Search

Object recognition is one of the most important functions of the human visual system, yet one of the least understood, this despite the fact that vision is certainly the most studied function of the brain. We understand relatively well how several processes in the cortical visual areas that support recognition capabilities take place, such as orientation discrimination and color constancy. This paper proposes a model of the development of object recognition capability, based on two main theoretical principles. The first is that recognition does not imply any sort of geometrical reconstruction, it is instead fully driven by the two dimensional view captured by the retina. The second assumption is that all the processing functions involved in recognition are not genetically determined or hardwired in neural circuits, but are the result of interactions between epigenetic influences and basic neural plasticity mechanisms. The model is organized in modules roughly related to the main visual biological areas, and is implemented mainly using the LISSOM architecture, a recent neural self-organizing map model that simulates the effects of intercortical lateral connections. This paper shows how recognition capabilities, similar to those found in brain ventral visual areas, can develop spontaneously by exposure to natural images in an artificial cortical model.

The paper describes a new adaptive neural network for edge-based pattern extraction problems. In the model, each neuron represents an edge with continuous state variables describing its location and orientation. The post-synaptic distribution is dependent on the state variables. Each neuron adjusts its state to increase its membrane potential, which results in highly adaptive dynamics of the synaptic weight distribution. The network allocates multiple neurons with different orientation modes for each edge. The strategy allows accurate modeling of multi-modal distributions at key-points such as corners and junctions. As a result, the network delineates edges at sub-pixel accuracy while preserving key-points. It is also capable of processing a sequence of images and following moving objects. The network is extended for contour integration and key-point detection tasks. The paper presents experiments conducted on both synthetic and non-synthetic data to demonstrate the effectiveness of the technique.

This paper uses visual, empirical and formal methods (Li & Zaidi, Vision Research, 40 (2000) 217; Li & Zaidi, Vision Research, 41 (22) (2001a) 2927) to examine the roles of oriented texture components in conveying veridical percepts of concave and convex surfaces that are pitched towards or away from the observer. The results show that pairs of components, oriented symmetrically around the axis of maximum curvature, combine to provide the geodesic orientation modulations that are critical for veridical shape perception. The degree of pitch determines the orientations of the critical pair of components. Perspective is crucial to the veridical perception of concavities and convexities, regardless of the degree of pitch. The results of this paper reconfirm that veridical shape perception depends on extracting critical patterns of oriented energy, but also show that the class of textures capable of conveying veridical percepts of developable shapes in general views is even more restricted than that identified by Li and Zaidi (Journal of Optical Society of America A, 18 (2001b), 2430).

The modern study of perception began when Fechner published his ‘Elements of Psychophysics’ in 1860. This book has guided most perception research ever since. It has become increasingly clear that there are problems with Fechner's approach, which assumes that the percept is completely determined by the sensory input. Fechner's approach cannot explain the processes that allow our percepts to be veridical. Post-Fechnerian schools (Helmholtzian, Structural, Gestalt and Gibsonian) have tried to deal with this problem, but have not been successful. An alternative to the Fechnerian approach is required. This paper describes an alternative that has been developing over the last 20 years within the computer vision community. It treats perceptual interpretation as a solution of an inverse problem that depends critically on the operation of a priori constraints. Contemporary research, which adopted this approach, has concentrated on verifying the usefulness of Bayesian and standard regularization methods. This paper takes the next step; it discusses theoretical and empirical aspects of studying human perception as an inverse problem. It reviews the literature that illustrates the power of the inverse problem approach. This review leads to the suggestion that progress in the study of perception will benefit if the inverse approach were to be adopted by experimentalists, as well as by the computational modelers, who have been actively exploring its potential to date.

A patient with severe left hemi-inattention was found to be influenced by neglected stimuli in free drawing tasks. He was presented 64 sheets of paper: 32 had geometric figures on the left side and 32 on the right side. The patient was asked what he saw and invited to draw a subject of his choice on the sheet of paper. The patient always denied seeing anything when the figures were on the left side, while he identified the figures on the right side. Four judges examined the patient's drawings and all of them agreed that 11 drawings were influenced by neglected stimuli and none by non-neglected ones. The drawings he made were similar but not the exact reproduction of neglected figures. This finding confirms that visual processing can proceed without awareness in a patient with hemi-inattention.

Many quantitative and qualitative studies in geoscience research are based on digital elevation models (DEMs) and 3D surfaces to aid understanding of natural and anthropogenically-influenced topography. As well as their quantitative uses, the visual representation of DEMs can add valuable information for identifying and interpreting topographic features. However, choice of viewpoints and rendering styles may not always be intuitive, especially when terrain data are augmented with digital image texture. In this paper, an information-theoretic framework for object understanding is applied to terrain visualization and terrain view selection. From a visibility channel between a set of viewpoints and the component polygons of a 3D terrain model, we obtain three polygonal information measures. These measures are used to visualize the information associated with each polygon of the terrain model. In order to enhance the perception of the terrain's shape, we explore the effect of combining the calculated information measures with the supplementary digital image texture. From polygonal information, we also introduce a method to select a set of representative views of the terrain model. Finally, we evaluate the behaviour of the proposed techniques using example datasets. A publicly available framework for both the visualization and the view selection of a terrain has been created in order to provide the possibility to analyse any terrain model. HighlightsInformation theory framework for terrain visualization.Polygonal information is used to enhance the terrain visualization.N-best view algorithm based on viewpoint quality measures.Publicly available framework.

