Shape Integration for Visual Object Recognition and Its Implication in Category-Specific Visual Agnosia

A series of experiments was conducted on a patient (ELM) with bilateral inferior temporal lobe damage and category-specific visual agnosia in order to specify the nature of his functional impairment. In Experiment 1, ELM performed a task of picture/word matching that used line drawings of fruits and vegetables as stimuli. The pattern of confusions exhibited by the patient suggested a failure in processing the full range of shape features necessary for the unique specification of the target relative to other structurally related items. This hypothesis of a shape integration impairment was tested and verified by subsequent visual recognition experiments (Experiments 2-4), which used synthetic stimuli with shapes precisely defined on the dimensions of elongation, curvature, and tapering. Furthermore, it was determined (Experiment 5) that the integration deficit is specific to the retrieval of shape knowledge from memory and does not affect the encoding of the properties of visual stimuli. It is argued that these findings have critical implications for cognitive theories of visual object recognition and for an interpretation of the visual function of the inferior temporal cortex. Finally, it was shown that the patient's deficit for structural knowledge integration is modulated by the semantic properties of the objects (Experiment 6), thereby demonstrating the applicability of the present findings to an explanation of category-specific visual agnosia.     

Understanding the structural determinants of object confusion in memory: an assessment of psychophysical approaches to estimating visual similarity

Computer-generated shapes varying on visual dimensions such as curvature, tapering, and thickness have been used to investigate identification deficits in the category-specific visual agnosia (CSVA) Patient E.L.M.. However, whether the implemented variations on each of these dimensions were perceived by novice observers as "similar amounts of change" is unknown. To estimate distance in psychophysical shape space, sets of shapes were developed using two different scaling methods--an objective method based on visual search, and a subjective method based on judgments of similarity--and a third approach that did not involve scaling. How well each method estimated psychophysical shape space was assessed by measuring the confusions within memory among the shapes. The results suggested that, although neither of the approaches perfectly reflected psychophysical shape space, subjective scaling was a better estimator of distance in psychophysical shape space than were other approaches. The number of confusions produced on each set of shapes was used to develop a new set of shapes that accurately estimated distance in psychophysical shape space. These results suggest that a combination of approaches is preferable in order to accurately estimate distance in psychophysical shape space.     

The eyes grasp,the hands see: Metric category knowledge transfers between vision and touch

Categorization of seen objects is often determined by the shapes of objects. However, shape is not exclusive to the visual modality: The haptic system also is expert at identifying shapes. Hence, an important question for understanding shape processing is whether humans store separate modality-dependent shape representations, or whether information is integrated into one multisensory representation. To answer this question, we created a metric space of computer-generated novel objects varying in shape. These objects were then printed using a 3-D printer, to generate tangible stimuli. In a categorization experiment, participants first explored the objects visually and haptically. We found that both modalities led to highly similar categorization behavior. Next, participants were trained either visually or haptically on shape categories within the metric space. As expected, visual training increased visual performance, and haptic training increased haptic performance. Importantly, however, we found that visual training also improved haptic performance, and vice versa. Two additional experiments showed that the location of the categorical boundary in the metric space also transferred across modalities, as did heightened discriminability of objects adjacent to the boundary. This observed transfer of metric category knowledge across modalities indicates that visual and haptic forms of shape information are integrated into a shared multisensory representation.     

Behavioral evidence for task-dependent "what" versus "where" processing within and across modalities

Task-dependent information processing for the purpose of recognition or spatial perception is considered a principle common to all the main sensory modalities. Using a dual-task interference paradigm, we investigated the behavioral effects of independent information processing for shape identificationand localization ofobject features within and across vision and touch. In Experiment 1, we established that color and texture processing (i.e., a "what" task) interfered with both visual and haptic shape-matching tasks and that mirror image and rotation matching (i.e., a "where" task) interfered with a feature-location-matching task in both modalities. In contrast, interference was reduced when a "where" interference task was embedded in a "what" primary task and vice versa. In Experiment 2, we replicated this finding within each modality, using the same interference and primary tasks throughout. In Experiment 3, the interference tasks were always conducted in a modality other than the primary task modality. Here, we found that resources for identification and spatial localization are independent of modality. Our findings further suggest that multisensory resources for shape recognition also involve resources for spatial localization. These results extend recent neuropsychological and neuroimaging findings and have important implications for our understanding of high-level information processing across the human sensory systems.     

