A new "fat face" illusion

We report a novel fat face illusion that when two identical images of the same face are aligned vertically, the face at the bottom appears 'fatter'. This illusion emerged when the faces were shown upright, but not inverted, with the size of the illusion being 4%. When the faces were presented upside down, the illusion did not emerge. Also, when upright clocks were shown in the same vertically aligned fashion, we did not observe the illusion, indicating that the fat illusion does not generalize to every category of canonically upright objects with similar geometric shape as a face.     

The Thatcher illusion and face processing in infancy

Adults readily detect changes in face patterns brought about by the inversion of eyes and mouth when the faces are viewed upright but not when they are viewed upside down. Research suggests that this illusion (the Thatcher illusion) is caused by the interfering effects of face inversion on the processing of second-order relational information (fine spatial information such as the distance between the eyes). In the current study, 6-month-olds discriminated 'thatcherized' faces when they were viewed upright but not when they were viewed upside down. These results are consistent with the notion that 6-month-olds are sensitive to second-order relational information while processing faces.     

Memory for facial expressions: The power of a smile

Faces with expressions (happy, surprise, anger, fear) were presented at study. Memory for facial expressions was tested by presenting the same faces with neutral expressions and asking participants to determine the expression that had been displayed at study. In three experiments, happy expressions were remembered better than other expressions. The advantage of a happy face was observed even when faces were inverted (upside down) and even when the salient perceptual feature (broad grin) was controlled across conditions. These findings are couched in terms of source monitoring, in which memory for facial expressions reflects encoding of the dispositional context of a prior event.     

The face inversion effect is not a consequence of aberrant eye movements

The face inversion effect is the finding that inverted faces are more difficult to recognize than other inverted objects. The present study explored the possibility that eye movements have a role in producing the face inversion effect. In Experiment 1, we demonstrated that the faces used here produce a robust face inversion effect when compared with another homogenous set of objects (antique radios). In Experiment 2, participants' eye movements were monitored while they learned a set of faces and during a recognition test. Although we clearly found a face inversion effect, the same features of a face were fixated during the learning and recognition test faces, whether the face was right side up or upside down. Thus, the face inversion effect is not a result of a different pattern of eye movements during the viewing of the face.     

Configural features in the context of upright and inverted faces

When faces are turned upside down, recognition is known to be severely disrupted. This effect is thought to be due to disruption of configural processing. Recently, Leder and Bruce (2000, Quarterly Journal of Experimental Psychology A 53 513-536) argued that configural information in face processing consists at least partly of locally processed relations between facial elements. In three experiments we investigated whether a local relational feature (the interocular distance) is processed differently in upside-down versus upright faces. In experiment 1 participants decided in which of two sequentially presented photographic faces the interocular distance was larger. The decision was more difficult in upside-down presentation. Three different conditions were used in experiment 2 to investigate whether this deficit depends upon parts of the face beyond the eyes themselves; displays showed the eye region alone, the eyes and nose, or the eyes and nose and mouth. The availability of additional features did not interact with the inversion effect which was observed strongly even when the eyes were shown in isolation. In experiment 3 all eyes were turned upside down in the inverted face condition as in the Thatcher illusion (Thompson, 1980 Perception 9 483-484). In this case no inversion effect was found. These results are in accordance with an explanation of the face-inversion effect in which the disruption of configural facial information plays the critical role in memory for faces, and in which configural information corresponds to spatial information that is processed in a way which is sensitive to local properties of the facial features involved.     

On facial expertise: processing strategies of twins' parents

Parents of monozygotic twins typically learn to recognise their own children and also to tell them apart on photographs. However, it is unknown whether such exemplar expertise can be generalised to unfamiliar faces of equal similarity (ie other twins). In the present study, parents of monozygotic twins were compared to 'control' parents whose children were non-twin siblings. In a same/different task with familiar and unfamiliar twin faces as stimuli, twins' parents were faster than control parents in distinguishing their own children, but slower when distinguishing unfamiliar twin faces. These unexpected findings can be interpreted as a generalisation of a habitual perceptual strategy that is efficient only with familiar face exemplars. When pictures of the twins' faces were inverted by 180degrees, the typical inversion costs for familiar faces were observed, but the twins' parents recognised unfamiliar twins' faces better when these were inverted than when upright. This inverted inversion effect could be caused by the high similarity of the face pairs as well as by experience with familiar faces of equally high similarity. We conclude that the facial expertise of twins' parents is idiosyncratic to their own children and not based on an enhanced general facial expertise.     