This paper introduces the “hybrid foraging” paradigm. In typical visual search tasks, observers search for one instance of one target among distractors. In hybrid search, observers search through visual displays for one instance of any of several types of target held in memory. In foraging search, observers collect multiple instances of a single target type from visual displays. Combining these paradigms, in hybrid foraging tasks observers search visual displays for multiple instances of any of several types of target (as might be the case in searching the kitchen for dinner ingredients or an X-ray for different pathologies). In the present experiment, observers held 8–64 targets objects in memory. They viewed displays of 60–105 randomly moving photographs of objects and used the computer mouse to collect multiple targets before choosing to move to the next display. Rather than selecting at random among available targets, observers tended to collect items in runs of one target type. Reaction time (RT) data indicate searching again for the same item is more efficient than searching for any other targets, held in memory. Observers were trying to maximize collection rate. As a result, and consistent with optimal foraging theory, they tended to leave 25–33% of targets uncollected when moving to the next screen/patch. The pattern of RTs shows that while observers were collecting a target item, they had already begun searching memory and the visual display for additional targets, making the hybrid foraging task a useful way to investigate the interaction of visual and memory search.

This paper hypothesizes a resource-limited mechanism, called a FINST, for individuating or indexing visual features, as distinct from encoding their type or location. FINSTs have the property that they index features in a way that is transparent to their retinal location, and hence under certain conditions succeed in "pointing to" scene locations. The basic assumption is that no operation upon sets of features can occur unless all the features to which the operation applies are first FINSTed. A number of applications of this hypothesis are explored in this paper, including applications to phenomena such as the spatial stability of visual percepts, the ability to track several independently moving targets in parallel, the ability to detect a class of spatial relations requiring the use of "visual routines", various mental imagery phenomena, and the ability to encode complex shapes for recognition. In addition, the possibility is examined that such indexes might be used to bind perceived locations to arguments in motor commands, thereby allowing some forms of perceptual-motor coordination. Several additional assumptions are introduced for this purpose, including the postulation of other indexes (called ANCHORS) for non-visually sensed locations. It is assumed that ANCHORS can be bound to FINSTs, thus allowing cross-referencing of visually detected locations to locations given within a proprioceptive or motor-command frame of reference.

Recent experimental evidence supports the idea that human observers are good at recognizing and categorizing materials. Fleming et al. reported that perceptual qualities and material classes are closely related using projected images (Journal of Vision 13(8) (2013) 9). In this paper, we further investigated their findings using real materials and degraded image versions of the same materials. We constructed a real material dataset, as well as four image datasets by varying chromaticity (color vs. gray) and resolution (high vs. low) of the material images. To investigate the fundamental properties of materials’ static surface appearance, we used stimuli that lacked shape and saturated color information. We then investigated the relationship between these perceptual qualities and the various types of image representation through psychophysical experiments. Our results showed that the representation method of some materials affected their perceptual qualities. These cases could be classified into the following three types: (1) perceptual qualities decreased by reproducing the materials as images, (2) perceptual qualities decreased by creating gray images, and (3) perceptual qualities such as “Hardness” and “Coldness” tended to increase when the materials were reproduced as low-quality images. Through methods such as principal component analysis and k-means clustering, we found that material categories are more likely to be confused when materials are represented as images, especially gray images.

Humans can visually recognize material categories of objects, such as glass, stone, and plastic, easily. However, little is known about the kinds of surface quality features that contribute to such material class recognition. In this paper, we examine the relationship between perceptual surface features and material category discrimination performance for pictures of materials, focusing on temporal aspects, including reaction time and effects of stimulus duration. The stimuli were pictures of objects with an identical shape but made of different materials that could be categorized into seven classes (glass, plastic, metal, stone, wood, leather, and fabric). In a pre-experiment, observers rated the pictures on nine surface features, including visual (e.g., glossiness and transparency) and non-visual features (e.g., heaviness and warmness), on a 7-point scale. In the main experiments, observers judged whether two simultaneously presented pictures were classified as the same or different material category. Reaction times and effects of stimulus duration were measured. The results showed that visual feature ratings were correlated with material discrimination performance for short reaction times or short stimulus durations, while non-visual feature ratings were correlated only with performance for long reaction times or long stimulus durations. These results suggest that the mechanisms underlying visual and non-visual feature processing may differ in terms of processing time, although the cause is unclear. Visual surface features may mainly contribute to material recognition in daily life, while non-visual features may contribute only weakly, if at all.

In 1945 Yves Le Grand published conditions, now largely forgotten, on the 4×4 matrix of an astigmatic eye for the eye to be emmetropic and an additional condition for retinal images to be undistorted. The conditions also applied to the combination of eye and the lens used to compensate for the refractive error. The conditions were presented with almost no justification. The purpose of this paper is to use linear optics to derive such conditions. It turns out that Le Grand’s conditions are correct for sharp images but his condition such that the images are undistorted prove to be neither necessary nor sufficient in general although they are necessary but not sufficient in most situations of interest in optometry and vision science. A numerical example treats a model eye which satisfies Le Grand’s condition of no distortion and yet forms elliptical and noncircular images of distant circles on the retina. The conditions for distant object are generalized to include the case of objects at finite distances, a case not examined by Le Grand.

