See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/51853319
Is the Face-Perception System Human-Specific
at Birth?
Article in Developmental Psychology · December 2011
DOI: 10.1037/a0026521 · Source: PubMed
CITATIONS
READS
31
181
4 authors, including:
Elisa Di Giorgio
Olivier Pascalis
18 PUBLICATIONS 158 CITATIONS
148 PUBLICATIONS 4,435 CITATIONS
Università degli Studi di Trento
SEE PROFILE
Grenoble-CNRS
SEE PROFILE
Francesca Simion
University of Padova
89 PUBLICATIONS 3,587 CITATIONS
SEE PROFILE
All content following this page was uploaded by Olivier Pascalis on 11 December 2016.
The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Developmental Psychology
Is the Face-Perception System Human-Specific at Birth?
Elisa Di Giorgio, Irene Leo, Olivier Pascalis, and Francesca Simion
Online First Publication, December 5, 2011. doi: 10.1037/a0026521
CITATION
Di Giorgio, E., Leo, I., Pascalis, O., & Simion, F. (2011, December 5). Is the Face-Perception
System Human-Specific at Birth?. Developmental Psychology. Advance online publication.
doi: 10.1037/a0026521
Developmental Psychology
2011, Vol. ●●, No. ●, 000 – 000
© 2011 American Psychological Association
0012-1649/11/$12.00 DOI: 10.1037/a0026521
Is the Face-Perception System Human-Specific at Birth?
Elisa Di Giorgio and Irene Leo
Olivier Pascalis
Università degli Studi di Padova
Universitè Pierre Mendes
Francesca Simion
Università degli Studi di Padova
The present study investigates the human-specificity of the orienting system that allows neonates to look
preferentially at faces. Three experiments were carried out to determine whether the face-perception
system that is present at birth is broad enough to include both human and nonhuman primate faces. The
results demonstrate that the newborns did not show any spontaneous visual preference for the human face
when presented simultaneously with a monkey face that shared the same features, configuration, and
low-level perceptual properties (Experiment 1). The newborns were, however, able to discriminate
between the 2 faces belonging to the 2 different species (Experiment 2). In Experiment 3, the newborns
were found to prefer looking at an upright, compared with an inverted, monkey face, as they do for
human faces. Overall, the results demonstrate that newborns perceive monkey and human faces in a
similar way. These findings are consistent with the hypothesis that the system underlying face preference
at birth is broad enough to bias newborns’ attention toward both human and nonhuman primate faces.
Keywords: face detection, newborns, human face, monkey face, face-perception system
The human visual world is complex and full of rich visual
stimuli. Among these numerous visual inputs, the human face,
because of its biological and social relevance, is one of the most
salient. A few hours after birth, the newborn’s visual attention is
preferentially attracted to faces over other complex visual stimuli
(Fantz, 1961; Goren, Sarty, & Wu, 1975; Johnson & Morton,
1991; Macchi Cassia, Turati, & Simion, 2004; Valenza, Simion,
Macchi Cassia, & Umiltà, 1996). These data suggest that humans
may be born with a template or representation of the human face
(Johnson & Morton, 1991). However, the role of these innate
biases and of the rapid early learning involved in shaping the
face-perception system remains an open question (de Schonen,
1989; de Schonen, Mancini, & Leigeois, 1998).
The hypothesized existence of special mechanisms for face
detection at birth suggests that newborns may begin life with a
broad face representation (Johnson & Morton, 1991). In contrast,
some authors maintain that newborns’ face representation might be
sufficiently detailed from birth (Meltzoff & Moore, 1977; Slater &
Kirby, 1998; Slater et al., 1998). Data supporting this hypothesis
have been generated by several studies that have demonstrated that
newborns imitated a variety of facial gestures. This result is
difficult to explain if the infant does not have a representation of
his or her own face (Meltzoff & Moore, 1977). Quinn and Slater
(2003) hypothesized that face representation is the product of
evolutionary pressure and would benefit from proprioception in
utero. Indeed, proprioceptive feedback provided by facial movements could contribute to the formation of a face-specific representation that is present at birth.
