Multiple parallel access in visual attention

Feature Counting Is Impaired When Shifting Attention Between the Eyes in Adults With Amblyopia

BACKGROUND

Feature counting requires rapid shifts of attention in the visual field and reflects higher-level cortical functions. This process is drastically impaired in the amblyopic eye of strabismic amblyopes. In this study, we hypothesized that feature counting performance in anisometropic and strabismic amblyopes is further impaired when shifts in attention is required between the eyes.

MATERIALS AND METHODS

Through a mirror stereoscope, highly visible Gabor patches were presented to the same eye within a block or randomly presented to the left eye or to the right eye with an equal probability within a block. The task was to report the number of Gabors (3 to 9) as accurately as possible. Counting performance was compared between the amblyopes and the normal-vision observers and between the viewing conditions (shifting attention between the eyes versus maintaining attention in the same eye).

RESULTS

When attention was maintained in the same eye, the amblyopic eye of both anisometropic and strabismic groups undercounted the number of Gabors, but achieved near-perfect performance with their fellow eye, compared to normal-vision observers. In contrast, when shifting attention randomly to the left or to the right eye, the amblyopic eye further undercounted the number of Gabors. Undercounting was also found in the fellow eye of strabismic amblyopes, but was not in the fellow eye of anisometropic amblyopes. Performance in normal-vision observers did not differ between shifting attention between the eyes and maintaining attention in the same eye.

CONCLUSION

Our data showed that the amblyopic eye of both anisometropic and strabismic amblyopes further undercounted features when shifting attention between the eyes, compared to when maintaining attention in the same eye. This suggests that the ability to quickly redirect attention, particularly under interocular suppression, is impaired in amblyopia. The fellow eye of strabismic amblyopes also undercounted features when shifting attention between the eyes. However, such fellow eye abnormality was not found in anisometropic amblyopes, suggesting that different patterns of visual deficits are associated with amblyopia of different etiologies. The inability to count multiple features accurately reflects dysfunctions of high-level cortices in the amblyopic brain.

Salience Models: A Computational Cognitive Neuroscience Review

The seminal model by Laurent Itti and Cristoph Koch demonstrated that we can compute the entire flow of visual processing from input to resulting fixations. Despite many replications and follow-ups, few have matched the impact of the original model-so what made this model so groundbreaking? We have selected five key contributions that distinguish the original salience model by Itti and Koch; namely, its contribution to our theoretical, neural, and computational understanding of visual processing, as well as the spatial and temporal predictions for fixation distributions. During the last 20 years, advances in the field have brought up various techniques and approaches to salience modelling, many of which tried to improve or add to the initial Itti and Koch model. One of the most recent trends has been to adopt the computational power of deep learning neural networks; however, this has also shifted their primary focus to spatial classification. We present a review of recent approaches to modelling salience, starting from direct variations of the Itti and Koch salience model to sophisticated deep-learning architectures, and discuss the models from the point of view of their contribution to computational cognitive neuroscience.

The effects of colour complexity and similarity on multiple object tracking performance

Surface features can be used during multiple object tracking (MOT). Previous studies suggested that surface features might be stored in visual working memory to assist object tracking, and attentive tracking and visual working memory share common attentional resources. However, it is still unknown whether features of both the target and distractor sets will be stored, or features of the target and distractor sets are processed differently. Moreover, how feature distinctiveness and similarity between the target and distractor sets affect tracking and allocation of attentional resources are still not clear. First, we manipulated the colour complexity of the target set (CT) and the colour complexity of the distractor set (CD), respectively, in two experiments, where colours of the target and distractor sets were always distinct, to test their effects on tracking performance. If features of the target and distractor sets are stored, manipulating feature complexity of the target and distractor sets would significantly affect tracking performance. Second, this study tested whether tracking performance was affected by different levels of distinctiveness between the target and distractor sets (DTD) and explored how distinctiveness affected tracking and allocation of attentional resources. Results showed that DTD and CT significantly affect tracking performance and allocation of attentional resources, but not CD. These results indicated that when targets and distractors have distinct features, only the surface features of the targets are maintained in visual working memory. And when targets have the same colour with the distractors, they are more difficult and consume more attentional resources to track.

