Search

Funder: University of Cambridge, Another person’s gaze direction is a rich source of social information, especially eyes gazing toward prominent or relevant objects. To guide attention to these important stimuli, visual search mechanisms may incorporate sophisticated coding of eye-gaze and its spatial relationship to other objects. Alternatively, any guidance might reflect the action of simple perceptual ‘templates’ tuned to visual features of socially relevant objects, or intrinsic salience of direct-gazing eyes for human vision. Previous findings that direct gaze (toward oneself) is prioritised over averted gaze do not distinguish between these accounts. To resolve this issue, we compared search for eyes gazing toward a prominent object versus gazing away, finding more efficient search for eyes ‘gazing toward’ the object. This effect was most clearly seen in target-present trials when gaze was task-relevant. Visual search mechanisms appear to specify gazer-object relations, a computational building-block of theory of mind.

Direct gaze - someone looking at you - is an important and subjectively-salient stimulus. Its processing is thought to be enhanced by the brain's internalised predictions - priors - that effectively specify it as the most likely gaze direction. Current consensus holds that, befitting its presumed importance, direct gaze attracts attention more powerfully than other gazes. Conversely, some Predictive Coding (PC) models, in which exogenous attention is drawn to stimuli that violate predictions, may be construed as making the opposite claim - i.e., exogenous attention should be biased away from direct gaze (which conforms to internal predictions), toward averted gaze (which does not). Here, searching displays with salient, 'odd-one-out' gazes, we observed attentional bias (in rapid, initial saccades) toward averted gaze, as would be expected by PC models. However, this pattern obtained only when conditions highlighted gaze-uniqueness. We speculate that, in our experiments, task requirements determined how prediction influenced perception.

Direct gaze-someone gazing at you-is an important social cue that might be expected to capture visual attention, even in the presence of other faces. Consistent with this, direct gazing eyes are often detected more rapidly in arrays of averted gazing eyes, than vice versa; a search asymmetry termed the stare-in-the-crowd effect (SITCE). Here, we examine top-down influences on the SITCE by manipulating observers' knowledge of the target's gaze prior to the search display. Our findings revealed two dissociable components of the SITCE. The first, which scaled with set size but was unaffected by prior knowledge, was attributed to noisy, parallel gaze processing that guides attention toward direct gaze (Process 1). The second, an overall response time advantage for direct versus averted gaze targets, irrespective of set size, was attributed to criteria for determining target presence versus absence (Process 2). Prior knowledge of the target's gaze direction increased the direct gaze advantage, rather than speeding up responses for both target types (typically expected for 100% valid cues). This unusual pattern suggests that top-down gaze-related influences may comprise an obligatory bias toward direct gaze. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

‘Heuristic’ theories of autism postulate that a single mechanism or process underpins the diverse psychological features of autism spectrum disorder. Although no such theory can offer a comprehensive account, the parsimonious descriptions they provide are powerful catalysts to autism research. One recent proposal holds that ‘noisy’ neuronal signalling explains not only some deficits in autism spectrum disorder, but also some superior abilities, due to ‘stochastic resonance’. Here, we discuss three distinct actions of noise in neural networks, arguing in each case that autism spectrum disorder symptoms reflect too little, rather than too much, neural noise. Such reduced noise, perhaps a function of atypical brainstem activation, would enhance detection and discrimination in autism spectrum disorder but at significant cost, foregoing the widespread benefits of noise in neural networks.