How dogs scan familiar and inverted faces: an eye movement study

Faces play an important role in communication and identity recognition in social animals. Domestic dogs often respond to human facial cues, but their face processing is weakly understood. In this study, facial inversion effect (deficits in face processing when the image is turned upside down) and responses to personal familiarity were tested using eye movement tracking. A total of 23 pet dogs and eight kennel dogs were compared to establish the effects of life experiences on their scanning behavior. All dogs preferred conspecific faces and showed great interest in the eye area, suggesting that they perceived images representing faces. Dogs fixated at the upright faces as long as the inverted faces, but the eye area of upright faces gathered longer total duration and greater relative fixation duration than the eye area of inverted stimuli, regardless of the species (dog or human) shown in the image. Personally, familiar faces and eyes attracted more fixations than the strange ones, suggesting that dogs are likely to recognize conspecific and human faces in photographs. The results imply that face scanning in dogs is guided not only by the physical properties of images, but also by semantic factors. In conclusion, in a free-viewing task, dogs seem to target their fixations at naturally salient and familiar items. Facial images were generally more attractive for pet dogs than kennel dogs, but living environment did not affect conspecific preference or inversion and familiarity responses, suggesting that the basic mechanisms of face processing in dogs could be hardwired or might develop under limited exposure.     

Removing eyebrows impairs recognition of famous faces, or doesn't, depending on how the eyebrows are removed

Matching photos of famous people's faces with their names was as fast with photos that had eyebrows masked by adhesive plaster as with unretouched original photos. But matching was slower with photos that had eyebrows erased and replaced with adjacent skin texture and colour than with original photos. The conjecture that erasure impairs recognition because it alters a face's configuration was examined by repeating the experiment with the photos shown upside down, the rationale being that because configural information is difficult to encode from inverted faces the erasure effect should diminish. With inverted faces, matching was no different for original, eyebrows-masked, and eyebrows-erased photos. Eyebrows appear to be less important for face recognition as informative parts and features than as sources of information about a face's configuration.     

The Thatcher illusion: rotating the viewer instead of the picture

Faces are difficult to recognise when presented upside down. This effect of face inversion was effectively demonstrated with the 'Thatcher illusion' by Thompson (1980 Perception 9 483-484). It has been tacitly assumed that this effect is due to inversion relative to retinal coordinates. Here we tested whether it is due to egocentric (i.e. retinal) inversion or whether the orientation of the body with respect to gravity also influences the face-inversion effect. A 3-D human turntable was used to test subjects in 5 different body-tilt (roll) orientations: 0 degree, 45 degrees, 90 degrees, 135 degrees, and 180 degrees. The stimuli consisted of 4 'normal' and 4 'thatcherised' faces and were presented in 8 different orientations in the picture plane. The subjects had to decide in a yes-no task whether the faces were 'normal' or 'thatcherised'. Analysis of the d' values revealed a significant effect of stimulus orientation and body tilt. The significant effect of body tilt was due to a drop in d' values in the 135 degrees orientation. This result is compared to findings of studies on the subjective visual vertical, where larger errors occurred in body-tilt orientations between 90 degrees and 180 degrees. The present findings suggest that the face-inversion effect relies mainly on retinal coordinates, but that in head-down body-tilt orientations around 135 degrees the gravitational reference frame has a major influence on the perception of faces.     

Neural evidence for the contribution of holistic processing but not attention allocation to the other-race effect on face memory

Multiple mechanisms have been suggested to contribute to the other-race effect on face memory, the phenomenon of better memory performance for own-race than other-race faces. Here, two of these mechanisms, increased attention allocation and greater holistic processing during memory encoding for own-race than other-race faces, were tested in two separate experiments. In these experiments event-related potentials were measured during study (the difference due to memory, Dm) and test phase (old/new effects) to examine brain activation related to memory encoding and retrieval, allowing for selective investigations of these memory sub-processes. In Experiment 1, participants studied own-race (Caucasian) and other-race (Chinese) faces under focused or divided attention. In Experiment 2, participants studied own-race (Caucasian) and other-race (African American) faces presented upright or upside down (i.e., inverted). Both experiments showed decreases in memory performance when attention allocation or holistic processing was reduced, but these effects were similar for own-race and other-race faces. Manipulations of holistic processing, but not attention allocation, influenced the neural other-race effects during memory encoding. Inverted own-race faces showed similar neural patterns as upright other-race faces, indicating that when holistic processing of own-race faces was reduced, these faces were encoded similarly as upright other-race faces. No influences of the experimental manipulations on other-race effects during memory retrieval were found. The present study provides the first neural evidence that increased holistic processing during memory encoding contributes to the other-race effect on face memory.     

