Isolating the Special Component of Face Recognition: Peripheral Identification and a Mooney Face.

A previous finding argues that, for faces, configural (holistic) processing can operate even in the complete absence of part-based contributions to recognition. Here, this result is confirmed using 2 methods. In both, recognition of inverted faces (parts only) was removed altogether (chance identification of faces in the periphery; no perception of a particularly hard-to-see Mooney face). Recognition of upright faces (configural plus parts), however, remained good. The simplicity of these new "isolation" techniques makes them ideal for (a) assessing configural processing in specialist populations (e.g., children, object experts) and (b) exploring properties of configural processing for faces in detail. As an example of the latter, orientation tuning was tested. Results argue against models in which faces are rotated to upright prior to identification. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A decisional component of holistic encoding.

It has been proposed (see, specifically, M. J. Farah, K. D. Wilson, M. Drain, & J. N. Tanaka, 1998) that human faces are used in cognition as undifferentiated wholes. General recognition theory (GRT; F. G. Ashby & J. T. Townsend, 1986) is used to represent hypotheses regarding the possible sources for the behavioral evidence supporting holistic representation. Specifically, it is suggested that holism can be understood in terms of violations of informational independence, informational separability, or decisional separability, as these constructs are defined in GRT. Stimuli were presented upright, inverted, and in an encoding task that emphasized the meaningful nature of the stimuli. Patterns of performance (recognition hit rates) were consistent with prior studies. However, there were only a handful of violations of informational separability. Instead, consistent violations of decisional separability suggested a decisional basis for holistic effects. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Categorical perception of face identity in noise isolates configural processing.

Neuropsychological evidence suggests that face recognition based on configural (holistic) information can occur in isolation from recognition based on local feature cues. The present study shows that configural processing can be isolated experimentally in normal subjects. A phenomenon is reported that exists only for upright whole faces, namely categorical perception (CP) of face identity in noise. Three discrimination tasks (ABX, better likeness, and similarity ratings) were used to test for perceptual distortion across the category boundary predicted from binary classification of face morphs. Noise was added such that any single local region provided unreliable cues to identity. Under these conditions, CP was found for upright faces but not for inverted faces or single features, even with more than 10,000 trials. The CP-in-noise signature phenomenon was then used to show that configural processing survives image plane rotations of 45°–90°. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Configural processing and face viewpoint.

Configural/holistic processing, a key property of face recognition, has previously been examined only for front views of faces. Here, 6 experiments tested front (0°), three-quarter (45°), and profile views (90°), using composite and peripheral inversion tasks. Results showed an overall disadvantage in identifying profiles. This arose entirely from part-based processing: View effects were as strong for disrupted-configuration faces (inverted, misaligned, scrambled) as for normal-configuration faces. In contrast, configural processing (aligned-misaligned difference, upright-inverted difference) was equally strong for all views under both clear and degraded viewing conditions. Findings argue that, although part-based processing is weakened by lower natural frequency of the profile view and/or occlusion of key face features, neither of these variables influences configural processing. This suggests that the functional role of configural processing is to allow reliable face identification despite substantial variance in local information across different natural images. Results also show that only image-plane rotation of faces (upright through inverted) affects configural processing; the contrast with depth rotation has potential implications for understanding the origin of configural processing in terms of innate versus experience-based expertise contributions. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Encoding Operations and Recognition Memory for Faces.

Two experiments examined the effects of encoding operations on forced-choice recognition memory for upright and inverted photographs of faces. In Experiment 1, with distractors closely matched to targets, performance was better on upright than on inverted faces, but was unaffected by whether subjects judged faces for distinctive features, distinctive traits or distinctive expressions. In Experiment 2, where distractors were either absent or weakly matched to distractors, accuracy was again higher on upright than on inverted faces, and was similar for the three encoding operations on upright faces. In contrast, it was poorer for distinctive expression judgments than for distinctive feature or for distinctive trait judgments on inverted faces. These results support Winograd's (1981) claim that distinctive feature and distinctive trait judgments both lead to the isolation of distinctive features. However, it was argued that distinctive expression judgments led to configural processing that was disrupted by inversion. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Orienting to Eye Gaze and Face Processing.