In this paper we propose a novel approach to the task of salient object detection. In contrast to previous salient object detectors that are based on a spotlight attention theory, we follow an object-based attention theory and incorporate the notion of an object directly into our saliency measurements. Particularly, we consider proto-objects as units of the analysis, where a proto-object is a connected image region that can be converted into a plausible object or object-part, once a focus of attention reaches it. As the object-based attention theory suggests, we start with segmenting a complex image into proto-objects and then assess saliency for each proto-object. The most salient proto-object is considered as being a salient object. We distinguish two types of object saliency. Firstly, an object is salient if it differs from its surrounding, which we call center-surround saliency. Secondly, an object is salient if it contains rare or outstanding details, which we measure by integrated saliency. We demonstrate that these two types of object saliency have complementary characteristics; moreover, the combination of the two performs at the level of state-of-the-art in salient object detection.

This paper reports the first ever detailed study about eye movement patterns during single object recognition in visual agnosia. Eye movements were recorded in a patient with an integrative agnosic deficit during two recognition tasks: common object naming and novel object recognition memory. The patient showed normal directional biases in saccades and fixation dwell times in both tasks and was as likely as controls to fixate within object bounding contour regardless of recognition accuracy. In contrast, following initial saccades of similar amplitude to controls, the patient showed a bias for short saccades. In object naming, but not in recognition memory, the similarity of the spatial distributions of patient and control fixations was modulated by recognition accuracy. The study provides new evidence about how eye movements can be used to elucidate the functional impairments underlying object recognition deficits. We argue that the results reflect a breakdown in normal functional processes involved in the integration of shape information across object structure during the visual perception of shape.

Learning word-referent mappings is complex because the word and its referent tend to co-occur with multiple other words and potential referents. Such complexity has led to proposals for a host of constraints on learning, though how these constraints may interact has not yet been investigated in detail. In this paper, we investigated interactions between word co-occurrence constraints and cross-situational statistics in word learning. Analyses of child-directed speech revealed that when both object-referring and non-referring words occurred in the utterance, referring words were more likely to be preceded by a determiner than when the utterance contained only referring words. In a word learning study containing both referring and non-referring words, learning was facilitated when non-referring words contributed grammatical constraints analogous to determiners. The complexity of multi-word utterances provides an opportunity for co-occurrence constraints to contribute to word-referent mapping, and the learning mechanism is able to integrate these multiple sources of information.

This paper investigates the classification of shapes into broad natural categories such as animal or leaf. We asked whether such coarse classifications can be achieved by a simple statistical classification of the shape skeleton. We surveyed databases of natural shapes, extracting shape skeletons and tabulating their parameters within each class, seeking shape statistics that effectively discriminated the classes. We conducted two experiments in which human subjects were asked to classify novel shapes into the same natural classes. We compared subjects’ classifications to those of a naive Bayesian classifier based on the natural shape statistics, and found good agreement. We conclude that human superordinate shape classifications can be well understood as involving a simple statistical classification of the shape skeleton that has been “tuned” to the natural statistics of shape.

In the preceding paper (Lacey, S., Flueckiger, P., Stilla, R., Lava, M., Sathian, K., 2009a. Object familiarity modulates involvement of visual imagery in haptic shape perception), we showed that the activations evoked by visual imagery overlapped more extensively, and their magnitudes were more correlated, with those evoked during haptic shape perception of familiar, compared to unfamiliar, objects. Here we used task-specific analyses of functional and effective connectivity to provide convergent evidence. These analyses showed that the visual imagery and familiar haptic shape tasks activated similar networks, whereas the unfamiliar haptic shape task activated a different network. Multivariate Granger causality analyses of effective connectivity, in both a conventional form and one purged of zero-lag correlations, showed that the visual imagery and familiar haptic shape networks involved top-down paths from prefrontal cortex into the lateral occipital complex (LOC), whereas the unfamiliar haptic shape network was characterized by bottom-up, somatosensory inputs into the LOC. We conclude that shape representations in the LOC are flexibly accessible, either top-down or bottom-up, according to task demands, and that visual imagery is more involved in LOC activation during haptic shape perception when objects are familiar, compared to unfamiliar.

The reorientation task is a paradigm that has been used extensively to study the types of information used by humans and animals to navigate in their environment. In this task, subjects are reinforced for going to a particular location in an arena that is typically rectangular in shape. The subject then has to find that location again after being disoriented, and possibly after changes have been made to the arena. This task is used to determine the geometric and featural cues that can be used to reorient the agent in the arena. The purpose of the present paper is to present several simulation results that show that a simple neural network, a perceptron, can be used to generate many of the traditional findings that have been obtained using the reorientation task. These results suggest that reorientation task regularities can be explained without appealing to a geometric module that is a component of spatial processing.