Categories in the pigeon brain: A reverse engineering approach

Pigeons are well known for their visual capabilities as well as their ability to categorize visual stimuli at both the basic and superordinate level. We adopt a reverse engineering approach to study categorization learning: Instead of training pigeons on predefined categories, we simply present stimuli and analyze neural output in search of categorical clustering on a solely neural level. We presented artificial stimuli, pictorial and grating stimuli, to pigeons without the need of any differential behavioral responding while recording from the nidopallium frontolaterale (NFL), a higher visual area in the avian brain. The pictorial stimuli differed in color and shape; the gratings differed in spatial frequency and amplitude. We computed representational dissimilarity matrices to reveal categorical clustering based on both neural data and pecking behavior. Based on neural output of the NFL, pictorial and grating stimuli were differentially represented in the brain. Pecking behavior showed a similar pattern, but to a lesser extent. A further subclustering within pictorial stimuli according to color and shape, and within gratings according to frequency and amplitude, was not present. Our study gives proof‐of‐concept that this reverse engineering approach—namely reading out categorical information from neural data—can be quite helpful in understanding the neural underpinnings of categorization learning.     

Category learning in rhesus monkeys: a study of the Shepard, Hovland, and Jenkins (1961) tasks

In influential research, R. N. Shepard, C. I. Hovland, and H. M. Jenkins (1961) surveyed humans' categorization abilities using tasks based in rules, exclusive-or (XOR) relations, and exemplar memorization. Humans' performance was poorly predicted by cue-conditioning or stimulus-generalization theories, causing Shepard et al. to describe it in terms of hypothesis selection and rule application that were possibly supported by verbal mediation. The authors of the current article surveyed monkeys' categorization abilities similarly. Monkeys, like humans, found category tasks with a single relevant dimension the easiest and perceptually chaotic tasks requiring exemplar memorization the most difficult. Monkeys, unlike humans, found tasks based in XOR relations very difficult. The authors discuss the character and basis of the species difference in categorization and consider whether monkeys are the generalization-based cognitive system that humans are not.     

Blinded by the accent! The minor role of looks in ethnic categorization

The categories that social targets belong to are often activated automatically. Most studies investigating social categorization have used visual stimuli or verbal labels, whereas ethnolinguistic identity theory posits that language is an essential dimension of ethnic identity. Language should therefore be used for social categorization. In 2 experiments, using the "Who Said What?" paradigm, the authors investigated social categorization by using accents (auditory stimuli) and looks (visual stimuli) to indicate ethnicity, either separately or in combination. Given either looks or accents only, the authors demonstrated that ethnic categorization can be based on accents, and the authors found a similar degree of ethnic categorization by accents and looks. When ethnic cues of looks and accents were combined by creating cross categories, there was a clear predominance of accents as meaningful cues for categorization, as shown in the respective parameters of a multinomial model. The present findings are discussed with regard to the generalizability of findings using one channel of presentation (e.g., visual) and the asymmetry found with different presentation channels for the category ethnicity.     

Hemispheric predominance assessment of phonology and semantics: a divided visual field experiment

The aim of the present behavioural experiment was to evaluate the most lateralized among two phonological (phoneme vs. rhyme detection) and the most lateralized among two semantic ("living" vs. "edible" categorization) tasks, within the dominant hemisphere for language. The reason of addressing this question was a practical one: to evaluate the degree of the hemispheric lateralization for several language tasks, by using the divided visual presentation of stimuli, and then choose the most lateralized semantic and phonological for mapping language in patients by using fMRI in future studies. During the divided visual field experiment by using words (semantic tasks) and pseudo-words (phonological tasks) as stimuli, thirty-nine right-handed participants were examined. Our results have shown that all tasks were significantly left hemisphere lateralized. Furthermore, the rhyme was significantly more lateralized than phoneme detection and "living" was significantly more lateralized than "edible" categorization. The rhyme decision and "living" categorization will be used in future fMRI studies for assessing hemispheric predominance and cerebral substrate for semantics and phonology in patients. Our results also suggest that the characteristics of stimuli could influence the degree of the hemispheric lateralization (i.e., the emotional charge of stimuli for words and the position of the phoneme to be detected, for pseudo-words).     