The author conducted 7 experiments to examine possible interactions between orienting to eye gaze and specific forms of face processing. Participants classified a letter following either an upright or inverted face with averted, uninformative eye gaze. Eye gaze orienting effects were recorded for upright and inverted faces, irrespective of whether the faces were simple, schematic faces or more realistic faces. In contrast, inversion affected orienting to targets appearing along the vertical axis. Switching the contrast between the iris and sclera reversed orienting to eye gaze. Lifting the eyelid to expose more of the iris-sclera contrast led to a potentiation of orienting to eye gaze. Raising the eyebrow alone without the eyelid did not affect orienting. The findings suggest that local perceptual information is critical for orienting to eye gaze and that the effect can occur with a degree of independence from certain types of face processing. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Rhesus monkeys (Macaca mulatta) lack expertise in face processing.

Faces are salient stimuli for primates that rely predominantly on visual cues for recognizing conspecifics and maintaining social relationships. While previous studies have shown similar face discrimination processes in chimpanzees and humans, data from monkeys are unclear. Therefore, three studies examined face processing in rhesus monkeys using the face inversion effect, a fractured face task, and an individual recognition task. Unlike chimpanzees and humans, the monkeys showed a general face inversion effect reflected by significantly better performance on upright compared to inverted faces (conspecifics, human and chimpanzees faces) regardless of the subjects' expertise with those categories. Fracturing faces alters first- and second-order configural manipulations whereas previous studies in chimpanzees showed selective deficits for second-order configural manipulations. Finally, when required to individuate conspecific's faces, i.e., matching two different photographs of the same conspecific, monkeys showed poor discrimination and repeated training. These results support evolutionary differences between rhesus monkeys and Hominoids in the importance of configural cues and their ability to individuate conspecifics' faces, suggesting a lack of face expertise in rhesus monkeys. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Visual selection contributes to artists' advantages in realistic drawing.

To test a venerable explanation for artists' drawing ability, superior skill at visual selection, artists and nonartists traced a photograph of a face using 70 pieces of tape—not enough to depict everything. Artists and nonartists judged the drawings on accuracy. A mixed-model analysis of variance yielded a reliable advantage for artist drawers, no main effect of judge group, and a strong interaction, in which artist judges' ratings distinguished artist versus nonartist drawers, but nonartist judges' ratings did not. Thus, artists appear to make superior decisions about what to include in drawings and are also more sensitive to others' decisions. In a second study, using the same task, nonartists drew upright or inverted faces. An interaction was found in which artist judges rated faces drawn inverted as more accurate than faces drawn upright, but nonartist judges' ratings showed no differences. Results are discussed in terms of reconciling top-down and bottom-up accounts of skilled drawing. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Inversion and processing of component and spatial–relational information in faces.

This study compared effects of inversion on perceptual processing of faces with distorted components (eyes and mouths) and faces distorted by altering spatial relations between components. In a rating task, participants inversion reduced the rated grotesqueness of spatially distorted faces but not that of faces with altered components. In a comparison task, pairs of faces were shown side by side; participants judged whether they were identical or different. Inversion greatly reduced the rate at which participants responded within 3 s to pairs that differed spatially, but not pairs that differed componentially. Also, latencies for detecting spatial differences were lengthened by inversion more than latencies for detecting componential differences. Results support the hypothesis that inversion impairs encoding of spatial-relational information more than, or instead of, componential information, depending on the task. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The face of aging: Sensitivity to facial feature relations changes with age.

Fundamental to face processing is the ability to encode information about the spatial relations among facial features (configural information). Using a bizarreness rating paradigm, we found older adults differed from young adults in rating configurally distorted faces (eyes, mouth inverted) as less bizarre across all tested orientations (0° to 180°), and were more vulnerable to orientation effects when faces were rotated beyond 90°. No age-related differences in perception of either unaltered faces or featurally distorted faces (eyes whitened, teeth blackened) occurred. These findings identify changes in sensitivity to configural information as an important factor in age-related differences in face perception. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Are faces of different species perceived categorically by human observers?