Atypical high-level vision in autism is sometimes attributed to a core deficit in the function of lateral geniculate nucleus magnocells or their retinal drives. While some physiological measures provide indirect, suggestive evidence for such a deficit, support from behavioural measures is lacking and contradictory. We assessed luminance contrast increment thresholds on pulsed- and steady- pedestals in 17 children with autism spectrum conditions (ASC) compared to 17 typically developing children; these two conditions correspond to widely-used indices of magnocellular and parvocellular function. As a group, children with ASC had strikingly elevated thresholds on the steady pedestal-paradigm, yet performed similarly to controls on the pulsed pedestal paradigm, a finding that would typically be interpreted to reflect impaired magnocellular function. The effect size of the impairment was large and a substantial minority (41.2%) of the ASC group showed significantly impaired performance on an individual basis. This finding is consistent with a selective magnocellular deficit. It directly contradicts previous claims that such deficits are confined to 'complex' visual stimuli and likely does not reflect atypical attention, adaptation or high-level vision. The pattern of results is not clearly predicted by notions of imbalance of excitation versus inhibition, atypical lateral connectivity or enhanced perceptual function that account for a range of other findings associated with perception in autism. It may be amenable to explanation in terms of decreased endogenous neural noise, a novel alternative we outline here.

When presented with a face stimulus whose gaze is diverted, observers' attention shifts to locations fixated by the face. Such "gaze following" has been characterized by some previous studies as a consequence of sophisticated theory of mind processes, but by others (particularly those employing the "gaze-cuing" paradigm) as an involuntary response that is triggered directly and reflexively by the physical features of a face. To address this apparent contradiction, we modified the gaze-cuing paradigm using a deception procedure to convince observers that prerecorded videos of an experimenter making head turns and wearing mirrored goggles were a "live" video link to an adjacent room. In two experiments, reflexive gaze following was found when observers believed that the model was wearing transparent goggles and could see, but it was significantly reduced when they believed that the experimenter wore opaque goggles and could not see. These results indicate that the attribution of the mental state "seeing" to a face plays a role in controlling even reflexive gaze following.

Previous studies in animals have revealed effects of both prenatal and current testosterone on brain organization and behavior. However, it is unclear how these effects translate to the human brain. Here, we refine previous procedures to clarify the relative contribution of prenatal versus current testosterone indices to cognitive function. Sixty-nine subjects performed four computerized tasks measuring mental rotation, targeting, figure-disembedding and perceptual discrimination. Using stepwise regression analyses, performance was related to free testosterone assayed in saliva samples and to second-to-fourth finger length ratios (2D:4D), a putative index of prenatal testosterone exposure. The exclusive predictor for mental rotation was found to be sex, while 2D:4D was found to be the sole predictor of targeting, exhibiting a curvilinear relation, and figure-disembedding performance, showing a linear relation. These findings suggest a substantial role for prenatal testosterone but not current testosterone in determining cognitive performance.

Several studies have reported that individuals with autism and Asperger's syndrome show a local processing bias on tasks involving features and configurations. This study assessed whether this bias results from differences in the perception of features or a cognitive bias to attend to features in autism as a consequence of a deficit in attending to configurations. Children with autism and typically developing children performed a task assessing the initial perceptual representation of features and configurations following a 50 ms stimulus display and the development of the perceptual representation by grouping processes following an 800 ms stimulus display. No differences were observed between the two groups, suggesting that the perceptual and attentional mechanisms marshalled by this task operate typically in children with autism.

Subitization, the rapid enumeration of small numbers of objects, is often held to reflect a general ability to process three or four items simultaneously. However, questions as to which types of visual representations count as items for subitization have, thus far, received relatively little attention. In several new experiments, we instructed participants to enumerate three or four colored regions in a display. In some trials, all of these regions belonged to separate visual objects, whereas in others, two or more belonged to the same object(s). We found far slower enumeration in the latter case than in the former case, a result that cannot be attributed to Stroop interference from the number of attended objects in a display. We conclude that subitization mechanisms cannot operate on small fragments of shape and color alone, even when such fragments are easily individuated from one another. Rather, such mechanisms are compelled to operate on object representations.