Multiple-Target Tracking: A Role for Working Memory?

In order to identify the cognitive processes associated with target tracking, a dual-task experiment was carried out in which participants undertook a dynamic multiple-object tracking task first alone and then again, concurrently with one of several secondary tasks, in order to investigate the cognitive processes involved. The research suggests that after designated targets within the visual field have attracted preattentive indexes that point to their locations in space, conscious processes, vulnerable to secondary visual and spatial task interference, form deliberate strategies beneficial to the tracking task, before tracking commences. Target tracking itself is realized by central executive processes, which are sensitive to any other cognitive demands. The findings are discussed in the context of integrating dynamic spatial cognition within a working memory framework.

Additivity of Feature-Based and Symmetry-Based Grouping Effects in Multiple Object Tracking

Multiple object tracking (MOT) is an attentional process wherein people track several moving targets among several distractors. Symmetry, an important indicator of regularity, is a general spatial pattern observed in natural and artificial scenes. According to the “laws of perceptual organization” proposed by Gestalt psychologists, regularity is a principle of perceptual grouping, such as similarity and closure. A great deal of research reported that feature-based similarity grouping (e.g., grouping based on color, size, or shape) among targets in MOT tasks can improve tracking performance. However, no additive feature-based grouping effects have been reported where the tracking objects had two or more features. “Additive effect” refers to a greater grouping effect produced by grouping based on multiple cues instead of one cue. Can spatial symmetry produce a similar grouping effect similar to that of feature similarity in MOT tasks? Are the grouping effects based on symmetry and feature similarity additive? This study includes four experiments to address these questions. The results of Experiments 1 and 2 demonstrated the automatic symmetry-based grouping effects. More importantly, an additive grouping effect of symmetry and feature similarity was observed in Experiments 3 and 4. Our findings indicate that symmetry can produce an enhanced grouping effect in MOT and facilitate the grouping effect based on color or shape similarity. The “where” and “what” pathways might have played an important role in the additive grouping effect.

On the Efficiency of Visual Selective Attention: Efficient Visual Search Leads to Inefficient Distractor Rejection

The ability to ignore irrelevant peripheral distractors was assessed as a function of the efficiency in visual search for a target at the center of a display Efficient target search, among dissimilar nontargets, led to greater distraction than inefficient search, among similar nontargets This seemingly paradoxical result is predicted by the recent proposal (Lavie, 1995a) that irrelevant processing can be prevented only by increasing the load for relevant processing Varying the set size of similar items in the central search task demonstrated that interference from irrelevant distractors was eliminated only with more than four relevant items These results demonstrate how capacity limits determine the efficiency of selective attention, and raise questions about some standard assumptions of most visual search models

Selection of new objects by onset capture and visual marking

Visual search is easier after looking at some distractors in advance because previewed distractors are excluded from the search (preview benefit). A dominant explanation for preview benefit is that it occurs because of the inhibition of old objects (visual marking). However, another view claims that preview benefit simply reflects automatic attentional orienting to new objects (onset capture). To address the question of whether visual marking plays any role in addition to onset capture, we compared the search performance for a target that always appeared as a new item ("marking" condition) with the performance for a target that appeared equally as a new or old item ("capture" condition). When items were presented at random positions in an invisible matrix, the slope in the "marking" condition was shallower than that in the "capture" condition, favoring the involvement of visual marking (Experiments 1 and 2). In contrast, no difference in slope was found among the search conditions regardless of changes in old items when items were arranged around the circumference of a circle (Experiment 3). These findings suggest that the contribution of visual marking depends on the configuration of search items; with complex displays, prioritizing selection for new objects is more effective if coupled with de-prioritizing de-selection for old objects.