Human beings perceive 3D shapes veridically, but the underlying mechanisms remain unknown. The problem of producing veridical shape percepts is computationally difficult because the 3D shapes have to be recovered from 2D retinal images. This paper describes a new model, based on a regularization approach, that does this very well. It uses a new simplicity principle composed of four shape constraints: viz., symmetry, planarity, maximum compactness and minimum surface. Maximum compactness and minimum surface have never been used before. The model was tested with random symmetrical polyhedra. It recovered their 3D shapes from a single randomly-chosen 2D image. Neither learning, nor depth perception, was required. The effectiveness of the maximum compactness and the minimum surface constraints were measured by how well the aspect ratio of the 3D shapes was recovered. These constraints were effective; they recovered the aspect ratio of the 3D shapes very well. Aspect ratios recovered by the model were compared to aspect ratios adjusted by four human observers. They also adjusted aspect ratios very well. In those rare cases, in which the human observers showed large errors in adjusted aspect ratios, their errors were very similar to the errors made by the model.

In this paper we present an improved model for line and edge detection in cortical area V1. This model is based on responses of simple and complex cells, and it is multi-scale with no free parameters. We illustrate the use of the multi-scale line/edge representation in different processes: visual reconstruction or brightness perception, automatic scale selection and object segregation. A two-level object categorization scenario is tested in which pre-categorization is based on coarse scales only and final categorization on coarse plus fine scales. We also present a multi-scale object and face recognition model. Processing schemes are discussed in the framework of a complete cortical architecture. The fact that brightness perception and object recognition may be based on the same symbolic image representation is an indication that the entire (visual) cortex is involved in consciousness.

The 3D cube figures used by Shepard and Metzler [Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-dimensional objects. Science, 171, 701-703] have been applied in a broad range of studies on mental rotation. This note provides a brief background on these figures, their general use in cognitive psychology and their role in studying spatial behavior. In particular, it is pointed out that large sex differences with the 3D mental rotation figures tend to be observed only in particular tasks, such as the Vandenberg and Kuse test [Vandenberg, S. G., & Kuse, A. R. (1978). Mental rotations, a group test of three-dimensional spatial visualization. Perceptual and Motor Skills, 47, 599-604] that involve multiple figures within a single problem. In contrast, pairwise presentation of the same 3D figures yields either small or no significant sex differences. In the context of the very broad range of ongoing research done with 3D figures, and the desirability of uniformity in the stimulus material used, we introduce a library of 16 cube mental rotation figures, each presented in orientations ranging from 0 to 360 degr in 5 degr steps, and with its mirror image, for a total of 2336 figures. This library, freely available to researchers, will help in the creation of mental rotation tasks both for presentation on the computer screen and for pencil and paper applications.

Contour interpolation is the process whereby spatially separated object fragments (inducers) are connected on the basis of their contour relations. An important characteristic of interpolation between simultaneously presented inducers is that observers rely on interpolation regions to perform a discrimination task. However, it is unclear if the same property holds when inducers are separated in both space and time. To address this question of spatiotemporal interpolation, we had participants discriminate spatiotemporally presented “fat” and “thin” noise-corrupted figures, when the figures were stationary (Experiment 1) or moving (Experiment 2), and when the connections across vertical gaps were either real, interpolated (illusory), or absent. Classification images from both experiments showed that noise regions near interpolated boundaries affect performance comparably to when real contours appear, but very little in the absence of interpolation. The classification images also revealed information about the time course of interpolation and suggested that contour interpolation between simultaneously visible inducers may be a special case of a more general spatiotemporal contour interpolation process.1The following acronyms appear in this paper: RMS—root mean square; SNR—signal to noise ratio; CI—Classification image.1

End-stopped cells in cortical area V1, which combine outputs of complex cells tuned to different orientations, serve to detect line and edge crossings, singularities and points with large curvature. These cells can be used to construct retinotopic keypoint maps at different spatial scales (level-of-detail). The importance of the multi-scale keypoint representation is studied in this paper. It is shown that this representation provides very important information for object recognition and face detection. Different grouping operators can be used for object segregation and automatic scale selection. Saliency maps for focus-of-attention can be constructed. Such maps can be employed for face detection by grouping facial landmarks at eyes, nose and mouth. Although a face detector can be based on processing within area V1, it is argued that such an operator must be embedded into dorsal and ventral data streams, to and from higher cortical areas, for obtaining translation-, rotation- and scale-invariant detection.

When humans scan their visual environment, relevant objects are selectively attended for enhanced processing. It is still unclear in what ways processing is modified by attention, and whether attentional selection operates on an individual feature (such as colour, orientation or motion) or on binding together different features. In the experiments reported in this paper, these two stages were characterized using psychophysical reverse correlation. Subjects viewed eight patches, briefly flashed and symmetrically arranged around fixation. Each patch consisted of segments that could vary in both colour and orientation. One of the patches (‘target’) differed from the remaining ‘distractor’ patches with respect to either its orientation, colour, or both (in three different experiments). Subjects were asked to detect the target patch. The stimulus was preceded by a cue. On some trials (‘cued’ trials), the cue informed observers that the target patch could only appear at two of the eight possible locations. On remaining (‘uncued’) trials, all eight positions were valid. Psychophysical reverse correlation was then applied to derive linear estimates of sensory filters for orientation only, colour only, and their conjunction. In line with the properties of single neurons in cortex, attentional cueing did not affect sensory tuning for detecting individual features. However, it affected the way in which features were subsequently (and very inefficiently) combined in a multiplicative fashion. The results are consistent with a model in which attention recalibrates internal responses to the statistics of the stimulus by having signals from different features mutually control each other through reciprocal inhibition.