In divided-attention tasks, observers must make speeded (or near threshold accuracy) judgements concerning two target features in a display. Typically, when the two features belong to the same object they are more rapidly judged than when they belong to separate objects, a pattern of findings referred to here as a “same-object benefit”. However, we note here that many of these studies share common features, in particular the use of pre-exposed, outline, and/or overlapping objects, and their findings may not generalize to other types of display. Building substantially on previous work by Davis, Welch, Holmes, and Shepherd (2001), we show in four new studies that once these features are not present in a divided-attention task, no same-object benefits are reported. Rather we now find “same-object costs”, where features belonging to a single object are less rapidly judged than features belonging to separate objects.

Much previous research into visual processing in migraine supports a model of abnormal cortical processing, in between the headache attacks, that is characterised by hyperexcitability, heightened responsiveness, a lack of habituation and/or a lack of intra-cortical inhibition. Shepherd (2001) reported two adaptation studies that challenged this view, one using the tilt after-effect, the second using the motion after-effect. Models of cortical function in migraine based on hyperexcitability and a lack of inhibition lead to specific predictions in an adaptation study: there should have been smaller after-effects in people with migraine than in people without. Both experiments, however, revealed larger after-effects in the migraine group than in the control group. Here, these results are extended to the simultaneous tilt illusion and an identical pattern of results was obtained: there were consistently larger effects in the migraine group than in the control group. The results from the three experiments are not consistent with a lack of inhibition in migraine. The results are discussed in terms of alternative models of cortical function, including a lack of excitation and reduced central energy reserves.

Several previous studies have suggested that we may attend only a fixed number of ‘objects’ at a time. However, whereas findings from two-target experiments suggest that we can attend only one object at a time, other results from object-tracking and enumeration paradigms point instead to a four-object limit. Here, we note that in these previous studies the number of objects covaried with the overall size and complexity of the stimulus, such that apparent one-object or four-object limits in those tasks may reflect changes in the complexity of attended stimuli, rather than the number of objects per se. Accordingly, in the current experiments we employ stimuli in which the number of objects varies, while overall size and complexity are held constant. Using these refined measures of object-based effects, we find no evidence for a one-object or four-object limit on attention. Indeed, we conclude that the number of attended objects does not affect how efficiently we can attend a given stimulus. We propose and test an alternative approach to object-based attention limitations based on within-object and between-object feature-binding mechanisms in human vision.

Many previous studies have found that we can attend pairs of visual features (e.g., colour, orientation) more efficiently when they belong to the same “object” compared to when they belong to separate, neighbouring objects (e.g., Behrmann, Zemel, & Mozer, 1998; Egly, Rafal, & Driver, 1994). This advantage for attending features from the same object may reflect stronger binding between these features than arises for pairs of features belonging to separate objects. However, recent findings described by Davis, Welch, Holmes, and Shepherd (in press) suggest that under specific conditions this same-object advantage can be reversed, such that attention now spreads more readily between features belonging to separate neighbouring objects than between features of the same object. In such cases it would appear that features belonging to separate visual objects are more strongly bound than features of the same object. Here I review these findings and present the results of a new study. Together these data suggest th...

An enduring controversy in neuroscience concerns how the brain “binds” together separately coded stimulus features to form unitary representations of objects. Recent evidence has indicated a close link between this binding process and 40-hertz (gamma-band) oscillations generated by localized neural circuits. In a separate line of research, the ability of young infants to perceive objects as unitary and bounded has become a central focus for debates about the mechanisms of perceptual development. Here we demonstrate that binding-related 40-hertz oscillations are evident in the infant brain around 8 months of age, which is the same age at which behavioral and event-related potential evidence indicates the onset of perceptual binding of spatially separated static visual features.

Four experiments examined whether Kanizsa subjective figures can induce amodal completion of a notched circle at parallel stages of visual search. Search for the notched circle among full circles was slow and inefficient when the notched circle appeared stereoscopically behind an abutting subjective surface, as if occluded by it. However, search became efficient and parallel when the notched and completed circles appeared nearer so that the subjective figures could not act as occluders. Control studies ruled out explanations in terms of low spatial frequencies, grouping of aligned edges, or the depth of the circles per se. Multiple Kanizsa subjective figures can be coded in parallel as occluding surfaces; such coding is obligatory because it arises even when highly detrimental to performance.