Currently, most theorists agree that the face-perception system
at birth is the product of a conjunction of evolutionary inheritance,
in utero learning, and rapid learning after birth (de Schonen, 1989;
Pascalis & Kelly, 2009; Sai, 2005; Slater et al., 2010). In other
words, in the same way that evolutionary selection builds a neural
“scaffolding” that facilitates language learning, a similar mechanism facilitates the development of the face-perception system.
Several studies support the notion that early learning with regard
to faces happens from birth. According to Walton and Bower
(1993), newborns may even be able to form a face prototype in less
than 1 min after birth. The early impact of experience is evidenced
by the recognition of individual faces, such as the mother’s face
over a stranger’s face (Bushnell, Sai, & Mullin, 1989; Field,
Cohen, Garcia, & Greendburg, 1984; Pascalis, de Schonen, Morton, Deruelle, & Fabre-Grenet, 1995). In addition, intersensory
contingency facilitates learning at birth; newborns show a preference for their mother’s face if they have had postnatal exposure to
the mother’s voice and face paired together (Sai, 2005).
According to some authors, the face-perception system becomes
“tuned” to human faces as a direct consequence of extensive
Elisa Di Giorgio and Irene Leo, Dipartimento di Psicologia dello Sviluppo e della Socializzazione, Università degli Studi di Padova, Padova,
Italy; Olivier Pascalis, Laboratoire de Psychologie et Neurocognition,
Universitè Pierre Mendes, Grenoble, France; Francesca Simion, Dipartimento di Psicologia dello Sviluppo e della Socializzazione, Università
degli Studi di Padova, Padova, Italy.
Research was supported by Ministero dell Università Grant
2007XFM93B_004 and University of Padua Grant 2007-CPDA075245,
awarded to Francesca Simion. Elisa Di Giorgio was supported by a PhD
Grant from Fondazione della Cassa di Risparmio di Padova e Rovigo. We
thank Dalla Barba and the nursing staff at the Pediatric Clinic for their
collaboration.
Correspondence concerning this article should be addressed to Elisa Di
Giorgio, Dipartimento di Psicologia dello Sviluppo e della Socializzazione,
Università degli Studi di Padova, via Venezia 8, 35131 Padova, Italy.
E-mail: elisa.digiorgio@unipd.it
1
2
DI GIORGIO, LEO, PASCALIS, AND SIMION
experience of faces provided by the species-typical environment
within the first days of life (Nelson, 2003; Scott, Pascalis, &
Nelson, 2007). However, how general the face-perception system
is at birth and the role of visual experience in shaping this system
remain open questions. Specifically, it appears to be relevant to
determine whether the face-perception system at birth is speciesspecific or broad enough to draw newborns’ attention toward faces
belonging to different species, such as human and nonhuman
primate faces (de Schonen, 1989).
A dramatic demonstration of the existence of a broad faceperception system, which becomes specialized as a function of
early visual input, is given in Sugita’s (2008) study. Infant Japanese macaques (Macaca Fuscata) were separated from their parents at birth and reared by human caregivers who wore masks to
ensure that the monkeys were not exposed to faces of any kind for
a period of 6 –24 months. Following the face-deprivation period,
the monkeys were then introduced to either fellow macaques or
humans. When tested for face preference during the deprivation
period, the monkeys showed a preference for both monkey and
human faces over objects but no preference for either category of
face when they were presented simultaneously. When the monkeys
were tested a month after the end of the deprivation period, their
behavioral performance had changed dramatically. The monkeys
that were exposed to human faces retained their preference for
human faces over objects, but they also preferred to look at human
faces over monkey faces. The monkeys that were exposed to
monkey faces preferred monkey faces over human faces. Surprisingly, the period of deprivation did not have a significant impact
on their performance in preference or recognition tasks at any
stage. This study demonstrated that the face-perception system is
not species-specific at birth, at least in monkeys, and that only
extensive experience with one kind of face had an impact on
shaping and specializing the system.
