Capacity limits for face processing

We present three experiments in which subjects were asked to make speeded sex judgements (Experiment 1) or semantic judgements (Experiments 2 and 3) to face targets and nonface items, while ignoring a solitary flanking distractor face or a nonface stimulus. Distractors could be either congruent (same response category) or incongruent (different response category) with the target. Distractor congruency effects were consistently observed in all combinations of target-distractor stimulus pairs, except when a distractor face flanked a target face. The failure to find congruency effects in this condition was explored further in a fourth experiment, in which four task-irrelevant flankers were simultaneously presented. Once again, no face-face congruency effects were found, even though comparison distractors interfered with face and nonface targets alike. However, four simultaneously presented distractor faces did not interfere with nonface targets either. We suggest that these experiments demonstrate a capacity limit for visual processing in these conditions, such that no more than one face is processed at a time.     

Similarity modulates the face-capturing effect in change detection

We investigated whether similarity among faces could modulate the face-capturing effect in change detection. In Experiment 1, a singleton search task was used to demonstrate that a face stimulus captures attention and the odd-one-out hypothesis cannot account for the results. Searching for a face target was faster than searching for a nonface target no matter whether distractor–distractor similarity was low or high. The fast search, however, did not lead to a face-detection advantage in Experiment 2 when the pre- and postchange faces were highly similar. When participants in Experiment 3 had to divide their attention between two faces in stimulus displays for change detection, detection performance was worse than performance in detecting nonface changes. The face-capturing effect alone is insufficient to produce the face-detection advantage. Face processing is efficient but its effect on performance depends on the stimulus–task context.     

Inhibition of return for the discrimination of faces

When a target appears unpredictably in the same rather than a different location relative to a preceding onset cue, reaction times (RTs) of participants tasked with responding to the target are slowed. This pattern of results, referred to as inhibition of return (IOR), is believed to reflect the operation of a mechanism that prevents perseverative search of nontarget locations. On the grounds that an evolved mechanism might be sensitive to social stimuli, Taylor and Therrien (2005) examined IOR for localization responses under conditions in which cues and targets could be intact face configurations or nonface configurations; contrary to their predictions, there was no influence of cue or target configuration on the magnitude of IOR, indicating that the mere occurrence of task-irrelevant face and nonface stimuli does not alter IOR. In the present study, we further examined this issue in a task that required a face/nonface target discrimination. When target configuration was thereby made task relevant, we found that IOR differed for face and nonface targets in terms of magnitude (when a single cue-target stimulus onset asynchrony was employed) and time course. We suggest that the RT delay associated with IOR may enable additional processing time and/or response selection when a task-relevant face is presented at the cued location.     

Perceptual load effects on processing distractor faces indicate face-specific capacity limits

The claim that face perception is mediated by a specialized “face module” that proceeds automatically, independently of attention (e.g., Kanwisher, 2000) can be reconciled with load theory claims that visual perception has limited capacity (e.g., Lavie, 1995) by hypothesizing that face perception has face-specific capacity limits. We tested this hypothesis by comparing the effects of face and nonface perceptual load on distractor face processing. Participants searched a central array of either faces or letter strings for a pop star versus politician's face or name and made speeded classification responses. Perceptual load was varied through the relevant search set size. Response competition effects from a category-congruent or -incongruent peripheral distractor face were eliminated with more than two faces in the face search task, but were unaffected by perceptual load in the name search task. These results support the hypothesis that face perception has face-specific capacity limits and resolve apparent discrepancies in previous research.     

The role of attention and familiarity in face identification

How is attention allocated during face identification? Previous work using famous and unfamiliar faces suggests that either no attention or a special attentional mechanism is required. We used a conventional attentional blink (AB) procedure to measure face identification with temporarily reduced attention. The participants viewed a rapid series of face images with one embedded nonface abstract pattern (T1). They judged the texture of T1 and then detected a prespecified face (T2) presented at varying lags after T1. T2 was either famous or unfamiliar, as were distractor faces. Regardless of distractor type, detection of an unfamiliar T2 face was significantly impaired at short versus long T1-T2 lags, indicating an attentional requirement for face identification. Detection of a famous T2 face was unaffected by lag, suggesting that familiarity protects against atemporal attentional bottleneck These findings do not support propositions that face identification is "special" in its need for attentional control     

Fearful faces modulate looking duration and attention disengagement in 7-month-old infants