Strategies of semantic categorization in the cerebral hemispheres

Strategies of semantic categorization in intact cerebral hemispheres were studied in two experiments by presenting names of typical and atypical category instances to the left visual field (LVF) (right hemisphere) or to the right visual field (RVF) (left hemisphere). The results revealed that the typicality of instances had a large effect on categorization times in the LVF in both experiments, suggesting that the right hemisphere relies strongly on a holistic, similarity-based comparison strategy. In Experiment 1, the typicality effect was weaker in the RVF than in the LVF. In Experiment 2, a typicality effect in the RVF was observed for the "four-footed animal" category but not for the "bird" category. The hypothesis that the left hemisphere employs a strategy based on defining or necessary features is not supported by the observed typicality effect in the "four-footed animal" category. Instead, it is suggested that the left hemisphere may be able to categorize on the basis of prestored instance-category knowledge. When such knowledge is not available (e.g., as for four-footed animals), a similarity-based comparison strategy is employed by the left hemisphere.     

Surface geometry influences the shape of illusory contours

Geometric and neural models of illusory-contour (IC) synthesis currently use only local contour geometry to derive the shape of ICs. Work on the visual representation of shape, by contrast, points to the importance of both contour and surface geometry. We investigated the influence of surface-based geometric factors on IC shape. The local geometry of inducing-contour pairs was equated in stereoscopic IC displays, and the shape of the enclosed surface was varied by manipulating sign of curvature, cross-axial shape width, and medial-axis geometry. IC shapes were measured using a parametric shape-adjustment task (Experiment 1) and a dot-adjustment task (Experiment 2). Both methods revealed large influences of surface geometry. ICs enclosing locally concave regions were perceived to be systematically more angular than those enclosing locally convex regions. Importantly, the influence of sign of curvature was modulated significantly by shape width and medial-axis geometry: IC shape difference between convex and concave inducers was greater for narrow shapes than wider ones, and greater for shapes with straight axis and symmetric contours (diamond versus bowtie), than those with curved axis and parallel contours (bent tubes). Even at the level of illusory "contours," there is a contribution of region-based geometry which is sensitive to nonlocal shape properties involving medial geometry and part decomposition. Models of IC synthesis must incorporate the role of nonlocal region-based geometric factors in a way that parallels their role in organizing visual shape representation more generally.     

Widgets: A new set of parametrically defined 3D objects for use in haptic and visual categorization tasks

IntroductionMost research to date on human categorization ability has concentrated on the visual and auditory domains. However, a limited – but non-negligible – range of studies has also examined the categorization of familiar or unfamiliar (i.e., novel) objects in the haptic (i.e., tactile-kinesthetic) modality.ObjectiveIn this paper, we describe how we developed a new set of parametrically defined objects, called widgets, that can be used as 3D (or 2D) materials for haptic (or visual) categorization purposes.MethodWidgets are unfamiliar complex 3D shapes with an ovoid body and four types of elements attached to it (eyes, tail, crest, and legs). The stimulus set comprises 24 objects divided into four categories of six exemplars each (the files used for 3D printing are provided as Supplementary Material).ResultsWe also assessed and demonstrated the validity of our stimulus set by conducting two separate studies of haptic and visual categorization, involving participants of different ages: young adults (Study 1), and children and adolescents (Study 2). Results showed that humans can categorize our 3D complex shapes on the basis of both haptically and visually perceived similarities in shape attributes.ConclusionWidgets are very useful new experimental stimuli for categorization studies using 3D printing technology.     

