Early computation of part structure: evidence from visual search

The visual system represents object shapes in terms of intermediate-level parts. The minima rule proposes that the visual system uses negative minima of curvature to define boundaries between parts. We used visual search to test whether part structures consistent with the minima rule are computed preattentively--or at least, rapidly and early in visual processing. The results of Experiments 1 and 2 showed that whereas the search for a non-minima-segmented shape is fast and efficient among minima-segmented shapes, the reverse search is slow and inefficient. This asymmetry is expected if parsing at negative minima occurs obligatorily. The results of Experiments 3 and 4 showed that although both minima- and non-minima-segmented shapes pop out among unsegmented shapes, the search for minima-segmented shapes is significantly slower. Together, these results demonstrate that the visual system segments shapes into parts, using negative minima of curvature, and that it does so rapidly in early stages of visual processing.     

An attention account of neural priming

Repetition priming of familiar stimuli (e.g., objects) produces a decrease in visual cortical activity for repeated versus novel items, which has been attributed to more fluent processing for repeated items. By contrast, priming of unfamiliar stimuli (e.g., abstract shapes) produces an increase in visual cortical activity. The mechanism for priming-related increases in activity for repeated unfamiliar stimuli is unknown. We hypothesised that such increases in activity may reflect attentional allocation to these items. We tested this hypothesis using a priming-spatial attention paradigm. During Phase 1 of Experiment 1, participants viewed unfamiliar abstract shapes and familiar objects. During Phase 2, participants identified target letters (S or H). Each target letter was preceded by a non-informative shape or object cue that was repeated (from Phase 1) or novel in the same (valid) or opposite (invalid) hemifield. In Experiment 2, we manipulated shape familiarity by presenting shapes once or six times during Phase 1. For both experiments, at valid locations, target identification accuracy was higher following repeated versus novel unfamiliar item cues and lower following repeated versus novel familiar item cues. These findings support our hypothesis that priming-related increases in visual cortical activity for repeated unfamiliar items may, in part, reflect attentional allocation.     

Parts of visual objects: an experimental test of the minima rule

Three experiments were conducted to test Hoffman and Richards's (1984) hypothesis that, for purposes of visual recognition, the human visual system divides three-dimensional shapes into parts at negative minima of curvature. In the first two experiments, subjects observed a simulated object (surface of revolution) rotating about a vertical axis, followed by a display of four alternative parts. They were asked to select a part that was from the object. Two of the four parts were divided at negative minima of curvature and two at positive maxima. When both a minima part and a maxima part from the object were presented on each trial (experiment 1), most of the correct responses were minima parts (101 versus 55). When only one part from the object--either a minima part or a maxima part--was shown on each trial (experiment 2), accuracy on trials with correct minima parts and correct maxima parts did not differ significantly. However, some subjects indicated that they reversed figure and ground, thereby changing maxima parts into minima parts. In experiment 3, subjects marked apparent part boundaries. 81% of these marks indicated minima parts, 10% of the marks indicated maxima parts, and 9% of the marks were at other positions. These results provide converging evidence, from two different methods, which supports Hoffman and Richard's minima rule.     

Parts, cavities, and object representation in infancy

Part representation is not only critical to object perception but also plays a key role in a number of basic visual cognition functions, such as figure-ground segregation, allocation of attention, and memory for shapes. Yet, virtually nothing is known about the development of part representation. If parts are fundamental components of object shape representation early in life, then the infant visual system should give priority to parts over other aspects of objects. We tested this hypothesis by examining whether part shapes are more salient than cavity shapes to infants. Five-month-olds were habituated to a stimulus that contained a part and a cavity. In a subsequent novelty preference test, 5-month-olds exhibited a preference for the cavity shape, indicating that part shapes were more salient than cavity shapes during habituation. The differential processing of part versus cavity contours in infancy is consistent with theory and empirical findings in the literature on adult figure-ground perception and indicates that basic aspects of part-based object processing are evident early in life.     

Imperfect invariance to object translation in the discrimination of complex shapes

The positional specificity of short-term visual memory for a variety of 3-D shapes was investigated in a series of 'same'/'different' discrimination experiments, with computer-rendered stimuli displayed either at the same or at different locations in the visual field. For animal-like shapes, we found complete translation invariance, regardless of the interstimulus similarity, and irrespective of direction and size of the displacement (experiments 1 and 2). Invariance to translation was obtained also with animal-like stimuli that had been 'scrambled' by randomizing the relative locations of their parts (experiment 3). The invariance broke down when the stimuli were made to differ in their composition, but not in the shapes of the corresponding parts (experiments 4 and 5). We interpret this pattern of findings in the context of several current theories of recognition, focusing in particular on the issue of the representation of object structure.     