What are the species boundaries of face processing? Using a face-feature morphing algorithm, image series intermediate between human, monkey (macaque), and bovine faces were constructed. Forced-choice judgement of these images showed sharply bounded categories for upright face images of each species. These predicted the perceptual discrimination boundaries for upright monkey-cow and cow-human images, but not human-monkey images. Species categories were also well-judged for inverted face images, but these did not give sharpened discrimination (categorical perception) at the category boundaries. While categorical species judgements are made reliably, only the distinction between primate faces and cow faces appears to be categorically perceived, and only in upright faces. One inference is that humans may judge monkey faces in terms of human characteristics, albeit distinctive ones.     

Does face inversion change spatial frequency tuning?

The authors examined spatial frequency (SF) tuning of upright and inverted face identification using an SF variant of the Bubbles technique (F. Gosselin & P. G. Schyns, 2001). In Experiment 1, they validated the SF Bubbles technique in a plaid detection task. In Experiments 2a–c, the SFs used for identifying upright and inverted inner facial features were investigated. Although a clear inversion effect was present (mean accuracy was 24% higher and response times 455 ms shorter for upright faces), SF tunings were remarkably similar in both orientation conditions (mean r = .98; an SF band of 1.9 octaves centered at 9.8 cycles per face width for faces of about 6°). In Experiments 3a and b, the authors demonstrated that their technique is sensitive to both subtle bottom-up and top-down induced changes in SF tuning, suggesting that the null results of Experiments 2a–c are real. The most parsimonious explanation of the findings is provided by the quantitative account of the face inversion effect: The same information is used for identifying upright and inverted inner facial features, but processing has greater sensitivity with the former. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Search asymmetries with real faces: Testing the anger-superiority effect.

The anger-superiority hypothesis states that angry faces are detected more efficiently than friendly faces. Previously research used schematized stimuli, which minimizes perceptual confounds, but violates ecological validity. The authors argue that a confounding of appearance and meaning is unavoidable and even unproblematic if real faces are presented. Four experiments tested carefully controlled photos in a search-asymmetry design. Experiments 1 and 2 revealed more efficient detection of an angry face among happy faces than vice versa. Experiment 3 indicated that the advantage was due to the mouth, but not to the eyes, and Experiment 4, using upright and inverted thatcherized faces, suggests a perceptual basis. The results are in line with a sensory-bias hypothesis that facial expressions evolved to exploit extant capabilities of the visual system. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Local and relational aspects of face distinctiveness

Distinctiveness contributes strongly to the recognition and rejection of faces in memory tasks. In four experiments we examine the role played by local and relational information in the distinctiveness of upright and inverted faces. In all experiments subjects saw one of three versions of a face: original faces, which had been rated as average in distinctiveness in a previous study (Hancock, Burton, & Bruce, 1996), a more distinctive version in which local features had been changed (D-local), and a more distinctive version in which relational features had been changed (D-rel). An increase in distinctiveness was found for D-local and D-rel faces in Experiment 1 (complete faces) and 3 and 4 (face internals only) when the faces had to be rated in upright presentation, but the distinctiveness of the D-rel faces was reduced much more than that of the D-local versions when the ratings were given to the faces presented upside-down (Experiments 1 and 3). Recognition performance showed a similar pattern: presented upright, both D-local and D-rel revealed higher performance compared to the originals, but in upside-down presentation the D-local versions showed a much stronger distinctiveness advantage. When only internal features of faces were used (Experiments 3 and 4), the D-rel faces lost their advantage over the Original versions in inverted presentation. The results suggest that at least two dimensions of facial information contribute to a face's apparent distinctiveness, but that these sources of information are differentially affected by turning the face upside-down. These findings are in accordance with a face processing model in which face inversion effects occur because a specific type of information processing is disrupted, rather than because of a general disruption of performance.     

A theoretical framework for understanding the simultaneous base-rate and payoff manipulations on decision criterion learning in perceptual categorization.

