Nonaccidental properties underlie shape recognition in Mammalian and nonmammalian vision

An infinite number of 2D patterns on the retina can correspond to a single 3D object. How do visual systems resolve this ill-posed problem and recognize objects from only a few 2D retinal projections in varied exposure conditions? Theories of object recognition rely on the nonaccidental statistics of edge properties, mainly symmetry, collinearity, curvilinearity, and cotermination. These statistics are determined by the image-formation process (i.e., the 2D retinal projection of a 3D object ); their existence under a range of viewpoints enables viewpoint-invariant recognition. An important question in behavioral biology is whether the visual systems of nonmammalian animals have also evolved biases to utilize nonaccidental statistics . Here, we trained humans and pigeons to recognize four shapes. With the Bubbles technique, we determined which stimulus properties both species used to recognize the shapes. Both humans and pigeons used cotermination, the most diagnostic nonaccidental property of real-world objects, despite evidence from a model computer observer that cotermination was not the most diagnostic pictorial information in this particular task. This result reveals that a nonmammalian visual system that is different anatomically from the human visual system is also biased to recognize objects from nonaccidental statistics.     

Class similarity and viewpoint invariance in the recognition of 3D objects

 In human vision, the processes and the representations involved in identifying specific individuals are frequently assumed to be different from those used for basic level classification, because classification is largely viewpoint-invariant, but identification is not. This assumption was tested in psychophysical experiments, in which objective similarity between stimuli (and, consequently, the level of their distinction) varied in a controlled fashion. Subjects were trained to discriminate between two classes of computer-generated three-dimensional objects, one resembling monkeys and the other, dogs. Both classes were defined by the same set of 56 parameters, which encoded sizes, shapes, and placement of the limbs, ears, snout, etc. Interpolation between parameter vectors of the class prototypes yielded shapes that changed smoothly between monkey and dog. Within-class variation was induced in each trial by randomly perturbing all the parameters. After the subjects reached 90% correct performance on a fixed canonical view of each object, discrimination performance was tested for novel views that differed by up to 60° from the training view. In experiment 1 (in which the distribution of parameters in each class was unimodal) and in experiment 2 (bimodal classes), the stimuli differed only parametrically and consisted of the same geons (parts), yet were recognized virtually independently of viewpoint in the low-similarity condition. In experiment 3, the prototypes differed in their arrangement of geons, yet the subjects’ performance depended significantly on viewpoint in the high-similarity condition. In all three experiments, higher interstimulus similarity was associated with an increase in the mean error rate and, for misorientation of up to 45°, with an increase in the degree of viewpoint dependence. These results suggest that a geon-level difference between stimuli is neither strictly necessary nor always sufficient for viewpoint-invariant performance. Thus, basic and subordinate-level processes in visual recognition may be more closely related than previously thought. Received: 15 November 1993/Accepted in revised form: 14 July 1994     

Viewpoint-specific scene representations in human parahippocampal cortex

The "parahippocampal place area" (PPA) responds more strongly in functional magnetic resonance imaging (fMRI) to scenes than to faces, objects, or other visual stimuli. We used an event-related fMRI adaptation paradigm to test whether the PPA represents scenes in a viewpoint-specific or viewpoint-invariant manner. The PPA responded just as strongly to viewpoint changes that preserved intrinsic scene geometry as it did to complete scene changes, but less strongly to object changes within the scene. In contrast, lateral occipital cortex responded more strongly to object changes than to spatial changes. These results demonstrate that scene processing in the PPA is viewpoint specific and suggest that the PPA represents the relationship between the observer and the surfaces that define local space.     

Medial axis shape coding in macaque inferotemporal cortex

The basic, still unanswered question about visual object representation is this: what specific information is encoded by neural signals? Theorists have long predicted that neurons would encode medial axis or skeletal object shape, yet recent studies reveal instead neural coding of boundary or surface shape. Here, we addressed this theoretical/experimental disconnect, using adaptive shape sampling to demonstrate explicit coding of medial axis shape in high-level object cortex (macaque monkey inferotemporal cortex or IT). Our metric shape analyses revealed a coding continuum, along which most neurons represent a configuration of both medial axis and surface components. Thus, IT response functions embody a rich basis set for simultaneously representing skeletal and external shape of complex objects. This would be especially useful for representing biological shapes, which are often characterized by both complex, articulated skeletal structure and specific surface features.     