Evidence suggests that subliminal stimuli can influence ostensibly volitional, executive processes but it is unclear whether this is highly task-specific. To address this we used a set-switching task. Volunteers saw a word pair and reported either if both words had the same number of syllables or if both were concrete. Task selection was random and instructed by a hexagon/triangle preceding the words. A subliminally-presented square or diamond reliably preceded each of these consciously perceived instruction-shapes. Significant congruency effects were observed in a subsequent Test Phase in which primes no longer reliably predicted the task (and in which high/low tones now served as conscious instructions). The Generalization Phase required novel phonological (rhyme) or semantic (category) judgments. Remarkably, unconscious priming congruency effects carried over in those participants who had shown priming in the Test Phase, the degree correlating across the two conditions. In a final phase of the study, participants were asked to discriminate between the two originally presented prime shapes. Those participants whose discriminations were more accurate showed reduced priming relative to participants with less accurate discriminations. The results suggest that, rather than being highly task specific, priming can operate at the level of a generalizable process and that greater awareness of primes may lessen their impact on behavior.

Regions of objects that are partially obscured in the current retinal image are often perceptually filled in by the visual system (Kantzsa, 1979) In some cases (modal completion), this causes the filled-in region to appear tinged with the color and brightness of unobscured parts of that object, but in other cases (amodal completion), it does not (Michotte & Burke, 1951) It has recently been argued that modal and amodal completion both arise in preattentive vision, and may operate equivalently at that level (Davis & Driver, 1994, He & Nakayama, 1992, Shipley & Kellman, 1992) In this article, we show that they have different effects on attentive vision, with attention tending to spread to (and from) modally completed regions and their visible inducers, but not to (or from) comparable amodally completed regions and their inducers This finding is consistent with visual attention operating on surfaces (eg, He & Nakayama, 1995) in a viewer-centered representation of the scene, after the operation of filling-in processes.

Face perception skills are fundamental to social cognition, yet they vary markedly across individuals. This diversity has stimulated controversy over the relative contributions of genetic factors, prenatal environment, and postnatal experience to face perception. Recent twin studies have found that face perception is heritable, highlighting the potential for substantial prenatal determination of these skills. In contrast, previous work on potential influences of the prenatal hormonal environment on social cognition have found no association between 2D:4D, a marker for prenatal androgen exposure, and processing of facial emotional expressions, apparently precluding a major role for prenatal hormones in governing face perception. We propose that substantial predictive relationships between 2D:4D and face perception may have been masked in previous work because the task employed required both face perception and processing of others’ emotions, the latter component being greatly influenced by circulating hormones in adults. To assess prenatal hormone influences on face perception without requiring emotion processing, we related 2D:4D to the face inversion effect (FIE), a measure of the recruitment of specialist face processes to visual perception. The magnitude of the resulting predictive relationship (r = .52) compared surprisingly well with that found between identical twins’ face perception (rs = .27–.61), suggesting that mechanisms of 2D:4D may account for a substantial proportion of perinatal influences on face processing. Furthermore, we employed 2D:4D as a common scale to map individual differences in the FIE onto prenatal testosterone:estrogen ratios, assayed by Lutchmaya, Baron-Cohen, Raggatt, Knickmeyer, and Manning (Early Human Development 77:23–28, 2004). The substantial overlap between the two data sets further implicates prenatal steroids in adult face perception skills.