Lateralized effects of categorical and coordinate spatial processing of component parts on the recognition of 3D non-nameable objects

Participants performed two object-matching tasks for novel, non-nameable objects consisting of geons. For each original stimulus, two transformations were applied to create comparison stimuli. In the categorical transformation, a geon connected to geon A was moved to geon B. In the coordinate transformation, a geon connected to geon A was moved to a different position on geon A. The Categorical task consisted of the original and the categorically transformed objects. The Coordinate task consisted of the original and the coordinately transformed objects. The original object was presented to the central visual field, followed by a comparison object presented to the right or left visual half-fields (RVF and LVF). The results showed an RVF advantage for the Categorical task and an LVF advantage for the Coordinate task. The possibility that categorical and coordinate spatial processing subsystems would be basic computational elements for between- and within-category object recognition was discussed.     

Is visual image segmentation a bottom-up or an interactive process?

Visualimage segmentation is the process by which the visual system groups features that are part of a single shape. Is image segmentation a bottom-up or an interactive process? In Experiments 1 and 2, we presented subjects with two overlapping shapes and asked them to determine whether two probed locations were on the same shape or on different shapes. The availability of top-down support was manipulated by presenting either upright or rotated letters. Subjects were fastest to respond when the shapes corresponded to familiar shapes—the upright letters. In Experiment 3, we used a variant of this segmentation task to rule out the possibility that subjects performed same/different judgments after segmentation and recognition of both letters. Finally, in Experiment 4,we ruled out the possibility that the advantage for upright letters was merely due to faster recognition of upright letters relative to rotated letters. The results suggested that the previous effects were not due to faster recognition of upright letters; stimulus familiarity influenced segmentation per se. The results are discussed in terms of an interactive model of visual image segmentation.     

Combined perceptual or motor-related expectancies modulated by type of cue

The interaction of two expectancies was examined. These were either two perceptual or two response-related expectancies. Perceptual expectancies were induced by combining spatial cuing with feature cuing on a trial-by-trial basis. Cues consisted either of two integrated parts, such as two arrows, or two separated pieces, such as an arrow and a word. Spatial-cuing effects were reduced on trials with invalid feature cues, as compared with valid ones. However, the interaction of spatial cuing and feature cuing was modulated by the type of cue used to induce expectancies. With integrated cues, spatial-cuing effects were reduced about twice as much as with separated cues. The same effect of type of cue was found in Experiment 2, although finger cuing was combined with hand cuing. With integrated cues, finger-cuing effects were much smaller on trials with invalid hand cues, as compared with valid ones. With separated cues, however, finger-cuing effects were additive to hand-cuing effects. The similarity of the results within perceptual- and motor-cuing tasks suggests that a general principle governs the combination of expectancies, such as that outlined in the framework of the proposed adjusted expectancy model.     

Visual mental rotation of possible and impossible objects

Participants were tested on two visual mental rotation tasks using three-dimensional "possible" and "impossible" shapes. Both types of stimuli can be easily encoded by their parts and how they are spatially organized. However, while possible shapes can also be easily encoded as a global image, it is more difficult to encode impossible shapes in such a way. Participants visually rotated both types of stimuli at comparable rates, reflecting that local representations were used in the process of visual mental rotation.     

Central visual persistences: II. Effects of hand and head rotations

In an earlier paper, kinesthetic effects on central visual persistences (CPs) were reported, including the ability to move these images by hand following eye closure. While all CPs could be translated anywhere within the frontal field, the present report documents a more selective influence of manual rotations on CPs in the same subjects. When common objects or figures drawn on cards were rotated (while holding one end of the object or one corner of a card between thumb and forefinger), it was found that CPs of larger objects rotated with the hand. By contrast, CPs of smaller objects, parts of objects, and textures remained stable in space as the hand rotated. It is proposed that CPs of smaller stimuli and textures are represented mainly by the ventral stream (temporal cortex) while larger CPs, which rotate, are represented mainly by the dorsal stream (parietal cortex). A second discovery was that CPs of small objects (but not of line segments or textures) could be rotated when the thumb and fingers surrounded the edges of the object. It is proposed that neuronal convergence of visual and tactile information about shape increases parietal responses to small objects, so that their CPs will rotate. Experiments with CPs offer new tools to infer visual coding differences between ventral and dorsal streams in man.     