Suggested Interactivity: Seeking Perceived Affordances for Information Visualization

In this article, we investigate methods for suggesting the interactivity of online visualizations embedded with text. We first assess the need for such methods by conducting three initial experiments on Amazon's Mechanical Turk. We then present a design space for Suggested Interactivity (i. e., visual cues used as perceived affordances-SI), based on a survey of 382 HTML5 and visualization websites. Finally, we assess the effectiveness of three SI cues we designed for suggesting the interactivity of bar charts embedded with text. Our results show that only one cue (SI3) was successful in inciting participants to interact with the visualizations, and we hypothesize this is because this particular cue provided feedforward.

Nonverbal communicative signals modulate attention to object properties

We investigated whether the social context in which an object is experienced influences the encoding of its various properties. We hypothesized that when an object is observed in a communicative context, its intrinsic features (such as its shape) would be preferentially encoded at the expense of its extrinsic properties (such as its location). In 3 experiments, participants were presented with brief movies, in which an actor either performed a noncommunicative action toward 1 of 5 different meaningless objects, or communicatively pointed at 1 of them. A subsequent static image, in which either the location or the identity of an object changed, tested participants' attention to these 2 kinds of information. Throughout the 3 experiments we found that communicative cues tended to facilitate identity change detection and to impede location change detection, whereas in the noncommunicative contexts we did not find such a bidirectional effect of cueing. The results also revealed that the effect of the communicative context was a result the presence of ostensive-communicative signals before the object-directed action, and not to the pointing gesture per se. We propose that such an attentional bias forms an inherent part of human communication, and function to facilitate social learning by communication.

Indexing and the object concept: developing `what' and `where' systems

The study of object cognition over the past 25 years has proceeded in two largely non-interacting camps. One camp has studied object-based visual attention in adults, while the other has studied the object concept in infants. We briefly review both sets of literature and distill from the adult research a theoretical model that we apply to findings from the infant studies. The key notion in our model of object representation is the `sticky' index, a mechanism of selective attention that points at a physical object in a location. An object index does not represent any of the properties of the entity at which it points. However, once an index is pointing to an object, the properties of that object can be examined and featural information can be associated with, or `bound' to, its index. The distinction between indexing and feature binding underwrites the distinction between object individuation and object identification, a distinction that turns out to be crucial in both the adult attention and the infant object-concept literature. By developing the indexing model, we draw together two disparate sets of literature and suggest new ways to study object-based attention in infancy.

Neural dynamics of object-based multifocal visual spatial attention and priming: object cueing, useful-field-of-view, and crowding

How are spatial and object attention coordinated to achieve rapid object learning and recognition during eye movement search? How do prefrontal priming and parietal spatial mechanisms interact to determine the reaction time costs of intra-object attention shifts, inter-object attention shifts, and shifts between visible objects and covertly cued locations? What factors underlie individual differences in the timing and frequency of such attentional shifts? How do transient and sustained spatial attentional mechanisms work and interact? How can volition, mediated via the basal ganglia, influence the span of spatial attention? A neural model is developed of how spatial attention in the where cortical stream coordinates view-invariant object category learning in the what cortical stream under free viewing conditions. The model simulates psychological data about the dynamics of covert attention priming and switching requiring multifocal attention without eye movements. The model predicts how "attentional shrouds" are formed when surface representations in cortical area V4 resonate with spatial attention in posterior parietal cortex (PPC) and prefrontal cortex (PFC), while shrouds compete among themselves for dominance. Winning shrouds support invariant object category learning, and active surface-shroud resonances support conscious surface perception and recognition. Attentive competition between multiple objects and cues simulates reaction-time data from the two-object cueing paradigm. The relative strength of sustained surface-driven and fast-transient motion-driven spatial attention controls individual differences in reaction time for invalid cues. Competition between surface-driven attentional shrouds controls individual differences in detection rate of peripheral targets in useful-field-of-view tasks. The model proposes how the strength of competition can be mediated, though learning or momentary changes in volition, by the basal ganglia. A new explanation of crowding shows how the cortical magnification factor, among other variables, can cause multiple object surfaces to share a single surface-shroud resonance, thereby preventing recognition of the individual objects.