Research has suggested the presence of subtle long-term cognitive changes in otherwise well-functioning individuals who have previously sustained a mild head injury (MHI). This paper investigated the long-term effects of MHI on visual, spatial, and visual-spatial short-term memory in well-functioning university students. Sixteen students who reported having sustained a MHI were compared to 16 controls on tests of short-term memory (STM) for abstract polygons in haphazardly arranged locations. The three tests differed only in the requirements for recall (shapes for the visual task, locations for the spatial task, and the shapes in their respective locations for the visual-spatial task). MHI participants were selectively impaired on spatial memory, suggesting that tasks of spatial STM may be more sensitive, compared to tasks of visual STM, to the subtle long-term cognitive changes that may be present after a MHI.

Recent studies show that what connotes an object is first of all a certain spatio-temporal structure. In this paper we describe some of the temporal features characterizing the temporal structure of objects: pre-existence, persistence, conservation of identity in spite of perceptive discontinuity, surviving changes in colour, size, and shape. We argue that time is an indispensable attribute for every type of object and briefly discuss the implication of this view with respect to a specific neuropsychological syndrome: unilateral spatial neglect.

In this paper, we propose a space-variant image representation model based on properties of magnocellular visual pathway, which perform motion analysis, in human retina. Then, we present an algorithm for the tracking of multiple objects in the proposed space-variant model. The proposed space-variant model has two effective image representations for object recognition and motion analysis, respectively. Each image representation is based on properties of two types of ganglion cell, which are the beginning of two basic visual pathways; one is parvocellular and the other is magnocellular. Through this model, we can get the efficient data reduction capability with no great loss of important information. And, the proposed multiple objects tracking method is restricted in space-variant image. Typically, an object-tracking algorithm consists of several processes such as detection, prediction, matching, and updating. In particular, the matching process plays an important role in multiple objects tracking. In traditional vision, the matching process is simple when the target objects are rigid. In space-variant vision, however, it is very complicated although the target is rigid, because there may be deformation of an object region in the space-variant coordinate system when the target moves to another position. Therefore, we propose a deformation formula in order to solve the matching problem in space-variant vision. By solving this problem, we can efficiently implement multiple objects tracking in space-variant vision.

Attending to objects implies the concurrent process of features that are analyzed in different visual subsystems or domains. Previous works have shown that attention cannot be simultaneously directed to the components of motion present in two transparent surfaces [M. Valdés et al., Cognition 66 (1998) B13-B23], even though they occupy overlapping regions of space. In this paper, possible across-domain effects in object-based attention were examined using a conjunction of form and motion in transparent superimposed surfaces. After directing attention to one surface, different combinations of motion and form judgements were performed. If both attributes belonged to the same surface, no interference was found. If the two judgements concerned features from different surfaces, a large performance cost was present for the attribute belonging to the uncued surface. The fact that these effects cut across feature domains supports the integrated competition hypothesis [J. Duncan, Attention and Performance XVI, The MIT Press, 1996, pp. 549-578].

The possibility of a link between medial axes (hereafter called symmetries) and figure salience has recently been proposed [Vision Res. 38 (1998) 2323; Vision Res. 38 (1998) 2429]. In this paper we investigated the characteristics of transient visual evoked potentials (VEPs) associated with stimulus figures designed to have different symmetries. Significant trends were observed for VEP components N1 (160–190 ms) and P2 (220–250 ms). N1 and P2 had shorter peak latencies for stimuli with greater symmetry peak amplitudes.

Contours are believed to play a key role in the visual analysis of scenes by the primate brain. In dynamic scenes, the presence of contours is often signaled by discontinuities in motion fields. However, it is unclear whether the motion fields over which the visual system extracts discontinuities, correspond to the local optic-flow or the pattern motion fields obtained by integrating local estimates. A resolution of this issue would provide important clues about the organization of visual motion and form analysis processes. In this paper, we present experimental evidence which suggests that the perception of motion defined contours is strongly dependent on motion integration - an operation that is believed to take place relatively late in the visual stream.

This paper presents a novel hypothesis on the function of massive feedback pathways in mammalian visual systems. We propose that the cortical feature detectors compete not for the right to represent the output at a point, but for exclusive rights to abstract and represent part of the underlying input. Feedback can do this very naturally. A computational model that implements the above idea for the problem of line detection is presented and based on that we suggest a functional role for the thalamo-cortical loop during perception of lines. We show that the model successfully tackles the so called Cross problem. Based on some recent experimental results, we discuss the biological plausibility of our model. We also comment on the relevance of our hypothesis (on the role of feedback) to general sensory information processing and recognition. (C) 1998 Published by Elsevier Science Ltd. All rights reserved.

