. With NSF support Dr. Mary Peterson will follow this lead to investigate scene separation in human perception, specifically the effects of previous experience with objects on later perception of the same objects.
Broader impacts include supervision and training of students from underrepresented groups and outreach activities to local high schools. The funded research will also supply equipment to the University of Arizona for data analysis and storage. Findings will also be disseminated through a laboratory website.

We spend nearly every waking moment searching with our eyes. Even a bored casual glance around a room is, at heart, an attempt to look for something. In modern life, visual search includes such important tasks as a doctor's inspection of medical X-ray images, a homeland security check of luggage for anomalies, or a pilot's quick glance across vital gauges on aircraft instrument panels. All living things capable of movement search incessantly, whether simple one-celled organisms or brainy hominids. In the laboratory, visual search has become a way to study the mind, whether as part of an intelligence test, a neurological exam, or in esoteric studies of attention, language, and social perception.
Dr. Robert Rauschenberger and Dr. Mary A. Peterson will use support from the National Science Foundation to study a hitherto unrecognized fact about visual search: that the targets of search (the proverbial "needle in the haystack") can become more similar to their distractors (the "hay") as time goes on. That is, if you present the "needle and haystack" for a 'long' time (1/4 of a second), search is paradoxically more difficult than if you present the very same display for only a brief time (1/10 of a second). Common sense suggests the opposite outcome. The insights to be gained from this research have the potential to alter our understanding of a fundamental component of human behavior significantly. The broader impact of the project may include innovations in the design of modern instruments, increases in airport security, and better tests for medical problems. The funded project also includes an educational component to fund training for Dr. Rauschenberger in theory and methods of complex nonlinear systems.

Visual perception seems trivially easy. Nevertheless, it has long withstood the attempts of both computer scientists and vision scientists to crack its code, most likely because past theories were based predominantly on conscious vision. In order for perceivers to experience a coherent visual world, perceptual input must be organized into separate objects, or "figures." As part of this process, the brain decides where a figure lies with respect to every border between two contiguous regions of space. Consider a vertical border, for instance. The brain decides whether that border is a boundary for a figure lying on the left or the right. When the border is perceived as a boundary of a figure lying on the left, the region on the right seems simply to continue behind the figure at the border; no shape is perceived there. These "figure-ground" decisions are made outside of conscious awareness. Because of this, it is extremely difficult to investigate the mechanisms that produce figure-ground perception. An important, unanswered, question is whether figure-ground perception is accomplished via fast, feed-forward mechanisms or whether iterative mechanisms involving feedback from higher processing levels are involved. In feed-forward models, the input is processed in successive stages until a coherent percept emerges. In iterative models, feed-forward processing is not sufficient; feedback from higher to lower levels is necessary to create the percept.

Mary Peterson and her colleagues at the University of Arizona have designed visual displays that allow them to investigate this question. One display type is a small symmetric, bounded silhouette lying on a larger ground. Portions of familiar shapes are hidden along the groundside of the silhouette's border; these shapes are not perceived consciously, only the shape of the enclosed silhouette is perceived consciously. (The classic "face-vase illusion" is an example of this.) Previous experiments have shown that the shape of the hidden object is suppressed when it is not perceived, supporting the view that figure-ground perception results from inhibitory competition between shapes that might be seen on opposite sides of a border; the winner is perceived as the shaped figure, whereas the loser is suppressed and the portion of space where it might have been seen is perceived as a shapeless ground. These displays are designed to isolate competition at the shape processing stage. Dr. Peterson and colleagues will also conduct a series of experiments which test whether suppression can be observed at lower levels where individual parts are represented and at higher levels where shape descriptions ("semantics") are represented. According to a feed-forward account, suppression of the losing shape would prevent access to its semantics; hence, no effects should be evident at higher levels. Without feedback, no effects should be evident at lower, part-processing, levels either. An iterative view could account for suppression at lower and/or higher levels by assuming that the outcome of the between-shape competition was relayed to higher and lower levels. Thus the proposed experiments will adjudicate between these competing views of how perception occurs. The researchers will also attempt to identify the processes involved in segregating figures from grounds in crowded real-world scenes. The perception of these more complex displays constitutes a challenge to current iterative models of figure-ground perception. The planned research will provide a foundation for neurophysiological experiments and formal computational models of vision and will contribute to our understanding of the temporal and spatial dynamics of shape perception.

The goal of Women in Cognitive Science is to improve the visibility of women scientists by fostering an environment that welcomes and nurtures young women scholars, to contribute to the professional development of scholars throughout their career, and to facilitate creation of a network that will provide contacts and connections to other women in science. Several workshops are designed for women in cognitive science, especially women in the early stages of their academic career. The workshops focus on negotiation techniques to create opportunities and optimize mechanisms to sustain research visibility and productivity. A second focus is on grant application writing for predoctoral, postdoctoral, and early career scientists. Workshops will take place at meetings of the Psychonomic Society, the Cognitive Science Society, and the Association for Psychological Sciences. The workshops will take the form of a public forum with invited speaker-panelists to initiate discussion about best practices for the professional advancement of women in cognitive science at the individual and institutional level. By partnering with these established societies, the workshops will maximize the outreach potential to a group that continues to be underrepresented in senior academic positions in the cognitive sciences.