The face-inversion effect as a deficit in the encoding of configural information: direct evidence

We report four experiments leading to conclusions that: (i) the face-inversion effect is mainly due to the deficits in processing of configural information from inverted faces; and (ii) this effect occurs primarily at the encoding stage of face processing, rather than at the storage stage. In experiment 1, participants discriminated upright faces differing primarily in configuration with 81% accuracy. Participants viewing the same faces presented upside down scored only 55%. In experiment 2, the corresponding discrimination rates for faces differing mainly in featural information were 91% (upright) and 90% (inverted). In experiments 3 and 4, the same faces were used in a memory paradigm. In experiment 3, a delayed matching-to-sample task was used, in which upright-face pairs differed either in configuration or features. Recognition rates were comparable to those for the corresponding upright faces in the discrimination tasks in experiments 1 and 2. However, there was no effect of delay (1 s, 5 s, or 10 s). In experiment 4, we repeated experiment 3, this time with inverted faces. Results were comparable to those of inverted conditions in experiments 1 and 2, and again there was no effect of delay. Together these results suggest that an 'encoding bottleneck' for configural information may be responsible for the face-inversion effect in particular, and memory for faces in general.     

Identification reaction time as a function of target and field object orientation

Four experiments were conducted investigating the effects of (a) target versus field object orientation, (b) target upside up versus target upside down and (c) inverting versus noninverting lenses on object identification reaction time. Inverting versus noninverting lenses had no significant effect. Target versus field orientation was significant when objects commonly found on a desk were used. Target upside up versus upside down was not significant. With letters as target and field objects, orientation was not significant. When the target letters were simply inverted (rotated 180 degrees on the horizontal axis), however, forming nonfamiliar patterns, the target upside up versus upside down was significant, with the upside down condition resulting in a longer mean reaction time. The complexity and similarity of target and field objects is offered as an explanation of the results.     

Prior knowledge about display inversion in biological motion perception

Display inversion severely impedes veridical perception of point-light biological motion (Pavlova and Sokolov, 2000 Perception & Psychophysics 62 889-899; Sumi, 1984 Perception 13 283-286). Here, by using a spontaneous-recognition paradigm, we ask whether prior information about display orientation improves biological motion perception. Participants were shown a set of 180 degrees inverted point-light stimuli depicting a human walker and quadrupeds (dogs). In experiment 1, one group of observers was not aware of the orientation of stimuli, whereas the other group was told beforehand that stimuli will be presented upside down. In experiment 2, independent groups of participants informed about stimulus orientation saw the same set of stimuli, in each of which either a moving or a static background line was inserted. The findings indicate that information about display inversion is insufficient for reliable recognition of inverted point-light biological motion. Instead, prior information facilitates display recognition only when it is complemented by additional contextual elements. It appears that visual impressions from inverted point-light stimuli remain impenetrable with respect to one's knowledge about display orientation. The origins of orientation specificity in biological motion perception are discussed in relation to the recent neuroimaging data obtained with point-light stimuli and fragmented Mooney faces.     

Upright versus upside-down faces: How interface attractiveness varies with orientation

A choice experiment is reported in which all pairs and triples of faces from a set of eight moderately attractive faces were presented, both upright and upside down, to 103 subjects. In each orientation, the subjects had to select the face that appeared more (pairs) or most (triples) attractive to them. For each orientation, the preference probabilities that arose from the pair and triple comparisons could be described by the BTL rule (Luce, 1959). Thus, each face was represented by two scores, one reflecting its attractiveness in the upright orientation and the other reflecting its attractiveness in the inverted orientation. Orientation affected the preference probabilities. Qualitatively, score ratios between faces decreased from upright to inverted orientation, suggesting that the faces became less discriminable by inversion. Quantitatively, the effect of inversion could be described by a simple rule that assumes a face’s two attractiveness scores to be affinely related across orientations. This result indicates that inversion affected all faces about equally. The present findings are discussed with respect to faces’ first- and second-order relational properties, a distinction emphasized in current theories of face perception. They suggest that the processing of first- and secondorder relational properties is impaired by inversion to roughly the same degree.     