The present study investigated whether facial expressions modulate visual attention in 7-month-old infants. First, infants' looking duration to individually presented fearful, happy, and novel facial expressions was compared to looking duration to a control stimulus (scrambled face). The face with a novel expression was included to examine the hypothesis that the earlier findings of greater allocation of attention to fearful as compared to happy faces could be due to the novelty of fearful faces in infants' rearing environment. The infants looked longer at the fearful face than at the control stimulus, whereas no such difference was found between the other expressions and the control stimulus. Second, a gap/overlap paradigm was used to determine whether facial expressions affect the infants' ability to disengage their fixation from a centrally presented face and shift attention to a peripheral target. It was found that infants disengaged their fixation significantly less frequently from fearful faces than from control stimuli and happy faces. Novel facial expressions did not have a similar effect on attention disengagement. Thus, it seems that adult-like modulation of the disengagement of attention by threat-related stimuli can be observed early in life, and that the influence of emotionally salient (fearful) faces on visual attention is not simply attributable to the novelty of these expressions in infants' rearing environment.     

Orientation of attention to nonconsciously recognised famous faces

The nonconscious orientation of attention to famous faces was investigated using masked 17 ms stimulus exposure. Each trial presented a simultaneous pair of one famous and one unfamiliar face, matched on physical characteristics, one each in left visual field (LVF) and right visual field (RVF). These were followed by a dot probe in either LVF or RVF to which participants made a speeded two-alternative forced-choice discrimination response. Participants subsequently evaluated the affective valence (good/evil) of the famous persons on a 7-point scale. Higher accuracy of dot probe discrimination in the same visual field (VF) as the famous face suggested that attention was oriented towards faces of persons evaluated “good”, but a reverse orientation effect was observed for those evaluated “evil”. The awareness check presented the same face pairs under the same conditions, and participants were at chance in a task of selecting the famous face in each pair. The results suggest that famous faces can be identified without awareness, and that attention is attracted by the faces of famous persons not regarded as “evil”.     

Development of selective attention in young infants: Enhancement and attenuation of startle reflex by attention

This study examined the effect of level of attention engagement on the modification of the blink reflex in young infants. Infants at 8, 14, 20, or 26 weeks of age were presented with interesting visual or auditory stimuli. At delays defined by changes in heart rate known to be associated with sustained attention or attention disengagement, blink reflexes were elicited by visual or auditory blink reflex stimuli. Blink amplitude varied according to the level of attention, and the match between the foreground and blink reflex stimulus. If the infant was attending to the foreground stimulus, a blink reflex stimulus in the same modality resulted in enhanced blink reflex magnitude. A blink reflex stimulus in the other modality resulted in an attenuated blink reflex magnitude. If attention was not engaged with the foreground stimulus, this modulation of the blink reflex did not occur. This ‘selective modality effect’ showed an increasing tendency to occur between 8 and 26 weeks of age. These results show that selective attention to modalities increases over this age range.     

Competition in early exogenous orienting between 7 and 21 weeks

Exogenous (stimulus-driven) orienting between 7 and 21 weeks of age was examined in 2 experiments using a display with multiple potential targets of attention. On each trial a small moving probe was used to draw attention to one side of the display or the other. This moving probe appeared simultaneously with 27 static bars. In the first experiment, sensitivity to the moving target was affected significantly by the spatial distribution of these red and green static bars for 14-week-olds but not for 8-week-olds. Sensitivity to the moving target was lower for 14-week-olds when most of the red bars appeared contralaterally to the moving target. This effect replicated a similar effect observed in J. L. Dannemiller (1998). The lack of a contralateral competition effect in Experiment 1 for the 8-week-olds may have occurred because I used a stronger motion stimulus for the younger infants in an attempt to hold the overall performance constant at the 2 ages. A second experiment using a weaker motion stimulus showed that this contralateral competition effect was observable over the entire age range from 7 to 21 weeks of age. Thus as early as 7 weeks of age, sensitivity for a small moving stimulus can be significantly influenced by the simultaneous presence of competing targets of attention in the visual field. Large increases in overall sensitivity were also found across the age range from 7 to 21 weeks. Results are discussed in terms of the development of putative competition mechanisms involved in exogenous orienting.     