The Bourdon illusion in haptic space

A strong Bourdon illusion-the apparent bentness of a straight edge-in the active haptic mode was established in two experiments. Blindfolded subjects clasped the opposite surfaces of an object with the same frontal profile as the visual figure between thumb and forefinger and moved the latter together from end to end across the object. When the two triangular components of the objects were rotated through 180 degrees so that their bases instead of their apexes were contiguous, the illusion was reversed in direction and reduced by about half, as with the visual illusion. It was concluded that the active haptic Bourdon illusion is basically the same as the visual illusion and can be accounted for by a compromise in perception between the orientation of the test surface and that of the object of which it is an integral part.     

Object segmentation by top-down processes

In cluttered scenes, some object boundaries may not be marked by image cues. In such cases, the boundaries must be defined top-down as a result of object recognition. Here we ask if observers can retain the boundaries of several recognized objects in order to segment an unfamiliar object. We generated scenes consisting of neatly stacked objects, and the objects themselves consisted of neatly stacked coloured blocks. Because the blocks were stacked the same way within and across objects, there were no visual cues indicating which blocks belonged to which objects. Observers were trained to recognize several objects and we tested whether they could segment a novel object when it was surrounded by these familiar, studied objects. The observer's task was to count the number of blocks comprising the target object. We found that observers were able to accurately count the target blocks when the target was surrounded by up to four familiar objects. These results indicate that observers can use the boundaries of recognized objects in order to accurately segment, top-down, a novel object.     

The structure of three-dimensional object representations in human vision: evidence from whole-part matching

This article examines how the human visual system represents the shapes of 3-dimensional (3D) objects. One long-standing hypothesis is that object shapes are represented in terms of volumetric component parts and their spatial configuration. This hypothesis is examined in 3 experiments using a whole-part matching paradigm in which participants match object parts to whole novel 3D object shapes. Experiments 1 and 2, consistent with volumetric image segmentation, show that whole-part matching is faster for volumetric component parts than for either open or closed nonvolumetric regions of edge contour. However, the results of Experiment 3 show that an equivalent advantage is found for bounded regions of edge contour that correspond to object surfaces. The results are interpreted in terms of a surface-based model of 3D shape representation, which proposes edge-bounded 2-dimensional polygons as basic primitives of surface shape.     

The role of surface-based representations of shape in visual object recognition

This study contrasted the role of surfaces and volumetric shape primitives in three-dimensional object recognition. Observers (N?=?50) matched subsets of closed contour fragments, surfaces, or volumetric parts to whole novel objects during a whole–part matching task. Three factors were further manipulated: part viewpoint (either same or different between component parts and whole objects), surface occlusion (comparison parts contained either visible surfaces only, or a surface that was fully or partially occluded in the whole object), and target–distractor similarity. Similarity was varied in terms of systematic variation in nonaccidental (NAP) or metric (MP) properties of individual parts. Analysis of sensitivity (d′) showed a whole–part matching advantage for surface-based parts and volumes over closed contour fragments—but no benefit for volumetric parts over surfaces. We also found a performance cost in matching volumetric parts to wholes when the volumes showed surfaces that were occluded in the whole object. The same pattern was found for both same and different viewpoints, and regardless of target–distractor similarity. These findings challenge models in which recognition is mediated by volumetric part-based shape representations. Instead, we argue that the results are consistent with a surface-based model of high-level shape representation for recognition.     

Concavities as basic features in visual search: evidence from search asymmetries

Concave cusps and negative curvature minima play an important role in many theories of visual shape perception. Cusps and minima are taken to be part boundaries, used to segment an object into parts. Because of their important role in determining object structure and because there is some evidence that object structure is processed in parallel, it might be expected that concave cusps and negative curvature minima are processed preferentially. We tested this conjecture in several visual search experiments. Visual search for a target with a concave cusp among totally convex distractors yields nearly flat slopes (< 10 msec/item) for both present and absent trials. Reversing the roles of the target and the distractor results in inefficient search. The same asymmetry is found when the concave cusp is replaced by other types of concavity. We conclude, therefore, the concavities can serve as basic features in visual search experiments. This conclusion implies that the unit of selection in a visual search task is an object, rather than a location.     

