DeGutis and colleagues (2012) have demonstrated that adults with developmental prosopagnosia show a typical holistic processing effect for the mouth, but not for the eye region. These findings are consistent with previous speculations that holistic processing of the eye region may be particularly important for successful face recognition. The present study examined 30 children, recruited based on their exceptionally high or low scores on the Cambridge Face Memory Task–Children (CFMT-C) from a database of more than 500 children. These children were separated into two groups roughly matched for age: those that performed very well at face recognition (high performers; N=15) and those that performed very poorly at face recognition (low performers; N=15). Average scores on the CFMT-C for each group were 91.31% (sd=5%) and 69.27% (sd=8%) respectively. For each group, we examined holistic face processing using the Part-Whole Task (Tanaka et al., 2010). This task examines the ability to recognize face parts (e.g. the eyes) both in isolation and in the context of the whole face, both of which differed by only one feature. As expected, the high performing group showed an overall holistic advantage, with greater accuracy on whole trials than part trials [t(14)=3.82, p<0.01]. This finding remained marginally significant when examining eye and mouth trials separately [Eye Trials: t(14)=1.97, p=0.068; Mouth Trials: t(14)=2.01, p=0.064]. Similar to the findings of DeGutis and colleagues (2012), the low performing group also showed an overall holistic advantage [t(14)=3.74, p<0.01]. However, this holistic advantage was carried by a holistic effect for mouth trials [t(14)=3.1, p<0.01] but not eye trials [t(14)= -0.29, n.s.]. These results replicate the finding that holistic processing of the eye region is particularly important for successful face recognition and may be impaired in cases of prosopagnosia. Furthermore, these data demonstrate the similarities in holistic processing between children and adults. Meeting abstract presented at VSS 2014

Corrow, Granrud, Mathison, and Yonas (2011, Perception, 40, 1376–1383) found evidence that 6-month-old infants perceive the hollow face illusion. In the present study we asked whether 6-month-old infants perceive illusory depth reversal for a nonface object and whether infants' perception of the hollow face illusion is affected by mask orientation inversion. In experiment 1 infants viewed a concave bowl, and their reaches were recorded under monocular and binocular viewing conditions. Infants reached to the bowl as if it were convex significantly more often in the monocular than in the binocular viewing condition. These results suggest that infants perceive illusory depth reversal with a nonface stimulus and that the infant visual system has a bias to perceive objects as convex. Infants in experiment 2 viewed a concave face-like mask in upright and inverted orientations. Infants reached to the display as if it were convex more in the monocular than in the binocular condition; however, mask orientation had no effect on reaching. Previous findings that adults' perception of the hollow face illusion is affected by mask orientation inversion have been interpreted as evidence of stored-knowledge influences on perception. However, we found no evidence of such influences in infants, suggesting that their perception of this illusion may not be affected by stored knowledge, and that perceived depth reversal is not face-specific in infants.

It is clear that adults can use pictorial depth cues to detect that a simple drawing specifies an impossible object. Recent studies have suggested that 4-month-old infants have a similar ability. Infants looked at a drawing of an impossible object for a longer period of time than a drawing of a possible object (Shuwairi and Johnson 2012). The findings suggest that sensitivity to depth order specified by the location of T-junctures as well as the ability to integrate local depth information into a global representation is present early in development. In addiction, work by Young and Deregowski (1981) has found that 7-year-old children were less able to discriminate possible objects compared to older children. The goal of this work is to use an eye tracking apparatus to investigate developmental changes in the sequence of fixations that take place when impossible objects are viewed. As part of this project fixation durations were recorded using a Tobii eye-tracker on 8-year-old-children. We found that 10 out of the 12 children (p<0.05 by sign test) fixated the relevant part of the impossible figure longer than the corresponding possible region. Future work will investigate developmental changes in the ability children to detect the impossible of a variety of drawings and the sequence of eye fixations used in the task. Meeting abstract presented at VSS 2013

Intro: At previous VSS meetings, we reported that: When 6-month-old infants were presented with the concave face illusion display, lit from below, they reach more often to the apparently closest center of mask when viewing it monocularly and to the actually closer edges of the mask when viewing it binocularly. This indicates that they perceive the concave mask as convex in the monocular viewing condition. In a second study the mask was inverted and lit from the side to reduce its face-like appearance. Infants reached more often to the center of mask in the monocular condition to the same degree as in the first study. This suggests that reducing the face-like appearance of the display may not influence the infant’s perception of the mask as strongly as it does with adults, who perceive a stronger illusion when the face is presented upright than when it is inverted (Papathomas & Bono, 2004). The current study aimed to investigate whether infants’ use of a general convexity assumption by presenting the infants with a concave object that is unfamiliar to the infant. Method: Using a within-subjects design, we presented 6-month-old infants (n=22) with an unfamiliar concave object and observed the trajectory of their reaches in the binocular and monocular viewing conditions. Results: Infants reached more often to the center of the concave object in the monocular condition (61%) than in the binocular condition (32%). A paired sample t-test revealed a significant difference between the means (t=-3.05, P <.01), with a large effect size, Cohen’s d= .9159. Discussion: These results suggest that the infants employ a general convexity assumption and perceive the concave object as convex when binocular information for depth is not available. When binocular cues specify the actual 3-D layout of the display they perceive it as convex. Meeting abstract presented at VSS 2013