Impaired face recognition does not preclude intact whole face perception

We studied intact and impaired processes in a prosopagnosic patient (RP). In Experiment 1, RP showed an inversion superiority effect with both faces and objects, with better performance when stimuli were presented upside down than in normal upright orientation. In Experiment 2, we studied the effect of face configuration directly by comparing matching performance with normal vs. scrambled faces. RP was worse with normal than with scrambled faces, whereas normal controls showed an advantage of a good face context. In Experiment 3, RP showed interference from external face features on the evaluation of internal face features. These results indicate, first, that although RP is impaired in face recognition and face matching, he does still encode the whole face rather than relying completely on parts-based procedures. Second, RP has a deficit at the level of the configural processes involved in finding subtle differences between individual faces, as his performance is worse when presented with a normal face configuration than with scrambled or inverted faces.     

The role of movement in the recognition of famous faces

The effects of movement on the recognition of famous faces shown in difficult conditions were investigated. Images were presented as negatives, upside down (inverted), and thresholded. Results indicate that, under all these conditions, moving faces were recognized significantly better than static ones. One possible explanation of this effect could be that a moving sequence contains more static information about the different views and expressions of the face than does a single static image. However, even when the amount of static information was equated (Experiments 3 and 4), there was still an advantage for moving sequences that contained their original dynamic properties. The results suggest that the dynamics of the motion provide additional information, helping to access an established familiar face representation. Both the theoretical and the practical implications for these findings are discussed.     

Effects of geometric distortions,Gaussian blur,and contrast negation on recognition of familiar faces

It is assumed that “configural processing” is important for recognizing people whom we know. Studies have shown better discrimination of relational changes between facial features when faces are presented upright than upside down and when presented as photographic negatives. However, studies have also demonstrated tolerance for complex changes to facial configuration of familiar faces when images are geometrically distorted. In two experiments, we explored effects of linear stretching, Gaussian blur, and contrast negation on recognition of familiar or unfamiliar faces (Experiment 1) and familiar or unfamiliar company brand logos (Experiment 2). Only contrast negation severely impaired recognition of familiar faces but not unfamiliar faces or company brand logos. Effects of contrast negation were additional to linear stretching, whereas linear stretching independently had no impact on recognition. These findings highlight an important deficit in the recognition of familiar faces caused by contrast negation.     

The other-race effect does not rely on memory: Evidence from a matching task

Viewers are typically better at remembering faces from their own race than from other races; however, it is not yet established whether this effect is due to memorial or perceptual processes. In this study, UK and Egyptian viewers were given a simultaneous face-matching task, in which the target faces were presented upright or upside down. As with previous research using face memory tasks, participants were worse at matching other-race faces than own-race faces and showed a stronger face inversion effect for own-race faces. However, subjects' performance on own and other-race faces was highly correlated. These data provide strong evidence that difficulty in perceptual encoding of unfamiliar faces contributes substantially to the other-race effect and that accounts based entirely on memory cannot capture the full data. Implications for forensic settings are also discussed.     

A sideways look at configural encoding: two different effects of face rotation

Inversion has a disproportionate disruptive effect on the recognition of faces. This may be due to the disruption of holistic or configural encoding employed to recognise upright faces. The paradigm developed by Tanaka and Farah (1993 Quarterly Journal of Experimental Psychology, Section A 46 225-246) was used to investigate the effect of 90 degrees (or orthogonal) rotation on configural encoding. Faces learnt in the orthogonal condition were not recognised as well as upright faces, but a whole-face advantage was found in both cases. This whole-face advantage did not occur for inverted faces. It appears that 90 degrees rotation affects recognition but not specifically configural encoding. It is concluded that rotating a face can have at least two different effects on face processing depending on the range of rotation. Implications for the nature of facial dimensions and the expertise account of the inversion effect are considered.     

The other-race effect does not rely on memory: Evidence from a matching task

Viewers are typically better at remembering faces from their own race than from other races; however, it is not yet established whether this effect is due to memorial or perceptual processes. In this study, UK and Egyptian viewers were given a simultaneous face-matching task, in which the target faces were presented upright or upside down. As with previous research using face memory tasks, participants were worse at matching other-race faces than own-race faces and showed a stronger face inversion effect for own-race faces. However, subjects' performance on own and other-race faces was highly correlated. These data provide strong evidence that difficulty in perceptual encoding of unfamiliar faces contributes substantially to the other-race effect and that accounts based entirely on memory cannot capture the full data. Implications for forensic settings are also discussed.