Apparent motion between shapes differing in location and orientation: a window technique for estimating path curvature

When a shape is alternately presented in two positions differing in both location and orientation, apparent motion tends to be experienced over a curved path. The curvature provides evidence about principles of object motion that may have been internalized in the perceptual system. This study introduces a technique for estimating deviation from a straight path. A shape was alternately presented on the two sides of a visual partition with a "window" just wide enough to accommodate the shape. Observers adjusted the location of the window to maximize the illusion of smooth passage of the shape through the window. In accordance with theoretical expectations, estimated deviations from rectilinear motion increased with the separation between the stimuli in spatial location, angular orientation, and time.     

Language-mediated object categorization: A longitudinal study with 16- to 20-month-old Serbian-speaking children

This study addressed the development of language-mediated categorization longitudinally, on a sample of Serbian-speaking children aged 16–20 months. Twenty four children were tested 3 consecutive times on: categorization of unknown objects with a shared label and a similar shape; and categorization of unknown objects with a shared label, but no visual similarity. The task was embedded in a game-like activity which included real object manipulation, categorization of novel objects labeled with made-up names, and socio-pragmatic cues provided by the experimenter. Our results indicate that an average child develops the ability to categorize novel objects with a similar shape and a shared label around 16 months of age, whereas for novel objects with a shared label, but no visual similarity, this ability is acquired sometime between 18 and 20 months of age. Our results partly confirm research findings obtained in other languages, thus emphasizing the need for cross-linguistic and cross-cultural validation of all language-related developmental studies.     

The influence of flankers on race categorization of faces

Context affects multiple cognitive and perceptual processes. In the present study, we asked how the context of a set of faces would affect the perception of a target face??s race in two distinct tasks. In Experiments 1 and 2, participants categorized target faces according to perceived racial category (Black or White). In Experiment 1, the target face was presented alone or with Black or White flanker faces. The orientation of flanker faces was also manipulated to investigate how face inversion effect would interact with the influences of flanker faces on the target face. The results showed that participants were more likely to categorize the target face as White when it was surrounded by inverted White faces (an assimilation effect). Experiment 2 further examined how different aspects of the visual context would affect the perception of the target face by manipulating flanker faces?? shape and pigmentation, as well as their orientation. The results showed that flanker faces?? shape and pigmentation affected the perception of the target face differently. While shape elicited a contrast effect, pigmentation appeared to be assimilative. These novel findings suggest that the perceived race of a face is modulated by the appearance of other faces and their distinct shape and pigmentation properties. However, the contrast and assimilation effects elicited by flanker faces?? shape and pigmentation may be specific to race categorization, since the same stimuli used in a delayed matching task (Experiment 3) revealed that flanker pigmentation induced a contrast effect on the perception of target pigmentation.     

The effect of category learning on the representation of shape: dimensions can be biased but not differentiated

Recent studies have suggested a profound influence of category learning on visual perception, resulting in independent processing of previously integral dimensions. The authors reinvestigate this issue for shape dimensions. They first extend previous findings that some shape dimensions (aspect ratio and curvature) are processed in a separable way, whereas others (radial frequency components) are not. They then show that a category-learning phase improved the discrimination of a relevant with respect to an irrelevant dimension, but only for separable dimensions. No similar effect was found on the relative sensitivity for integral shape dimensions. Thus, category learning is capable of biasing separable shape dimensions but does not alter the status of dimensions in the visual system as either separable or integral.     

An element in the visual field is just a conjunction of attributes: A critique of Bundesen (1991)

Nissen (1985)presented results showing that in partial-report experiments, selection by location was different from selection by color or by shape. Bundesen (1991) presented a reinterpretation of these results, claiming that they can also be explained by a model in which location selection is treated on a par with selection by color or by shape. We will show that the analysis given by Bundesen is based on a special relation between an “element” in the visual field and its location, which in fact reduces the expressions given by Bundesen to those given by Nissen. Furthermore, we will show that in Bundesen’s theory, it is assumed that in the process of perceptual categorization an element in the visual field can be referred to without the use of one of its attributes. This assumption is untenable from a theoretical point of view. In a1l the results presented by Nissen can still be regarded as evidence for the difference between selection by location and selection by color or by shape.     