Observers completed perceptual categorization tasks in which base rates and payoffs were manipulated separately or simultaneously across a range of category discriminabilities. Decision criterion estimates from the simultaneous base-rate/payoff conditions were closer to optimal than those predicted from the independence assumption, in line with predictions from the flat-maxima hypothesis. A hybrid model that instantiated the flat-maxima and competition between reward and accuracy maximization hypotheses was applied to the data as well as used in a reanalysis of C. J. Bohill and W. J. Maddox's (2001) study. The hybrid model was superior to a model that incorporated the independence assumption, suggesting that violations of the independence assumption are to be expected and are well captured by the flat-maxima hypothesis, without requiring any additional assumptions. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Face detection in peripheral vision: do faces pop out?

We examined whether faces can produce a 'pop-out' effect in visual search tasks. In the first experiment, subjects' eye movements and search latencies were measured while they viewed a display containing a target face amidst distractors. Targets were upright or inverted faces presented with seven others of the opposite polarity as an 'around-the-clock' display. Face images were either photographic or 'feature only', with the outline removed. Naive subjects were poor at locating an upright face from an array of inverted faces, but performance improved with practice. In the second experiment, we investigated systematically how training improved performance. Prior to testing, subjects were practised on locating either upright or inverted faces. All subjects benefited from training. Subjects practised on upright faces were faster and more accurate at locating upright target faces than inverted. Subjects practised on inverted faces showed no difference between upright and inverted targets. In the third experiment, faces with 'jumbled' features were used as distractors, and this resulted in the same pattern of findings. We conclude that there is no direct rapid 'pop-out' effect for faces. However, the findings demonstrate that, in peripheral vision, upright faces show a processing advantage over inverted faces.     

Learning and attention in multidimensional identification and categorization: Separating low-level perceptual processes and high-level decisional processes.

Four observers completed identification and categorization tasks. Learning and attention processes were examined by applying general recognition theory (F. G. Ashby & J. T. Townsend, 1986), which separates perceptual, decisional, and attentional processes. Learning led to decision regions that became more nearly optimal. Learning had little effect on perceptual processes in identification and decisional integration categorization tasks but affected perceptual processes in decisional selective attention categorization tasks, leading to perceptual selective attention. These findings suggest that (a) identification and categorization invoke decision strategies that are localized in the striatum (F. G. Ashby, E. M. Waldron, W. W. Lee, & A. Berkman, 2001) and (b) perceptual and decisional attention systems exist and are mediated by distinct brain structures (M. I. Posner & S. E. Petersen, 1990). (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Configural weighting in judgments of two- and four-outcome gambles.

This study tested branch independence, a key property distinguishing nonconfigural from configural theories of decision making. Sixty undergraduates judged buying and selling prices of 168 lotteries composed of 2 or 4 equally likely outcomes, (x, y, z, v). Branch independence requires that (x, y, z, v) is judged higher than (x′, y′, z, v) whenever (x, y, z′, v′) is judged higher than (x′, y′, z′, v′). Different violations observed in different viewpoints are consistent with the theory that the utility function is independent of viewpoint and that only configural weights differ between viewpoints. Lower ranked outcomes have greater weights in the buyer's than in the seller's viewpoint. Sellers place more weight than buyers on higher ranked outcomes. In both viewpoints, violations of branch independence are contrary to the inverse-S weighting function of cumulative prospect theory: Moderate outcomes receive more weight than adjacent extreme outcomes. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Holistic processing of faces: Perceptual and decisional components.

Researchers have used several composite face paradigms to assess holistic processing of faces. In the selective attention paradigm, participants decide whether one face part (e.g., top) is the same as a previously seen face part. Their judgment is affected by whether the irrelevant part of the test face is the same as or different than the relevant part of the study face. This failure of selective attention implies holistic processing. However, the authors show that this task alone cannot distinguish between perceptual and decisional sources of holism. The distinction can be addressed by the complete identification paradigm, in which both face parts are judged to be same or different, combined with analyses based on general recognition theory (F. G. Ashby & J. T. Townsend, 1986). The authors used a different paradigm, sequential responses, to relate these 2 paradigms empirically and theoretically. Sequential responses produced the same results as did selective attention and complete identification. Moreover, disruptions of holistic processing by systematic misalignment of the faces corresponded with systematic and significant changes in the decisional components, but not in the perceptual components, that were extracted using general recognition theory measures. This finding suggests a significant decisional component of holistic face processing in the composite face task. (PsycINFO Database Record (c) 2010 APA, all rights reserved)