Prosopagnosia is a socially debilitating disorder; especially for children faced with the task of developing social skills (Dalrymple, Corrow, Duchaine, & Yonas, 2012). Studies examining face recognition in adults have supported the hypothesis that holistic face processing underlies effective face recognition abilities. However, only recently has this hypothesis been directly examined in adults. A recent report by DeGutis, Wilmer, Mercado, & Cohen (2012) found that the relationship between measures of holistic processing and face recognition scores is robust. Understanding the role of holistic processing in the face recognition skills of children may aid in the creation of successful interventions for children with prosopagnosia. The present study examined individual differences in face recognition skills of typically developing children and tested two hypotheses; 1. like adults, children with higher scores of holistic face processing would exhibit better face recognition skills and 2. greater attention to the eye region would benefit face recognition whereas attention to extraneous areas, such as the chin, would inhibit successful face recognition. 75 8-year-old children completed childhood versions of the Cambridge Face Memory Task (Pellicano et al., unpublished; Dalrymple et al., 2012) and child-adapted versions of the part-whole task (Tanaka et al., 1993). Regression-based scores of holistic processing significantly predicted face recognition performance (p=0.007) with higher scores on the part-whole task predicting better CFMT scores. Furthermore, although attention to the eyes did not significantly predict face recognition skills, attention to the chin (p=0.04), forehead (p=0.01), and neck (p=0.001) negatively correlated with face recognition scores. These results suggest that individual differences in holistic processing play a significant role in face recognition for typically developing children. Furthermore, attention to extraneous areas of the face may interfere with this process. Meeting abstract presented at VSS 2013

Last year at VSS, we reported that 6 month-old-infants perceive the hollow-face illusion. Infants reached preferentially to the center of a concave mask during monocular viewing and to the edges of the mask during binocular viewing, indicating that they perceived the mask as convex when viewing it monocularly. However, it is unclear whether infants perceive this illusion as a result of an assumption that faces are convex or a more general convexity assumption. The current study investigated whether the infant visual system makes a face-specific convexity assumption in perceiving the hollow-face illusion. For adults, the illusion is less effective when the mask is inverted than when upright, suggesting that a face-specific convexity assumption plays a role in perceiving that a concave mask is convex. Method: Infants were presented with an upright and an inverted concave mask. Half of the infants viewed the mask monocularly (n=24) and the other half viewed the mask binocularly (n=28). Infants were allowed to reach for the mask and we observed the trajectory of their reaches. Results: Infants in both the upright and inverted conditions reached more often to the center of the display in the monocular condition (61% upright; 65% inverted) than in the binocular condition (15% upright; 16% inverted). Monocular versus binocular comparisons were significant for both the upright (p<0.001) and inverted (p<0.001) conditions. There was no effect of inversion. Conclusion: The results replicate the previous finding that 6-month-old infants perceive the hollow-face illusion. However, unlike adults, we found no evidence of an inversion effect, suggesting that their perception of the hollow-face illusion results from a general assumption that objects tend to be convex, not from an assumption that faces in particular are convex. Future work will examine this question by evaluating the strength of a "hollow potato" illusion Meeting abstract presented at VSS 2012

Two experiments investigated infants' and adults' perception of 3D shape from line junction information. Participants in both experiments viewed a concave wire half-cube frame. In Experiment 1, adults reported that the concave wire frame appeared to be convex when it was viewed monocularly (with one eye covered) and that it appeared to be concave when it was viewed binocularly. In Experiment 2, 5- and 7-month-old infants were shown the concave wire frame under monocular and binocular viewing conditions, and their reaching behavior was recorded. The infants in both age groups reached preferentially toward the center of the wire frame in the monocular condition and toward its edges in the binocular condition. Because infants typically reach to what they perceive to be closest to them, these reaching preferences provide evidence that they perceived the wire frame as convex when they viewed it monocularly and as concave when they viewed it binocularly. These findings suggest that, by 5 months of age, infants, like adults, use line junction information to perceive depth and object shape.