Bio-computational model of object-recognition: quantum Hebbian processing with neurally shaped Gabor wavelets

Theoretical and simulational evidence, as well as experimental indications, are accumulating that quantum associative memory and imaging are possible. We compare these data with biological evidence, since we find them to a significant extent compatible. This paper presents a computationally implementable integrative model of appearance-based viewpoint-invariant recognition of objects. The neuro-quantum hybrid model incorporates neural processing up to V1 and quantum associative processing in V1, achieving together an object-recognition result in V2 and ITC. Results of our simulation of the central quantum-like parts of the bio-model, receiving neurally pre-processed inputs, are presented. This part contains our original simulated storage by multiple quantum interference of image-encoding Gabor wavelets done in a Hebbian way, especially using the Griniasty et al. pose-sequence learning rule.     

A self-organizing multiple-view representation of 3D objects

We explore representation of 3D objects in which several distinct 2D views are stored for each object. We demonstrate the ability of a two-layer network of thresholded summation units to support such representations. Using unsupervised Hebbian relaxation, the network learned to recognize ten objects from different viewpoints. The training process led to the emergence of compact representations of the specific input views. When tested on novel views of the same objects, the network exhibited a substantial generalization capability. In simulated psychophysical experiments, the network's behavior was qualitatively similar to that of human subjects.     

Behavioral discrimination of water motions caused by moving objects.

We tested whether goldfish, Carassius auratus, discriminate hydrodynamic stimuli caused by moving objects. Blindfolded goldfish responded to a passing object with changes in inter-gill-movement intervals. To learn whether goldfish can discriminate water motions caused by different moving objects we habituated them to a certain object stimulus. If the stimulus was altered, e.g., by altering speed, direction of motion, or size or shape of the object, fish again showed a temporary suspension of breathing when the object passed by. If animals failed to respond to an altered stimulus, we paired this stimulus with a weak electric shock during training. Goldfish discriminated object motion direction. In addition, in two choice experiments goldfish discriminated water motions caused by objects which moved with different speeds (e.g., 5 cm s(-1) versus 6 cm s(-1)), or by objects which differed in size (e.g., 1 cm x 1 cm versus 1.4 cm x 1.4 cm cross section), or shape (e.g., a round versus a triangular object). If object size and/or shape was varied quasi-randomly such that the faster moving object not always caused the greatest water velocities, fish still discriminated object speed.     

The three-dimensional structure of trypsin-treated Staphylococcus aureus alpha-toxin.

Trypsin treatment of staphylococcal alpha-toxin cleaves the molecule into two roughly equally sized parts, which results in inactivation of the toxin. Tetragonal arrays of oligomers, closely resembling the native ones, can however be formed on lipid layers. From tilted views of negatively stained crystals a 3D structure to 23 A resolution has been determined by electron microscopy and image processing. On comparison with the 3D structure of the native alpha-toxin (Olofsson et al., J. Mol. Biol. 214, 299-306, 1990) the subdomains are more separated, confirming the differences found when comparing the projection maps (Olofsson et al., J. Struct. Biol. 106, 199-204, 1991). The tryptic cleavage takes place in a postulated hinge region. The results are consistent with the hypothesis that the conformational change required for inducing the membrane permeabilizing property takes place in this region. Furthermore, we present a refined projection map at approximately 10 A resolution based on the analysis of a large number of crystals using unbending methods.     

Neural representation of objects in space: a dual coding account.

I present evidence on the nature of object coding in the brain and discuss the implications of this coding for models of visual selective attention. Neuropsychological studies of task-based constraints on: (i) visual neglect; and (ii) reading and counting, reveal the existence of parallel forms of spatial representation for objects: within-object representations, where elements are coded as parts of objects, and between-object representations, where elements are coded as independent objects. Aside from these spatial codes for objects, however, the coding of visual space is limited. We are extremely poor at remembering small spatial displacements across eye movements, indicating (at best) impoverished coding of spatial position per se. Also, effects of element separation on spatial extinction can be eliminated by filling the space with an occluding object, indicating that spatial effects on visual selection are moderated by object coding. Overall, there are separate limits on visual processing reflecting: (i) the competition to code parts within objects; (ii) the small number of independent objects that can be coded in parallel; and (iii) task-based selection of whether within- or between-object codes determine behaviour. Between-object coding may be linked to the dorsal visual system while parallel coding of parts within objects takes place in the ventral system, although there may additionally be some dorsal involvement either when attention must be shifted within objects or when explicit spatial coding of parts is necessary for object identification.     