In this paper we analyse the time course of neuronal activity in temporal cortex to the sight of the head and body. Previous studies have already demonstrated the impact of view, orientation and part occlusion on individual cells. We consider the cells as a population providing evidence in the form of neuronal activity for perceptual decisions related to recognition. The time course of neural responses to stimuli provides an explanation of the variation in speed of recognition across different viewing circumstances that is seen in behavioural experiments. A simple unifying explanation of the behavioural effects is that the speed of recognition of an object depends on the rate of accumulation of activity from neurones selective for the object, evoked by a particular viewing circumstance. This in turn depends on the extent that the object has been seen previously under the particular circumstance. For any familiar object, more cells will be tuned to the configuration of the object's features present in the view or views most frequently experienced. Therefore, activity amongst the population of cells selective for the object's appearance will accumulate more slowly when the object is seen in an unusual view, orientation or size. This accounts for the increased time to recognise rotated views without the need to postulate `mental rotation' or `transformations' of novel views to align with neural representations of familiar views.

This paper reports the case of a patient with a peripheral spelling impairment who is much more severely impaired at writing in lower-case letters than in upper-case letters. This pattern can be observed when writing both words and single letters of the alphabet. Despite this, his problems in writing lower-case letters are no longer present when he is writing cursively. This case therefore indicates that the ability to print letters in lower-case can be selectively impaired in the absence of similar problems in printing,,upper-case letters or in writing lower-case letters cursively. In terms of the model of writing put forward by Ellis (1982, 1988), this suggests that allographic level representations for print handwriting styles can be functionally dissociated from allographic representations for cursive styles.

Monkeys with inferior temporal cortex lesions cannot discriminate between different shapes (e.g., + vs. 0) but can discriminate between shapes that differ only in orientation (e.g., 6 vs. 9). Lesions of the parietal cortex, on the other hand, impair the discrimination of rotated shapes but spare the ability to discriminate between different shapes. A similar dissociation is found between some visual agnosics who can match but not identify rotated views of objects and other patients who can identify and discriminate objects only if the view is conventional; any change in orientation disrupts performance. In this paper we argue that two mechanisms may be available for the perception of rotated shapes. Which mechanism is used depends on the degree of rotation. It is suggested that the different effects of parietal and temporal lesions reflect the relative contributions of the two areas to the task and disrupt different stages of the two strategies used. A framework for the cortical processing of rotated shapes in the non-human primate is presented.

A striking example of our sensitivity to dynamic information is our ability to infer motion from still images depicted in paintings, photographs or cartoons. What are the neural mechanisms that mediate this implied motion perception? In a recent paper, Krekelberg et al. demonstrate that form cues that imply motion are integrated with real motion information, and influence perception in both humans and monkeys and the neural processing in prototypical motion areas of the monkey brain. Perception and successful interaction with moving objects entail that the visual system integrates information from form and motion cues into unified dynamic perceptual events. However, traditionally, shape and motion processing have been attributed to anatomically and functionally separable neural pathways in the primate brain [1]. An extrastriate visual area in the medial temporal monkey brain (MT/V5) and its human analogue in the ascending limb of the inferior temporal sulcus (hMT+/V5) have been identified as one of the main regions involved in the analysis of visual motion [2] and [3]. By contrast, regions in the occipitotemporal cortex (V4, IT) have been implicated in the analysis of shape properties and object recognition in the monkey and the human brain [4] and [5]. Although significant progress has been made in uncovering the neural mechanisms that mediate motion and form perception, surprisingly little is known about possible interactions of these mechanisms that may underlie the unified perception of moving objects in our dynamic visual environments. Krekelberg et al.[6] provide evidence for such interactions by showing similar responses in monkey MT and MST cells to real motion and static form patterns that have no coherent physical motion but imply motion. These physiological findings are consistent with the perception of their human and monkey subjects.

Various visual cues provide information about depth and shape in a scene. When several of these cues are simultaneously available in a single location in the scene, the visual system attempts to combine them. In this paper, we discuss three key issues relevant to the experimental analysis of depth cue combination in human vision: cue promotion, dynamic weighting of cues, and robustness of cue combination. We review recent psychophysical studies of human depth cue combination in light of these issues. We organize the discussion and review as the development of a model of the depth cue combination process termed modified weak fusion (MWF). We relate the MWF framework to Bayesian theories of cue combination. We argue that the MWF model is consistent with previous experimental results and is a parsimonious summary of these results. While the MWF model is motivated by normative considerations, it is primarily intended to guide experimental analysis of depth cue combination in human vision. We describe experimental methods, analogous to perturbation analysis, that permit us to analyze depth cue combination in novel ways. In particular these methods allow us to investigate the key issues we have raised. We summarize recent experimental tests of the MWF framework that use these methods.