The visual system must employ assumptions, or constraints, to interpret pictorial (static monocular) displays. For example, for the depth cue of relative size to be effective, adults use an assumption that two objects are similar in size in perceiving the retinally larger one as closer. Our goal is to investigate if and when infants, like adults, use similar assumptions to perceive 3-D layout. Last year at VSS, we reported that six-month-old infants use line junctions to perceive a concave shape as convex when only static monocular information is available. In the present study, we investigated whether infants, like adults, use the assumption that faces are convex. Specifically, we used Richard Gregory's concave face illusion to examine if infants use prior knowledge about the 3-D layout of faces to direct their reaching to the apparently closest part of the display (the nose). Methods: Using a within-subjects design, we presented six-month-old infants (n = 11) with Gregory's concave face illusion and observed the trajectory of infants' reaching behavior under monocular and binocular viewing conditions. Reaching behavior was scored by a blind observer. Results: Infants reached more often to the center of the display in the monocular condition (67%) than in the binocular condition (12%, p = 0.001). These results indicate that infants perceived the display as convex under monocular viewing conditions and as concave when viewing the display with two eyes. In addition, in the monocular condition, some infants attempted to grasp the nose of the face. Discussion: These findings suggest that experience with the concave nature of faces may influence how infants respond to this illusion. This provides further evidence that infants use assumptions to perceive the 3-D structure of the environment.

Developmental prosopagnosia refers to an inability to recognize faces despite typical intellectual functioning, emotion and object recognition, and no evidence of brain injury. Very little is known about this disorder in children. We present a case study of a child, henceforth referred to as B, who reported extreme difficulty in recognizing faces. Participant: B is a healthy 7 year old male with no history of neurological or behavioral disorder. His parents report that he is above-average in intelligence and very social. He has no history of visual impairment other than a deficit in face recognition. Methods: We evaluated B's face recognition skills using a variety of measures for which we have been collecting normative data. We examined his ability to recognize faces, discriminate between faces, identify emotion, and recognize objects. A clinical neuropsychologist also evaluated B using a wide variety of measures. Finally, we employed a training program to improve B's ability to recognize the faces of familiar people in his life. Results: On our test of face recognition, B answered correctly on 46.8% of trials (50% is chance); average performance of 8 year olds on this test is 80% (sd = 10.7, n = 43). These results provide evidence of a severe deficit in face recognition. B showed no evidence of a similar deficit in the ability to discriminate faces, recognize objects, and identify emotion. The neuropsychological evaluation also found a deficit in face recognition despite intact object recognition, memory, emotion recognition, and above-average IQ. Finally, we successfully trained him to identify photographs of 30 familiar people; however, he used local characteristics of the photograph to recognize the faces. Discussion: This is one of very few studies of children with developmental prosopagnosia and the first comparison of a DP child using measures collected from a large sample of typical children.

Perceptual constancies allow us to perceive properties of the distal environment despite variations in proximal stimulation. For example, adults are able to perceive the reflectance of an object despite changes in illumination; i.e., they are capable of lightness constancy. We investigated whether 4-month-old infants (n = 17) show evidence of lightness constancy. Infants viewed two faces through an aperture in a lighted enclosure. To measure a natural preference for a dark or light face, infants were presented with two pre-test trials under either low (group 1) or high (group 2) illumination. Infants were then habituated to either two white faces under low illumination (group 2) or two dark grey faces under high illumination (group 1). The luminance values for the faces during habituation were matched for the two groups. In test trials, infants were again shown the white face and the gray face under the same illumination levels presented in the pretest trials. Therefore, after habituation, one of the two faces was novel in reflectance while the other was novel in luminance. For each infant, a novelty preference was calculated by dividing the amount of time spent looking at the face that was novel in reflectance by the total looking time. Following habituation, infants showed a 58% preference for the face with the novel reflectance (p = 0.015). We also calculated the change in looking preference between the pre-test trials and post-test trials. There was a 12% greater preference for the novel reflectance in the post-test trials than the pre-test trials (p = 0.007). These results indicate that 4-month-old infants recognize an object's reflectance across changes in luminance. This indicates a degree of lightness constancy by 4 months of age. Our findings are consistent with those of Chien, Bronson-Castain, Palmer, & Teller (2006), the only previous study on lightness constancy in infants.

In this study we investigated infants' perception of the hollow-face illusion. 6-month-old infants were shown a concave mask under monocular and binocular viewing conditions and the direction of their reaches toward the mask was recorded. Adults typically perceive a concave mask as convex under monocular conditions but as concave under binocular conditions, depending on viewing distance. Based on previous findings that infants reach preferentially toward the parts of a display that are closest to them, we expected that, if infants perceive the hollow-face illusion as adults do, they would reach to the center of the mask when viewing it monocularly and to the edges when viewing it binocularly. The results were consistent with these predictions. Our findings indicated that the infants perceived the mask as convex when viewing it with one eye and concave when viewing it with two eyes. The results show that 6-month-old infants respond to the hollow-face illusion. Our finding suggests that, early in life, the visual system uses the constraint, or assumption, that faces are convex when interpreting visual input.