Enhanced facial symmetry assessment in orthodontists

Assessing facial symmetry is an evolutionarily important process, which suggests that individual differences in this ability should exist. As existing data are inconclusive, the current study explored whether a group trained in facial symmetry assessment, orthodontists, possessed enhanced abilities. Symmetry assessment was measured using face and nonface stimuli among orthodontic residents and two control groups: university participants with no symmetry training and airport security luggage screeners, a group previously shown to possess expert visual search skills unrelated to facial symmetry. Orthodontic residents were more accurate at assessing symmetry in both upright and inverted faces compared to both control groups, but not for nonface stimuli. These differences are not likely due to motivational biases or a speed–accuracy tradeoff—orthodontic residents were slower than the university participants but not the security screeners. Understanding such individual differences in facial symmetry assessment may inform the perception of facial attractiveness.     

Expertise training with novel objects leads to left-lateralized facelike electrophysiological responses

Scalp event-related potentials (ERPs) in humans indicate that face and object processing differ approximately 170 ms following stimulus presentation, at the point of the N170 occipitotemporal component. The N170 is delayed and enhanced to inverted faces but not to inverted objects. We tested whether this inversion effect reflects early mechanisms exclusive to faces or whether it generalizes to other stimuli as a function of visual expertise. ERPs to upright and inverted faces and novel objects (Greebles) were recorded in 10 participants before and after 2 weeks of expertise training with Greebles. The N170 component was observed for both faces and Greebles. The results are consistent with previous reports in that the N170 was delayed and enhanced for inverted faces at recording sites in both hemispheres. For Greebles, the same effect of inversion was observed only for experts, primarily in the left hemisphere. These results suggest that the mechanisms underlying the electrophysiological face-inversion effect extend to visually homogeneous nonface object categories, at least in the left hemisphere, but only when such mechanisms are recruited by expertise.     

The control of stimulus-driven saccades is subject not to central, but to visual attention limitations

In three experiments, we investigated whether the control of reflexive saccades is subject to central attention limitations. In a dual-task procedure, Task 1 required either unspeeded reporting or ignoring of briefly presented masked stimuli, whereas Task 2 required a speeded saccade toward a visual target. The stimulus onset asynchrony (SOA) between the two tasks was varied. In Experiments 1 and 2, the Task 1 stimulus was one or three letters, and we asked how saccade target selection is influenced by the number of items. We found (1) longer saccade latencies at short than at long SOAs in the report condition, (2) a substantially larger latency increase for three letters than for one letter, and (3) a latency difference between SOAs in the ignore condition. Broadly, these results match the central interference theory. However, in Experiment 3, an auditory stimulus was used as the Task 1 stimulus, to test whether the interference effects in Experiments 1 and 2 were due to visual instead of central interference. Although there was a small saccade latency increase from short to long SOAs, this difference did not increase from the ignore to the report condition. To explain visual interference effects between letter encoding and stimulus-driven saccade control, we propose an extended theory of visual attention.     

A bottleneck in face identification: repetition priming from flanker images

There is evidence that face processing is capacity-limited in distractor interference tasks and in tasks requiring overt recognition memory. We examined whether capacity limits for faces can be observed with a more sensitive measure of visual processing, by measuring repetition priming of flanker faces that were presented alongside a face or a nonface target. In Experiment 1, we found identity priming for face flankers, by measuring repetition priming across a change in image, during task-relevant nonface processing, but not during the processing of a concurrently-presented face target. Experiment 2 showed perceptual priming of the flanker faces, across identical images at prime and test, when they were presented alongside a face target. In a third Experiment, all of these effects were replicated by measuring identity priming and perceptual priming within the same task. Overall, these results imply that face processing is capacity limited, such that only a single face can be identified at one time. Merely attending to a target face appears sufficient to trigger these capacity limits, thereby extinguishing identification of a second face in the display, although our results demonstrate that the additional face remains at least subject to superficial image processing.     

The development of sensitivity to visual structure.

Perceived complexity of visual patterns as a function of stimulus structure and contour was studied in 4–5-, 7–8-, and 9–10-yr-old children and adults. Subjects participated in a paired comparison task of visual complexity. Both amount of contour and the presence or absence of structure in the patterns were manipulated The results indicated that complexity judgments of all subjects were affected by the presence of structure at lower levels of contour. With age, gradually increasing weight was attributed to visual structure in high-contour patterns. The results suggested that sensitivity to visual structure may develop well into school age, taking the form of a gradual increase in the number of pattern elements which can be perceived to be organized.     