Reliance on visual and verbal information across ontological kinds: What do children know about animals and machines?

In two experiments, we systematically examined the reliance on visual (external shape and features) and verbal (origins and internal structure) information in isolation, and together in the identification of animals and machines by 3-, 4-, and 5-year-olds, and adults. Experiment 1 examined the use of visual and verbal information independently in a visual classification task, a verbal classification task, and an induction task. Experiment 2 examined the relative weighting of visual and verbal information in an induction task and a categorization task. The three most important findings from Experiment 1 were that (a) children and adults can use either visual or verbal information to distinguish animals from machines; (b) all age groups classified items with mixed visual information as machines, a tendency that increased with age; and (c) with age, children became increasingly able to induce non-obvious properties, especially the non-obvious properties of machines. The findings from Experiment 2 indicate that the youngest and oldest participants relied on both visual and verbal information in the identification of animals and machines in categorization and induction tasks. Five-year-olds, however, relied only on visual information. As in Experiment 1, we observed a tendency to judge items with contrasting information as machines, suggesting that individuals utilize a more strict definition (both visually and verbally) for the category of animals. We discuss the implication of these results with respect to developmental differences in the use of perceptual and conceptual information across the ontological distinction between artifacts and natural kinds.     

Effects of category learning on the stimulus selectivity of macaque inferior temporal neurons

Primates can learn to categorize complex shapes, but as yet it is unclear how this categorization learning affects the representation of shape in visual cortex. Previous studies that have examined the effect of categorization learning on shape representation in the macaque inferior temporal (IT) cortex have produced diverse and conflicting results that are difficult to interpret owing to inadequacies in design. The present study overcomes these issues by recording IT responses before and after categorization learning. We used parameterized shapes that varied along two shape dimensions. Monkeys were extensively trained to categorize the shapes along one of the two dimensions. Unlike previous studies, our paradigm counterbalanced the relevant categorization dimension across animals. We found that categorization learning increased selectivity specifically for the category-relevant stimulus dimension (i.e., an expanded representation of the trained dimension), and that the ratio of within-category response similarities to between-category response similarities increased for the relevant dimension (i.e., category tuning). These small effects were only evident when the learned category-related effects were disentangled from the prelearned stimulus selectivity. These results suggest that shape-categorization learning can induce minor category-related changes in the shape tuning of IT neurons in adults, suggesting that learned, category-related changes in neuronal response mainly occur downstream from IT.     

What does the occluding contour tell us about solid shape?

A new theorem is discussed that relates the apparent curvature of the occluding contour of a visual shape to the intrinsic curvature of the surface and the radial curvature. This theorem allows the formulation of general laws for the apparent curvature, independent of viewing distance and regardless of the fact that the rim (the boundary between the visible and invisible parts of the object) is a general, thus twisted, space curve. Consequently convexities, concavities, or inflextions of contours in the retinal image allow the observer to draw inferences about local surface geometry with certainty. These results appear to be counterintuitive, witness to the treatment of the problem by recent authors. It is demonstrated how well-known examples, used to show how concavities and convexities of the contour have no obvious relation to solid shape, are actually good illustrations of the fact that convexities are due to local ovoid shapes, concavities to local saddle shapes.     

Psychological Groups and Political Psychology: A Response to Huddy's "Critical Examination of Social Identity Theory"

In a recent article in this journal, Leonie Huddy (2001) asks whether the social identity approach developed by Tajfel, Turner, and their collaborators can "advance the study of identity within political science" (p. 128). She concludes that "various shortcomings and omissions in its research program" (p. 128) hinder the application of the approach to political phenomena. This paper presents a response to Huddy's evaluation of the social identity approach. Several aspects of her account of social identity work are challenged, especially her suggestion that it ignores subjective aspects of group membership. The interpretation of the minimal group paradigm is discussed in detail, as are issues of identity choice, salience, and variations in identity strength. The treatment of groups as process in social identity theory and self–categorization theory is given particular emphasis.