The effect of viewing eccentricity on enumeration

Visual acuity and contrast sensitivity progressively diminish with increasing viewing eccentricity. Here we evaluated how visual enumeration is affected by visual eccentricity, and whether subitizing capacity, the accurate enumeration of a small number (∼3) of items, decreases with more eccentric viewing. Participants enumerated gratings whose (1) stimulus size was constant across eccentricity, and (2) whose stimulus size scaled by a cortical magnification factor across eccentricity. While we found that enumeration accuracy and precision decreased with increasing eccentricity, cortical magnification scaling of size neutralized the deleterious effects of increasing eccentricity. We found that size scaling did not affect subitizing capacities, which were nearly constant across all eccentricities. We also found that size scaling modulated the variation coefficients, a normalized metric of enumeration precision, defined as the standard deviation divided by the mean response. Our results show that the inaccuracy and imprecision associated with increasing viewing eccentricity is due to limitations in spatial resolution. Moreover, our results also support the notion that the precise number system is restricted to small numerosities (represented by the subitizing limit), while the approximate number system extends across both small and large numerosities (indexed by variation coefficients) at large eccentricities.

GeneShelf: a web-based visual interface for large gene expression time-series data repositories

A widespread use of high-throughput gene expression analysis techniques enabled the biomedical research community to share a huge body of gene expression datasets in many public databases on the web. However, current gene expression data repositories provide static representations of the data and support limited interactions. This hinders biologists from effectively exploring shared gene expression datasets. Responding to the growing need for better interfaces to improve the utility of the public datasets, we have designed and developed a new web-based visual interface entitled GeneShelf (http://bioinformatics.cnmcresearch.org/GeneShelf). It builds upon a zoomable grid display to represent two categorical dimensions. It also incorporates an augmented timeline with expandable time points that better shows multiple data values for the focused time point by embedding bar charts. We applied GeneShelf to one of the largest microarray datasets generated to study the progression and recovery process of injuries at the spinal cord of mice and rats. We present a case study and a preliminary qualitative user study with biologists to show the utility and usability of GeneShelf.

Failures to ignore entirely irrelevant distractors: the role of load

In daily life (e.g., in the work environment) people are often distracted by stimuli that are clearly irrelevant to the current task and should be ignored. In contrast, much applied distraction research has focused on task interruptions by information that requires a response and therefore cannot be ignored. Moreover, the most commonly used laboratory measures of distractibility (e.g., in the response-competition and attentional-capture paradigms), typically involve distractors that are task relevant (e.g., through response associations or location). A series of experiments assessed interference effects from stimuli that are entirely unrelated to the current task, comparing the effects of perceptual load on task-irrelevant and task-relevant (response competing) distractors. The results showed that an entirely irrelevant distractor can interfere with task performance to the same extent as a response-competing distractor and that, as with other types of distractors, the interfering effects of the irrelevant distractors can be eliminated with high perceptual load in the relevant task. These findings establish a new laboratory measure of a form of distractibility common to everyday life and highlight load as an important determinant of such distractibility.

Distractor effects during processing of words under load

The perceptual load model of attention (Lavie, 1995) suggests that processing of irrelevant distractors depends on the extent to which a relevant task engages full perceptual capacity. Word recognition models suggest that letter perception is facilitated in words relative to nonwords. These models led us to hypothesize that increasing the number of letters would increase perceptual load more for nonwords than for words, and thus would be more likely to exhaust capacity and eliminate distractor processing for nonwords than for words. In support of this hypothesis, we found that increasing the number of search letters increases RTs more for nonwords than for words and only reduces distractor interference for nonwords. Thus, although readers process words more efficiently than nonwords, they also become more prone to distraction when processing words.