The interactive presentation of 3D information obtained from reconstructed datasets and 3D placement of single histological sections with the 3D portable document format

Interpretation of the results of anatomical and embryological studies relies heavily on proper visualization of complex morphogenetic processes and patterns of gene expression in a three-dimensional (3D) context. However, reconstruction of complete 3D datasets is time consuming and often researchers study only a few sections. To help in understanding the resulting 2D data we developed a program (TRACTS) that places such arbitrary histological sections into a high-resolution 3D model of the developing heart. The program places sections correctly, robustly and as precisely as the best of the fits achieved by five morphology experts. Dissemination of 3D data is severely hampered by the 2D medium of print publication. Many insights gained from studying the 3D object are very hard to convey using 2D images and are consequently lost or cannot be verified independently. It is possible to embed 3D objects into a pdf document, which is a format widely used for the distribution of scientific papers. Using the freeware program Adobe Reader to interact with these 3D objects is reasonably straightforward; creating such objects is not. We have developed a protocol that describes, step by step, how 3D objects can be embedded into a pdf document. Both the use of TRACTS and the inclusion of 3D objects in pdf documents can help in the interpretation of 2D and 3D data, and will thus optimize communication on morphological issues in developmental biology.     

A Representational Similarity Analysis of the Dynamics of Object Processing Using Single-Trial EEG Classification

The recognition of object categories is effortlessly accomplished in everyday life, yet its neural underpinnings remain not fully understood. In this electroencephalography (EEG) study, we used single-trial classification to perform a Representational Similarity Analysis (RSA) of categorical representation of objects in human visual cortex. Brain responses were recorded while participants viewed a set of 72 photographs of objects with a planned category structure. The Representational Dissimilarity Matrix (RDM) used for RSA was derived from confusions of a linear classifier operating on single EEG trials. In contrast to past studies, which used pairwise correlation or classification to derive the RDM, we used confusion matrices from multi-class classifications, which provided novel self-similarity measures that were used to derive the overall size of the representational space. We additionally performed classifications on subsets of the brain response in order to identify spatial and temporal EEG components that best discriminated object categories and exemplars. Results from category-level classifications revealed that brain responses to images of human faces formed the most distinct category, while responses to images from the two inanimate categories formed a single category cluster. Exemplar-level classifications produced a broadly similar category structure, as well as sub-clusters corresponding to natural language categories. Spatiotemporal components of the brain response that differentiated exemplars within a category were found to differ from those implicated in differentiating between categories. Our results show that a classification approach can be successfully applied to single-trial scalp-recorded EEG to recover fine-grained object category structure, as well as to identify interpretable spatiotemporal components underlying object processing. Finally, object category can be decoded from purely temporal information recorded at single electrodes.     

Spatial aspects of spontaneous object grouping by young chimpanzees(Pan troglodytes)

I report spontaneous spatial object grouping in five chimpanzees (Pan troglodytes)ranging from 1 to 4 years of age. I recorded subjects' spontaneous interactions with groups of objects, noting the spatial relations in the groupings and the constructive processes that the subjects adopted. Though one subject developed vertical stacks, none of them realized horizontal alignments or spatial correspondences between groups. All subjects showed consistent trends in their constructive processes toward manipulating objects in relation to the body: they increasingly manipulated objects globally, that is to say, they moved or held objects together with the same part of the body. They also increasingly placed objects on the body or in correspondence to their symmetrical body parts. The increasing importance assigned by chimpanzees to body- object relations over object- object relations contrasts with human developmental trends in the domain of manipulative space.     

Three-dimensional computer reconstructions from serial sections of cell nuclei

The analysis of ultrathin serial sections as 3-dimensional (3D) information requires interpretation and display of a large amount of data. We suggest a simple way to solve this problem; it permits presentation of a series of sections as a 3D color image of good quality. It involves a picture system with specialized hardware and software written for this purpose. 3D images of cellular organelles have been drawn either by manually defining the contour of the objects or by thresholding of the volumes in the structures. These 2 methods allow rapid drawing of the image on the screen. It is possible to determine the position, shape and size of 3D structures. This interactive system allows the user to choose between several options: colors, removal of parts of the object, and cutout.     

Viewer-centered object representation in the human visual system revealed by viewpoint aftereffects

Are there neurons representing specific views of objects in the human visual system? A visual selective adaptation method was used to address this question. After visual adaptation to an object viewed either 15 or 30 degrees from one side, when the same object was subsequently presented near the frontal view, the perceived viewing directions were biased in a direction opposite to that of the adapted viewpoint. This aftereffect can be obtained with spatially nonoverlapping adapting and test stimuli, and it depends on the global representation of the adapting stimuli. Viewpoint aftereffects were found within, but not across, categories of objects tested (faces, cars, wire-like objects). The magnitude of this aftereffect depends on the angular difference between the adapting and test viewing angles and grows with increasing duration of adaptation. These results support the existence of object-selective neurons tuned to specific viewing angles in the human visual system.     