Task selection, previously thought to operate only under conscious, voluntary control, can be activated by unconsciously-perceived stimuli. In most cases, such activation is observed for unconscious stimuli that closely resemble other conscious, task-relevant stimuli and hence may simply reflect perceptual activation of consciously established stimulus-task associations. However, other studies have reported ‘direct’ unconscious-stimulus influences on task selection in the absence of any conscious, voluntary association between that stimulus and task (e.g., Zhou and Davis, 2012). In new experiments, described here, these latter influences on cued- and free-choice task selection appear robust and long-lived, yet, paradoxically, are suppressed to undetectable levels following momentary conscious prime-task pairing. Assessing, and rejecting, three intuitive explanations for such suppressive effects, we conclude that conscious prime-task pairing minimizes non-strategic influences of unconscious stimuli on task selection, insulating endogenous choice mechanisms from maladaptive external control.

Unconscious stimuli activate task sets, mental programs that orchestrate performance of complex tasks, but the role of attention in such effects has not been addressed. In previous studies, unconscious prime stimuli appeared at attended locations and were explicitly specified in the task instructions; spatial attention to the prime and/or a specific conscious attentional set may thus be required for such unconscious activation to arise. In the present experiments, a learning phase established associations between unconscious prime stimuli and performance of two tasks. These associations influenced task performance in a subsequent test phase, even though the primes were not specified in current task instructions. This is the first demonstration that unconscious stimuli can prime task sets independently of a current attentional set that specifies stimulus-task mappings. Such priming was not influenced by spatial attention cues, in contrast to clear attention influences in comparison trials that mimicked conditions employed by previous studies.

A pervasive integration deficit could provide a powerful and elegant account of cognitive processing in autism spectrum disorders (ASD). However, in the case of visual Gestalt grouping, typically assessed by tasks that require participants explicitly to introspect on their own grouping perception, clear evidence for such a deficit remains elusive. To resolve this issue, we adopt an index of Gestalt grouping from the object-based attention literature that does not require participants to assess their own grouping perception. Children with ASD and mental- and chronological-age matched typically developing children (TD) performed speeded orientation discriminations of two diagonal lines. The lines were superimposed on circles that were either grouped together or segmented on the basis of color, proximity or these two dimensions in competition. The magnitude of performance benefits evident for grouped circles, relative to ungrouped circles, provided an index of grouping under various conditions. Children with ASD showed comparable grouping by proximity to the TD group, but reduced grouping by similarity. ASD seems characterized by a selective bias away from grouping by similarity combined with typical levels of grouping by proximity, rather than by a pervasive integration deficit.

Optimizing our responses to physical threats is of clear adaptive value, yet influences of threat perception specific to visual guidance of action have received little attention. Here, we assess effects of threat perception on a targeting task, finding improved performance for peripheral targets relative to central targets. These effects of threat were absent either when the task did not involve targeting or when targeting followed high-arousal, positive-valence stimuli. Additionally, the effects of threat showed a nasal- versus temporal-visual hemifield asymmetry and were absent when target stimuli isolated S-cone photoreceptor function. These findings suggest that the superior colliculus, which is activated following threat perception, shows clear visual-field asymmetry, receives no direct S-cone-based inputs from the retina, and forms part of a hypothesized direct brainstem-amygdala-cortex alarm system, mediates this effect.

In contrast to standard search templates that specify a target object's expected features, templates for rejection (TFR) may specify features of nontargets, biasing attention away from irrelevant objects. Little is known about TFR, and virtually nothing is known about their role in guiding search across natural scenes. In such scenes, targets and nontargets may not be easily distinguished on the basis of their visual features; it has been claimed that standard search templates may therefore specify target objects' semantic features to guide attention. Here, we ask whether TFR can do so. Noting a limitation of previous procedures used to study standard search templates, we trialed an alternative method to examine semantic templates for nontarget exclusion in natural scene search. We found that when nontargets belonged unpredictably to either of two physically distinct categories, search was less efficient than when targets belonged to one known category. This two-category cost, attributed to inefficient application of search templates, was absent for two physically dissimilar but semantically related categories. Adding a training phase to highlight semantic distinctiveness of two object categories reinstated the two-category cost, precluding stimulus-based accounts of the effect. These patterns were not observed for one-image displays or when observers searched for object categories rather than ignoring them, demonstrating their specificity to TFR, the inadequacy of search-and-destroy models to account for them, and likely basis in attentional guidance. TFR can specify semantic information to guide attention away from nontargets.