Shape constancy refers to the phenomenon in which the percept of the shape of a given object remains constant despite changes in the shape of the object's retinal image. The phenomenon of shape constancy is considered from historical, theoretical and empirical perspectives in this paper. First, four prior theories are discussed; specifically, (1) Helmholtzian theory, which assumes that shape constancy is achieved by taking an object's orientation into account, (2) Gestalt theory, which assumes that shape constancy involves a relationship between the perceived shape and perceived orientation of an object, (3) Gibsonian theory, which assumes that shape constancy is based on projective invariants and (4) multiple view theory, which assumes that shape constancy is achieved by memorizing a large set of different views of the object. It is shown, by an analysis of the prior literature, that none of these theories can actually explain the phenomenon of shape constancy. A new theory, which is based on new perspective invariants of a flat shape, is then proposed. The new Perspective Invariants Theory can account for all prior shape constancy experiments. New experiments, testing predictions of the Perspective Invariants Theory are then described. These experiments showed that: (1) a novel shape can be matched with its single perspective image in the absence of depth cues, (2) perceptual processing of shape is impaired when the range of possible values of tilt is wide, (3) perceptual processing of shape is not affected by the width of the range of possible values of slant. These results support predictions of Perspective Invariants Theory.

This paper reviews selected research on the neurophysiology of cutaneous mechanoreceptors and their relationship to tactile sensations. Particular emphasis is placed on studies in which neurophysiological results are directly comparable to psychophysical results. These studies provide the strongest evidence for correlating mechanoreceptor function with specific tactile sensations. Furthermore, these studies have implications for how one should administer tactile stimuli in order to selectively stimulate particular groups of mechanoreceptors.

Recent evidence concerning the neural basis of visual perception and learning has led to the development of a sequential neural model in which the striate, prestriate, inferior temporal, and ventromedial frontotemporal areas are viewed as successive links in a neocortical-limbic pathway mediating visual-affective associations [17, 20, 21]. In this paper an analogous model is proposed for touch, with the primary and secondary sensory areas (SI and SII) serving as the successive somatosensory links to the ventromedial limbic region. The model helps to account for some otherwise puzzling data reviewed here pertaining to (a) hierarchical organization in tactual perception, (b) localization of the tactual memory trace, and (c) sensory-limbic interaction in tactual learning. The pivotal link in this model is the secondary sensory area (SII), which is assigned a role in somesthesis similar to that attributed to the inferior temporal area in vision. Evidence from a variety of sources is introduced to support the parallel.

Contrast- or quasi-frequency-modulated masker gratings consisting of three high frequency components (8.8, 11 and 13.2 c/deg) affect the detectability of a 2.2 c/deg signal grating, to an extent that is strongly dependent upon the relative phase between signal and masker. Unmodulated high frequency maskers have no such phase-dependent effects. This paper explores the possibility that the visual system's nonlinear response to luminance is responsible for these phenomena. A specific hypothesis is proposed according to which the effects of the spatially modulated maskers are due entirely to a distortion product at 2.2 c/deg caused by the visual nonlinearity. Although some of the predictions of this hypothesis are borne out by the experimental findings, others are contradicted.

This paper derives the constraints on a set of channels that would be consistent with the results of several experiments on the temporal properties of the visual system, and it describes a specific set of channels that meet these constraints. Data on simultaneous detection and discrimination require a minimum of three channels. Temporal frequency discrimination at and above threshold constrain the bandwidths and locations of the channels. The shape of the temporal modulation sensitivity function constrains their sensitivities. The functions that meet these constraints are similar to those derived from masking data, and they can account for data on flicker matching, notch losses in modulation sensitivity, and changes of perceived temporal frequency with changes of modulation depth.

Visual information can be stored relative to a particular point of view or independently of any particular point of view. Research on mental rotation has shown that people can store and use viewer-centered visual representations of objects and scenes. Some theories of object recognition posit that object-centered representations are also stored and used to represent the shape of three-dimensional objects. In this paper a series of experiments provides evidence that people can store and use both viewer-centered and object-centered representations of three-dimensional objects.

In perceiving objects we may synthesize conjunctions of separable features by directing attention serially to each item in turn (A. Treisman and G. Gelade, Cognitive Psychology, 1980, 12, 97–136) . This feature-integration theory predicts that when attention is diverted or overloaded, features may be wrongly recombined, giving rise to “illusory conjunctions.” The present paper confirms that illusory conjunctions are frequently experienced among unattended stimuli varying in color and shape, and that they occur also with size and solidity (outlined versus filled-in shapes). They are shown both in verbal recall and in simultaneous and successive matching tasks, making it unlikely that they depend on verbal labeling or on memory failure. They occur as often between stimuli differing on many features as between more similar stimuli, and spatial separation has little effect on their frequency. Each feature seems to be coded as an independent entity and to migrate, when attention is diverted, with few constraints from the other features of its source or destination.

Gestalt psychologists have stressed the perceptual importance of grouping. On the basis of multichannel theory of early visual processing, grouping has been attributed to the operation of channels tuned to low spatial frequencies. If this explanation is correct and sufficient, grouping should not occur when looking at highpass filtered images, because they do not excite low spatial frequency channels. However this paper brings evidence that visual grouping also occurs in such a situation. Hence, a more comprehensive explanation of visual grouping is needed.