Anterior regions of monkey parietal cortex process visual 3D shape

The intraparietal cortex is involved in the control of visually guided actions, like reach-to-grasp movements, which require extracting the 3D shape and position of objects from 2D retinal images. Using fMRI in behaving monkeys, we investigated the role of the intraparietal cortex in processing stereoscopic information for recovering the depth structure and the position in depth of objects. We found that while several areas (CIP, LIP, and AIP on the lateral bank; PIP and MIP on the medial bank) are activated by stereoscopic stimuli, AIP and an adjoining portion of LIP are sensitive only to depth structure. Furthermore, only these two regions are sensitive to both the depth structure and the 2D shape of small objects. These results indicate that extracting 3D spatial information from stereo involves several intraparietal areas, among which AIP and anterior LIP are more specifically engaged in extracting the 3D shape of objects.     

Location-specific Responsiveness to Environmental Perturbations in Wedge-capped Capuchins (Cebus olivaceus)

We recorded the responses of the members of a captive group of wedge-capped capuchins to novel and familiar objects placed in different parts of their cage in a study of the spatial dependency of activity with objects. We focused on behavioral pattern variability across subjects and across object location. Results show that, according to the location of the object, a great deal of within-subject response variability exists. The dominant male was slow to interact physically with objects and presented social-like behaviors—essentially grooming—towards objects in only one site. Implicit to the ethological approach is the assumption that consistent spatial location is irrelevant or, at best, of little importance to the definition of stimuli. Nevertheless, stimuli would be best considered as perturbations insofar as the significance of an object or event depends on where and when it is encountered. In order to evaluate how monkey cognition operates, it seems essential to investigate the role of the primate's own spatial structure. As a working hypothesis, we introduce the processes of spatial facilitation and inhibition and suggest that they affect how an individual interacts with objects and events.     

Colour Terms Affect Detection of Colour and Colour-Associated Objects Suppressed from Visual Awareness

The idea that language can affect how we see the world continues to create controversy. A potentially important study in this field has shown that when an object is suppressed from visual awareness using continuous flash suppression (a form of binocular rivalry), detection of the object is differently affected by a preceding word prime depending on whether the prime matches or does not match the object. This may suggest that language can affect early stages of vision. We replicated this paradigm and further investigated whether colour terms likewise influence the detection of colours or colour-associated object images suppressed from visual awareness by continuous flash suppression. This method presents rapidly changing visual noise to one eye while the target stimulus is presented to the other. It has been shown to delay conscious perception of a target for up to several minutes. In Experiment 1 we presented greyscale photos of objects. They were either preceded by a congruent object label, an incongruent label, or white noise. Detection sensitivity (d’) and hit rates were significantly poorer for suppressed objects preceded by an incongruent label compared to a congruent label or noise. In Experiment 2, targets were coloured discs preceded by a colour term. Detection sensitivity was significantly worse for suppressed colour patches preceded by an incongruent colour term as compared to a congruent term or white noise. In Experiment 3 targets were suppressed greyscale object images preceded by an auditory presentation of a colour term. On congruent trials the colour term matched the object’s stereotypical colour and on incongruent trials the colour term mismatched. Detection sensitivity was significantly poorer on incongruent trials than congruent trials. Overall, these findings suggest that colour terms affect awareness of coloured stimuli and colour- associated objects, and provide new evidence for language-perception interaction in the brain.     

A biplanar reconstruction method based on 2D and 3D contours: application to the distal femur

A three-dimensional (3D) reconstruction algorithm based on contours identification from biplanar radiographs is presented. It requires, as technical prerequisites, a method to calibrate the biplanar radiographic environment and a surface generic object (anatomic atlas model) representing the structure to be reconstructed. The reconstruction steps consist of: the definition of anatomical regions, the identification of 2D contours associated to these regions, the calculation of 3D contours and projection onto the radiographs, the associations between points of the X-rays contours and points of the projected 3D contours, the optimization of the initial solution and the optimized object deformation to minimize the distance between X-rays contours and projected 3D contours. The evaluation was performed on 8 distal femurs comparing the 3D models obtained to CT-scan reconstructions. Mean error for each distal femur was 1 mm.