The mind and brain of short-term memory

The past 10 years have brought near-revolutionary changes in psychological theories about short-term memory, with similarly great advances in the neurosciences. Here, we critically examine the major psychological theories (the "mind") of short-term memory and how they relate to evidence about underlying brain mechanisms. We focus on three features that must be addressed by any satisfactory theory of short-term memory. First, we examine the evidence for the architecture of short-term memory, with special attention to questions of capacity and how--or whether--short-term memory can be separated from long-term memory. Second, we ask how the components of that architecture enact processes of encoding, maintenance, and retrieval. Third, we describe the debate over the reason about forgetting from short-term memory, whether interference or decay is the cause. We close with a conceptual model tracing the representation of a single item through a short-term memory task, describing the biological mechanisms that might support psychological processes on a moment-by-moment basis as an item is encoded, maintained over a delay with some forgetting, and ultimately retrieved.

Dynamic binding of identity and location information: a serial model of multiple identity tracking

Tracking of multiple moving objects is commonly assumed to be carried out by a fixed-capacity parallel mechanism. The present study proposes a serial model (MOMIT) to explain performance accuracy in the maintenance of multiple moving objects with distinct identities. A serial refresh mechanism is postulated, which makes recourse to continuous attention switching, a capacity-limited episodic buffer for identity-location bindings, indexed location information stored in the visuospatial short-term memory, and an active role of long-term memory. As identity-location bindings are refreshed serially, a location error is inherent for all other targets except the focally attended one. The magnitude of this location error is a key factor in predicting tracking accuracy. MOMIT's predictions were supported by the data of five experiments: performance accuracy decreased as a function of target set-size, speed, and familiarity. A mathematical version of MOMIT fitted nicely to the observed data with plausible parameter estimates for the binding capacity and refresh time.

Grouping and trajectory storage in multiple object tracking: impairments due to common item motions

In our natural viewing, we notice that objects change their locations across space and time. However, there has been relatively little consideration of the role of motion information in the construction and maintenance of object representations. We investigated this question in the context of the multiple object tracking (MOT) paradigm, wherein observers must keep track of target objects as they move randomly amid featurally identical distractors. In three experiments, we observed impairments in tracking ability when the motions of the target and distractor items shared particular properties. Specifically, we observed impairments when the target and distractor items were in a chasing relationship or moved in a uniform direction. Surprisingly, tracking ability was impaired by these manipulations even when observers failed to notice them. Our results suggest that differentiable trajectory information is an important factor in successful performance of MOT tasks. More generally, these results suggest that various types of common motion can serve as cues to form more global object representations even in the absence of other grouping cues.

Attentional selection of moving objects by a serial process

How does the efficiency of attentional selection depend on the number of attended objects in a display? We measured the channel capacity (CC) of human observers during the attentional tracking of moving targets. The relation between CC and target number was used to estimate target-sampling rate. The sampling rate was halved when the number of targets was doubled, indicating that tracking was accomplished by a mechanism whose processing rate did not vary with target number. Systematically varying the dynamic parameters of the display provided inconclusive evidence for the idea that the time interval between successive samples of the same target increased with target number. These results are consistent with the hypothesis that the selection of multiple moving objects involves a limited capacity processor.

Counting in everyday life: discrimination and enumeration

Enumerating the number of items in a set accurately and quickly is a basic mathematical skill. This ability is especially crucial in the more real-life situations, where relevant items have to be discriminated from irrelevant distracters. Although much work has been done on the brain mechanisms and neural correlates of the enumeration and/or discrimination process, no agreement has been reached yet. We used event-related potentials (ERPs) to show the time course of brain activity elicited by a task that involved both enumeration and discrimination at the same time. We found that even though the two processes run to some extent in parallel, discrimination seems to take place mainly in an earlier time window (from 100 ms after the stimulus onset) than enumeration (beyond 200 ms after the stimulus onset). Moreover, electrophysiological evidence based on the N2 and P3 components make it reasonable to argue for the existence of a dichotomy between subitizing (for sets of less than four items) and counting (for sets of four and more items). Source estimation suggests that subitizing and counting, though being distinct brain processes, do recruit similar brain areas.