Luck and Vogel (1997) have reported several striking results in support of the view that visual short-term memory (VSTM) has a fixed capacity of four objects, irrespective of how many relevant features those objects comprise. However, more recent studies have challenged this account, indicating only a weak effect of the number of objects once other factors are more evenly equated across conditions. Here, we employed a symmetry manipulation to verify object segmentation in our displays, to demonstrate that when spatial and masking factors are held constant, the number of objects per se has no effect on VSTM. Instead, VSTM capacity may reflect the number of object "parts" or feature conjunctions in a given display.

Unilateral brain damage frequently produces “extinction,” in which patients can detect brief single visual stimuli on either side but are unaware of a contralesional stimulus if presented concurrently with an ipsilesional stimulus. Explanations for extinction have invoked deficits in initial processes that operate before the focusing of visual attention or in later attentive stages of vision. Preattentive vision was preserved in a parietally damaged patient, whose extinction was less severe when bilateral stimuli formed a common surface, even if this required visual filling-in to yield illusory Kanizsa figures or completion of partially occluded figures. These results show that parietal extinction arises only after substantial processing has generated visual surfaces, supporting recent claims that visual attention is surface-based.

Subjective figures, seen in the absence of luminance gradients (Fig. 1), provide a phenomenal illusion that can be related to the properties of single cells in the visual cortex, offering a rare bridge between brain function and visual awareness. It remains controversial whether subjective figures arise from intelligent cognitive mechanisms, or from lower-level processes in early vision. The cognitive account implies that the perception of subjective figures may require serial attentive processing, whereas on the low-level account they should arise in parallel at earlier visual stages. Physiological evidence apparently fits a low-level account and indicates that some types of subjective contour may be detected earlier than the conventional Kanizsa type. Here we report that even Kanizsa subjective figures can be detected without focal attention at parallel stages of the human visual system.

The complex relationship between attention and STM forms a core issue in the study of human cognition, and Cowan's target article attempts, quite successfully, to elucidate an important part of this relationship. However, while I agree that aspects of STM performance may reflect the action mechanisms that we normally consider to subserve “attention” I shall argue here that attention is not subject to a fixed four-object capacity limit as Cowan suggests. Rather, performance in attention tasks as well as STM may be best accounted for in terms of decay and interference.

Comparisons between modally and amodally completed regions show that perceptual filling-in is not merely the ignoring of absences. Illusory filled-in colour arises for modal completion, but not for amodal completion in comparable displays. We find that attention spreads automatically to modally but not amodally completed regions from their inducers, revealing a functional effect of filled-in colour.

Support for object-based accounts of visual attention has been drawn from several different types of effect. One effect is found when observers try to restrict their attention to a particular region of a display. Other regions belonging to the same object are often selected as well, suggesting that attention spreads spatially over entire objects. Another effect is found when judging two visual attributes; performance is often less efficient when the attributes belong to separate objects rather than both belonging to a single object. This latter effect has been taken to imply that only one segmented object can be attended at a time. However, it may instead merely be a variant of the first effect. If, as we assume here, attention spreads to task-irrelevant regions of relevant objects, it will encompass a larger spatial region and more information when judging attributes of two objects rather than one. Here we compared judging one versus two objects, while manipulating whether the two objects occupied a wider extent than the single object condition (as in previous work), or not. Costs were found for judging two objects versus one only when together they occupied a wider spatial extent. We conclude that reported difficulties in attending two objects may be due to attention spreading across the entire spatial extent of objects when judging their parts, rather than a fixed inability to process more than object at a time.