Figural goodness effects in perception and memory

There is an apparent contradiction concerning configurational effects in visual information processing. Some studies have shown that when an array is organized into a “good” or unitary Gestalt, analysis of a single part of it is facilitated, while others have shown “good” arrays to impede search for a part. The three experiments reported here support the proposition that goodness of form can facilitate performance when memory is used, but that goodness impairs strictly perceptual search for a part of an array. These experiments compare detection of a single feature in faces (unitary figures) and nonfaces. They show that when the face or nonface is presented before the target feature (and must be held in memory), performance is better for faces than for nonfaces. When the target is presented before the face or nonface and perceptual search is required, faces give worse performance than nonfaces. Implications for perceptual phenomena, including the object-superiority and word-superiority effects, are discussed.     

Attention to eye contact opportunity and facial motion by three-month-old infants

Thirty-two 3-month-old infants participated in two experiments showing color videotapes of facial stimuli in a paired comparison format. In Experiment 1, the experimenter, serving as the stimulus, looked either directly at the infant or averted his gaze to the side; the face was presented either still or in motion. Eye contact opportunity had no effect while motion of the head was an effective attractor of visual fixation. In Experiment 2, the amount of available eye contact opportunity was parametrically varied by occluding the eyes with different patterns of blinking, each at the same rate. The no-motion 100% eye contact available condition received less attention than the three blinking stimuli, which were all equally attended to, though they varied with respect to the amount of eye contact opportunity they afforded. The contrast in effect of eye contact availability and rather subtle stimulus motion would imply that 3-month-old infants are comparatively insensitive to being the object of another's visual regard.     

A visual short-term memory advantage for faces

What determines how much can be stored in visual short-term memory (VSTM)? Studies of VSTM have focused largely on stimulus-based properties such as the number or complexity of the items stored. Recent work also suggests that capacity is severely reduced for items within the same category. However, the importance for VSTM capacity of more qualitative differences in processing for different categories has not been investigated. For example, faces are processed more holistically than other objects. In Experiments 1 and 2, we show that the processing of faces, objects that are crucial socially and for which we possess considerable expertise, overcomes these limitations. More faces can be stored in VSTM than objects from other complex nonface categories. As in prior studies, at short encoding durations we found that capacity for faces was less than that for other categories. However, at longer encoding durations, capacity for faces exceeded that for nonface objects, and this advantage was specific to upright faces. Because inversion reduces holistic processing, the interaction of orientation with VSTM capacity—which occurred for faces but not objects in Experiment 3—suggests that it is holistic processing that confers an advantage for face VSTM when sufficient encoding time is allowed.     

Discrimination of faces and houses by rhesus monkeys: the role of stimulus expertise and rotation angle

The face inversion effect, or impaired recognition of upside down compared to upright faces, is used as a marker for the configural processing of faces in primates. The inversion effect in humans and chimpanzees is strongest for categories of stimuli for which subjects have considerable expertise, primarily conspecifics’ faces. Moreover, discrimination performance decreases linearly as faces are incrementally rotated from upright to inverted. This suggests that rotated faces must be transformed, or normalized back into their most typical viewpoint before configural processing can ensue, and the greater the required normalization, the greater the likelihood of errors resulting. Previous studies in our lab have demonstrated a general face inversion effect in rhesus monkeys that was not influenced by expertise. Therefore, the present study examined the influence of rotation angle on the visual perception of face and nonface stimuli that varied in their level of expertise to further delineate the processes underlying the inversion effect in rhesus monkeys. Five subjects discriminated images in five orientation angles. Results showed significant linear impairments for all stimulus categories, including houses. However, compared to the upright images, only rhesus faces resulted in worse performance at rotation angles greater than 45°, suggesting stronger configural processing for stimuli for which subjects had the greatest expertise.     

Context superiority in a detection task with line-element stimuli: a low-level effect

Detection and identification performances for vertical and horizontal target elements embedded within an array of oriented noise elements were measured as a function of the orientation difference between the target and noise elements. Detection performances obtained with one vertical and three horizontal target elements clustered together and displayed such that they formed a schematic face-like pattern were significantly better than those obtained with the same clustered target elements displayed in an arbitrary, symmetrical or asymmetrical, configuration. This was so even though the identification of the face and nonface stimuli was well below the detection threshold of their parts. Detection thresholds for the clustered nonface patterns were slightly but significantly lower than those for the same target elements dispersed among the noise elements. Probability summation calculations based on the detection results obtained with one single target element predict detection thresholds which are intermediate between those of the clustered and dispersed targets, suggesting that inhibitory and facilitatory spatial interactions respectively are at work for the two types of stimuli. The existence of a context(face)-superiority effect at the detection level indicates top-down/bottom-up interactions between remote visual processing stages.