The Efficiency of Feature-Based Subitization and Counting

The enumeration of small numbers of objects (approximately 4) proceeds rapidly, accurately, and with little effort via a process termed subitization. Four experiments examined whether it was possible to subitize the number of features rather than objects present in a display. Overall, the findings showed that when features are presented randomly and are uncorrelated with object numerosity, efficient enumeration is not possible. This suggests that the visual system does not have parallel access to multiple feature maps and that subitization processes operate exclusively on representations coding the locations of objects. The data are discussed with respect to theories of visual enumeration and search.

Blink and Shrink: The Effect of the Attentional Blink on Spatial Processing

The detection or discrimination of the second of 2 targets in a rapid serial visual presentation (RSVP) task is often temporarily impaired-a phenomenon termed the attentional blink. This study demonstrated that the attentional blink also affects localization performance. Spatial cues pointed out the possible target positions in a subsequent visual search display. When cues were presented inside an attentional blink (as induced by an RSVP task), the observers' capacity to use them was reduced. This effect was not due to attention being highly focused, to general task switching costs, or to complete unawareness of the cues. Instead, the blink induced a systematic localization bias toward the fovea, reflecting what appears to be spatial compression.

Dynamics of attention in depth: evidence from multi-element tracking

We examined the allocation of attention in depth using a multi-element tracking paradigm. Observers were required to track a predefined subset of from two to eight elements in displays containing up to sixteen identical moving elements. We first show that depth cues, such as binocular disparity and occlusion through T-junctions, improve performance in a multi-element tracking task in the case where element boundaries are allowed to intersect in the depiction of motion in a single frontoparallel plane. We also show that the allocation of attention across two perceptually distinguishable planar surfaces, either frontoparallel or receding at a slanting angle and defined by coplanar elements, is easier than allocation of attention within a single surface. The same result was not found when attention was required to be deployed across items of two-color populations rather than across items of a single color. Our results suggest that, when surface information does not suffice to distinguish between targets and distractors that are embedded in these surfaces, division of attention across two surfaces aids in tracking moving targets. A final experiment with populations of elements moving within distinct volumes produced similar results, suggesting that spatial separation in three dimensions, rather than confinement to surfaces as such, may explain the improved performance for the two-surface case.

The role of perceptual load in processing distractor faces

It has been established that successful ignoring of irrelevant distractors depends on the extent to which the current task loads attention. However, the previous load studies have typically employed neutral distractor stimuli (e.g., letters). In the experiments reported here, we examined whether the perception of irrelevant distractor faces would show the same effects. We manipulated attentional load in a relevant task of name search by varying the search set size and found that whereas congruency effects from meaningful nonface distractors were eliminated by higher search load, interference from distractor faces was entirely unaffected by search load. These results support the idea that face processing may be mandatory and generalize the load theory to the processing of meaningful and more complex nonface distractors.

Visual marking inhibits singleton capture

This paper is concerned with how we prioritize the selection of new objects in visual scenes. We present four experiments investigating the effects of distractor previews on visual search through new objects. Participants viewed a set of to-be-ignored nontargets, with the task being to search for a target in a second set, added to the first after 1000ms. This second set could contain a salient feature singleton, defined in terms of its color, orientation, or both color and orientation. When the singleton was a distractor, search was slowed relative to when there was no singleton. Search was facilitated when the singleton was a target. Interestingly, both the interference and facilitation effects were modulated when the preview shared features with the singleton. Follow-up experiments showed that this reduction of singleton effects was not due to: (i) low-level sensory aspects of the displays, (ii) increased heterogeneity in the search set in the preview condition, or (iii) color-based grouping of old and new items. Instead, we suggest that there is an inhibitory carry-over from the first to the second set of items based on feature similarity. We suggest the suppression stems from a process termed visual marking, which suppresses irrelevant visual objects in anticipation of more relevant new objects (Watson & Humphreys, 1997). The findings argue against alternative explanations such as the automatic capture by abrupt new onsets account.

The development of visual attention in infancy

Over the past decade, the study of attention in infancy has seen dramatic progress. This review delineates four attentional functions (alertness, spatial orienting, attention to object features, and endogenous attention) that are relevant to infancy and uses these functions as a framework for summarizing the developmental course of attention in infancy. Rudimentary forms of various attentional functions are present at birth, but each of the functions exhibits different and apparently dissociable periods of postnatal change during the first years of life. The role of attention in development should therefore be considered in the context of interaction among different systems at different levels of maturity during the first years of life.