Forty children 1 1/2 to 3 1/2 years of age searched for stickers that were hidden underneath two discretely different classes of objects embedded in arrays containing several classes. The paper focuses on a previously unobserved strategy that appeared at 3 1/2. Whereas the younger children searched underneath those objects that looked like the objects the children knew had stickers, the 3 1/2-year-olds systematically explored not only the objects that did, but also the objects that did not, have stickers. Eight adults performed similarly to the 3 1/2-year-olds. Analyses of developmental trends in the data and the observation of individual children performing the tasks suggest that the children's increased inspection of the negative objects was not merely the result of a random or more exhaustive search. The children appear to have been attempting to determine which kinds of choices were the wrong ones. The attempt to define and delimit negative cases is a central aspect of reflective thought.

It is now generally accepted among neuroscientists that the sensory cortex of the brain is arranged in a structure that is simultaneously ''topographic'' (a pointwise mapping), layered, and columnar. The microcolumns in the columnar structure exhibit both a directional and an areal response in addition to the pointwise one. It is shown that these directional-areal response fields are contact elements upon the visual manifold that generate visual contours as the ''lifts'' of the form stimuli upon the retina into a contact bundle embodied in the visual cortex itself. Those invariances of form perception termed the ''psychological constancies'' -shape, size, motion, color constancies, etc.-represent the action of the conformal group CO(1,3) upon the visual manifold. It follows that the connection for the cortical contact bundle is a Lie group connection. The paper closes with an axiomatic treatment of visual perception and expression of Riegel's dialectical psychology in terms of the symmetric difference operation to provide an association between perceptual and cognitive function.

This paper describes three experiments concerned with the ability of 3-month-old infants to discriminate between stationary solid forms. Experiment 1 confirmed, using a habituation/recovery technique, the finding of Cook, Field, and Griffiths (1978) that the infants do not distinguish between a cube and a truncated pyramid (and a range of other solid forms). Experiment 2 showed that infants can distinguish the square face of the cube from a trapezoidal face of the truncated pyramid when these shapes are presented in isolation. Experiment 3 showed that this discrimination is not made when the corresponding faces (square and trapezoid) of the solid cube and trapezoid are colored distinctively. These results are discussed in terms of limitations in the infants' capacity to resolve solid form.

This paper presents a sampling of the psychophysical and electrophysiologic examinations which have been applied to determine the location of the visual processing defect in functional amblyopia. Areal summation and lateral inhibition are discussed along with methods for determining their extent at different locations in the visual system. Using psychophysical and visual-evoked response data from a series of these tests in humans, a case is made that the outer retina is the locus of a major defect leading to the loss of visual resolution in functional amblyopia.

The results reported in this paper indicate that images of simple two-dimensional patterns are formed sequentially. Segments of block letters were imaged in roughly the order in which most people draw the letters, and segments of novel patterns were imaged in the order in which the segments were learned. These results were shown not to be an artifact of how people scan over images once they are formed, and could not have been due to experimenter expectancy effects. In addition, a new objective methodology indicated that images of complex letters required more time to generate, and the increases in time with complexity were very similar to those obtained simply by measuring the time subjects required to report that images had been formed.

This paper presents an extension and refinement of the author's theory for human visual information processing, which is then applied to the problem of human facial recognition. Several fundamental processes are implicated: encoding of visual images into neural patterns, detection of simple facial features, size standardization, reduction of the neural patterns in dimensionality, and finally correlation of the resulting sequence of patterns with all visual patterns already stored in memory. In the theory presented here, this entire process is automatically "driven" by the storage system in what amounts to an hypothesis verification paradigm. Neural networks for carrying out these processes are presented and syndromes resulting from damage to the proposed system are analyzed. A correspondence between system component and brain anatomy is suggested, with particular emphasis on the role of the primary visual cortex in this process. The correspondence is supported by structural and electrophysiological properties of the primary visual cortex and other related structures. The logical (computational) role suggested for the primary visual cortex has several components: size standardization, size reduction, and object extraction. The result of processing by the primary visual cortex, it is suggested, is a neural encoding of the visual pattern at a size suitable for storage. (In this context, object extraction is the isolation of regions in the visual field having the same color, texture, or spatial extent.) It is shown in detail how the topology of the mapping from retina to cortex, the connections between retina, lateral geniculate bodies and primary visual cortex, and the local structure of the cortex itself may combine to encode the visual patterns. Aspects of this theory are illustrated graphically with human faces as the primary stimulus. However, the theory is not limited to facial recognition but pertains to Gestalt recognition of any class of familiar objects or scenes.

Many people have argued that natural categories are organized in terms of a family resemblance principle. Members of family resemblance categories tend to share properties with each other but have no properties that are singly necessary and jointly sufficient (defining) for category membership. This paper reports seven experiments using a sorting task to evaluate the conditions under which people prefer to construct categories according to a family resemblance principle. The first set of studies followed the typical practice of defining family resemblance in terms of independent sets of matching and mismatching values. Across a variety of stimulus materials, instructions, procedures, and category structures, family resemblance sorting was almost never observed. Despite procedures designed to prevent it, participants persisted in sorting on the basis of a single dimension. The second set of studies explored the idea that interproperty relationships rather than independent features serve to organize categories. We found that people will abandon unidimensional sorting in favor of sorting by correlated properties, especially when they can be causally connected. In addition, when conceptual knowledge is added which makes interproperty relationships salient, family resemblance sorting becomes fairly common. Implications of the results for the development of family resemblance categories and the practice of treating properties or features as additive and independent are discussed.