Visual object agnosia as a problem in integrating parts and part relations

Current models of vision generally assume that the recognition of visual objects is achieved by encoding their component parts, as well as the spatial relations among parts. The current study examined how the processing of parts and their configurations may be affected in visual agnosia due to brain damage. Both a visual agnosic patient (AR) and healthy control subjects performed a visual search task in which they had to discriminate between targets and distractors that varied according to whether they shared their parts and/or their configuration. The results show that AR's visual search rates are disproportionally slow when targets and distractors share the same configuration than when they have different configurations. AR is also found to be disproportionately slow in discriminating targets and distractors that share identical parts when the targets and distractors share the same configuration. With differently configured targets and distractors, AR shows no part sharing effect. For controls, in contrast, the part and configuration sharing effects occur independently of one another. It is concluded that AR's object recognition deficit arises from difficulties in discriminating objects that share their configuration, and from an abnormal dependency of part information processing upon object configuration.     

Object-centred orienting of attention

Abstract

In the present study two experiments are reported in which the subjects were presented on a computer screen with a two-dimensional line drawing that is perceived as a three-dimensionalobject (i.e. a cube). The cube could be seen as stationary, as rotating about the y-axis (Experiment 1A), or as rotating about the x-axis (Experiment 1B). The subject's attention was directed by a visual precue to a vertex of the cube. As the cube rotated, the precued location moved in viewer-centred co-ordinatesm, but the local feature of the cube that had been precued (i.e. a given vertex) did not move in object-centred co-ordinates. The imperative stimulus was presented at the precued location (valid trials) or at an uncued location (invalid trials). Precued and uncued locations were determined in object-centred coordinates. The subjects were required to signal detection of the imperative stimulus by pressing the space bar on the computer keyboard. There were also control conditions in which the procedure was identical, but the cube was not visible. When the cube was visible, valid trials were faster than invalid trials, regardless of whether it was seen as stationary or rotating. It was concluded that the subjects could allocate spatial attention in object-centred co-ordinates. There wer also indications that responses for invalid trials were faster when the imperative stimulus was presented on the same face as the precue than when it was presented on the opposite face.     

Encoding color and shape from different parts of an object in visual short-term memory

Can we find an object-based encoding benefit in visual short-term memory (VSTM) when the features to be remembered are from different parts of an object? Using object parts defined by either figure-ground separation or negative minima of curvature, results from five experiments in which the visual change detection paradigm was used showed that the object-based encoding benefit in VSTM is modulated by how features are assigned to parts of an object: Features are best retained when the color and shape features to be remembered belong to the same part of an object. Although less well retained in comparison, features from different parts of an object are still better remembered than features from spatially separated objects. An object-based feature binding therefore exists even when the color and shape features to be remembered are from different parts of an object.     

Cross‐modal facilitation of mental rotation: Effects of modality and complexity

Performance in a visual mental rotation (MR) task has been reported to predict the ability to recognize retrograde‐transformed melodies. The current study investigated the effects of melodic structure on the MR of sequentially presented visual patterns. Each trial consisted of a five‐segment sequentially presented visual pattern (standard) followed by a five‐tone melody that was either identical in structure to the standard or its retrograde. A visual target pattern was either the rotated version of the standard or unrelated to it. The task was to indicate whether the target pattern was a rotated version of the standard or not. Periodic patterns were not rotated but melodies facilitated the rotation of non‐periodic patterns. For these, rotation latency was determined by a quantitative index of complexity (number of runs). This study provides the first experimental confirmation for cross‐modal facilitation of MR.     

实物形状的知觉相似性对视觉隐喻加工的影响

视觉隐喻是众多研究者比较关注的领域, 当前多以物体简笔画、漫画及广告等方面的研究为主, 但对与生活密切相关的实物的视觉隐喻研究甚少。因此研究采用掩蔽启动范式、李克特9点量表及语音生成等研究方法, 在物体概念类型(概念相同、概念相异)及形状类型(形状相似、形状相异)条件下, 针对实物图片的形状知觉相似性是否对视觉隐喻加工产生影响进行探讨。结果发现无论物体概念是否相同, 形状相似性对个体的物体识别具有“促进”作用, 并有利于视觉隐喻加工; 在物体识别中, 当两个物体概念不同时, 个体会试图建立一个特设类别来对形状相似的物体进行归类, 从而实现了视觉隐喻的加工。因此研究进一步证明了实物形状的知觉相似性有利于视觉隐喻的加工。