Visual indexes, preconceptual objects, and situated vision

This paper argues that a theory of situated vision, suited for the dual purposes of object recognition and the control of action, will have to provide something more than a system that constructs a conceptual representation from visual stimuli: it will also need to provide a special kind of direct (preconceptual, unmediated) connection between elements of a visual representation and certain elements in the world. Like natural language demonstratives (such as 'this' or 'that') this direct connection allows entities to be referred to without being categorized or conceptualized. Several reasons are given for why we need such a preconceptual mechanism which individuates and keeps track of several individual objects in the world. One is that early vision must pick out and compute the relation among several individual objects while ignoring their properties. Another is that incrementally computing and updating representations of a dynamic scene requires keeping track of token individuals despite changes in their properties or locations. It is then noted that a mechanism meeting these requirements has already been proposed in order to account for a number of disparate empirical phenomena, including subitizing, search-subset selection and multiple object tracking (Pylyshyn et al., Canadian Journal of Experimental Psychology 48(2) (1994) 260). This mechanism, called a visual index or FINST, is briefly discussed and it is argued that viewing it as performing a demonstrative or preconceptual reference function has far-reaching implications not only for a theory of situated vision, but also for suggesting a new way to look at why the primitive individuation of visual objects, or proto-objects, is so central in computing visual representations. Indexing visual objects is also, according to this view, the primary means for grounding visual concepts and is a potentially fruitful way to look at the problem of visual integration across time and across saccades, as well as to explain how infants' numerical capacity might arise.

The role of attention in motion extrapolation: are moving objects 'corrected' or flashed objects attentionally delayed?

Objects flashed in alignment with moving objects appear to lag behind [Nijhawan, 1994 Nature (London) 370 256-257]. Could this 'flash-lag' effect be due to attentional delays in bringing flashed items to perceptual awareness [Titchener, 1908/1973 Lectures on the Elementary Psychology of Feeling and Attention first published 1908 (New York: Macmillan); reprinted 1973 (New York: Arno Press)]? We overtly manipulated attentional allocation in three experiments to address the following questions: Is the flash-lag effect affected when attention is (a) focused on a single event in the presence of multiple events, (b) distributed over multiple events, and (c) diverted from the flashed object? To address the first two questions, five rings, moving along a circular path, were presented while observers attentively tracked one or multiple rings under four conditions: the ring in which the disk was flashed was (i) known or (ii) unknown (randomly selected from the set of five); location of the flashed disk was (i) known or (ii) unknown (randomly selected from ten locations). The third question was investigated by using two moving objects in a cost-benefit cueing paradigm. An arrow cued, with 70% or 80% validity, the position of the flashed object. Observers performed two tasks: (a) reacted as quickly as possible to flash onset; (b) reported the flash-lag effect. We obtained a significant and unaltered flash-lag effect under all the attentional conditions we employed. Furthermore, though reaction times were significantly shorter for validly cued flashes, the flash-lag effect remained uninfluenced by cue validity, indicating that quicker responses to validly cued locations may be due to the shortening of post-perceptual delays in motor responses rather than the perceptual facilitation. We conclude that the computations that give rise to the flash-lag effect are independent of attentional deployment.

Visual marking in moving displays: feature-based inhibition is not necessary

Visual marking is a mechanism by which new visual stimuli can gain a selection advantage by the top-down attentional inhibition of stimuli already in the field. Previous work (Olivers, Watson, & Humphreys, 1999) has shown that, for moving stimuli, there must be a unique feature difference between the old items and the new items for marking to occur. The present study shows that this constraint is not necessary if the local spatial relationships between the old moving items remain constant. It is proposed that, with a fixed configuration, the old moving items can be grouped to form a single object. An inhibitory template set up to represent the object then coordinates the application of inhibition to the individual stimuli. Implications for the theory and ecological flexibility of visual marking are discussed.

Situating vision in the world

Recently, there has been a great deal of interest in what has been called 'situated cognition', which has included claims that certain forms of representation are inadequate for modeling active organisms or agents such as humans and robots. In this article, I suggest that a weakness in classical theories of visual representation is the way in which representations connect with the real world, which may account for many of the concerns expressed by the situated cognition community. Specifically, I claim that what current theories lack is any provision for a certain form of direct, preconceptual connection between objects in the visual world (visual objects or proto-objects) and their representations in the visual system. This type of connection is akin to what philosophers and semanticists have referred to as an 'indexical' or 'demonstrative' reference and what some cognitive scientists have referred to as 'deictic pointers'. I explain why such a mechanism is needed and suggest that many workers have, in fact, been studying precisely this under the term 'visual index'. The visual index hypothesis is illustrated with the results of some relevant experiments, including multiple object tracking, visual routines and subset-selected visual searches. Indexing theory provides a synthesis that has profound implications for explaining a wide range of psychophysical findings, certain results in infant cognitive development and also some ancient problems in the philosophy of mind.

Undercounting features and missing features: evidence for a high-level deficit in strabismic amblyopia

Abnormal visual development in strabismic amblyopia drastically affects visual perception and properties of neurons in primary visual cortex (V1). To test the notion that amblyopia also has consequences for higher visual areas, we asked humans with amblyopia to count briefly presented features. Using the amblyopic eye, strabismic amblyopes counted inaccurately, markedly underestimating the number of features. This inaccuracy was not due to low-level considerations (blur, visibility, crowding, undersampling or topographical jitter), as they also underestimated the number of features missing from a uniform grid. Rather, counting deficits in strabismic amblyopes reflected a higher-level limitation in the number of features the amblyopic visual system can individuate.

Visual marking: evidence for inhibition using a probe-dot detection paradigm

Watson and Humphreys (1997, 1998) have recently demonstrated that new objects can be prioritized for visual attentional processing by the top-down attentional inhibition of old objects already in the field, a mechanism they called visual marking. The experiments reported here show that the detection of a dim probe dot is impaired when it falls at the location of an old object (Experiments 1 and 3) but that this occurs only in conditions in which it is advantageous for subjects to mark (inhibit) old objects (Experiment 2). These results further support previous work showing that visual marking is based on the inhibition of the locations of old objects and that visual marking can be flexibly applied (or withheld), depending on the goals of current behavior.

Stereoscopic depth but not shape perception from second-order stimuli

Depth can be seen using either linear (first-order) or non-linear (second-order) stereo micropatterns when, in the latter, contrast envelopes contain the disparity information. We examined whether a second-order mechanism can contribute to the perception of 3-D surface shape. Using a variety of different stimulus types, we show that for each, shape is easy to see with linear stimuli. Over a wide range of parameters however, none of our observers perceived shape, however faintly, from the non-linear stimuli. To explore why these elements failed, we simplified our stimulus to a step-edge in depth and measured performance while varying the number of elements. We show how performance declined when more than two non-linear elements were used. We discuss reasons for the limitation found for non-matching elements, including a dissociation for stereopsis between seeing surface shape and depth.

Tracking multiple items through occlusion: clues to visual objecthood

In three experiments, subjects attempted to track multiple items as they moved independently and unpredictably about a display. Performance was not impaired when the items were briefly (but completely) occluded at various times during their motion, suggesting that occlusion is taken into account when computing enduring perceptual objecthood. Unimpaired performance required the presence of accretion and deletion cues along fixed contours at the occluding boundaries. Performance was impaired when items were present on the visual field at the same times and to the same degrees as in the occlusion conditions, but disappeared and reappeared in ways which did not implicate the presence of occluding surfaces (e.g., by imploding and exploding into and out of existence instead of accreting and deleting along a fixed contour). Unimpaired performance did not require visible occluders (i.e., Michotte's tunnel effect) or globally consistent occluder positions. We discuss implications of these results for theories of objecthood in visual attention.