A demonstration of the impact of experience in shaping the
face-perception system of humans can be found in a recent study
by Heron-Delaney, Wirth, and Pascalis (2011). This study demonstrated that a few days’ worth of visual experience is sufficient
to render the system species-specific, enabling newborns to prefer
a human face over a monkey face. Specifically, the nature of
infants’ early representations at birth was investigated to disentangle the question of whether newborns, as well as infants, could
discriminate humans from nonhuman primates and whether there
was a preference for humans from birth and throughout the first 6
months of life. The findings demonstrated that 3.5- and 6-monthold infants paid more attention to pictures of human beings than of
nonhuman primates (a gorilla or a macaque). The same preference
was observed when infants were presented with the whole body or
the face only. A preference for humans was also observed in
newborns only in the face-only condition. The authors concluded
that newborns’ exposure to human faces during their first few days
of life is enough to allow them to develop a representation of
human faces that is precise enough to differentiate human from
nonhuman primate faces.
However, the sensory hypothesis may explain these results. This
hypothesis suggests that certain classes of stimuli are preferred by
newborns as a result of the general properties of the early stages of
early visual processing (Banks & Salapatek, 1981). The sensory
hypothesis (based on the prediction of the linear system model)
maintains that the attractiveness of a pattern is determined solely
by the effective energy of that pattern. As the stimuli used in
Heron-Delaney et al.’s study (2011) were not equated as regards
their low-level perceptual properties, the preference that was observed may be due solely to a difference in the visibility of the two
stimuli (e.g., a difference in contrast). Moreover, in these pictures,
the nonhuman primate faces stand out, as their fur is dramatically
different from the humans’ skin. This means that the perceptual
cue that the newborns used may have been the presence of fur and
not the difference in the global configuration of the two faces.
In light of these considerations, the preference for human faces
over monkey faces needs to be replicated with stimuli that have
been matched for their low-level perceptual properties. The aim of
the present study is to determine whether the system which underpins face preference at birth is sufficiently general to bias
newborns’ attention equally toward human and nonhuman primate
faces that have been equated for all low-level variables. In other
words, we aim to investigate whether only a few hours of visual
experience with human faces are sufficient to bias newborns’
face-perception system toward human faces. To prevent differences in the perceptual characteristics of the stimuli, such as the
presence of fur or different external contours, we equated the two
faces for all low-level perceptual properties (i.e., low spatial frequencies and high-contrast areas). The only exception was the
eyes, because the contrast between the sclera and the iris of the
human eye differs from that of a monkey’s eye. The salience of
eyes plays an essential role in learning faces, as previously demonstrated by a computational model of face processing (Acerra,
Burnoud, & de Schonen, 2002) and by behavioral studies (Geldar,
Maurer, & Carney, 1999; Maurer & Barrera, 1981).
Three experiments were carried out to determine whether newborns (a) show a spontaneous preference between a human and a
monkey face equated for all low-level perceptual properties (Experiment 1), (b) are able to discriminate between a human face and
a monkey face (Experiment 2), and (c) show a preference for an
upright monkey face over an upside-down monkey face (Experiment 3).
Experiment 1
In line with previous studies carried out on humans and animals
that have demonstrated the existence of a preference for faces at
birth (Macchi Cassia et al., 2004; Sugita, 2008), the aim of Experiment 1 was to investigate whether newborns exhibit a spontaneous visual preference for human faces when contrasted with
nonhuman primate faces when the stimuli look similar in terms of
their global shape and the brightness and contrast are as equal as
possible.
Method
Participants. Eighteen typical, healthy, full-term Caucasian
newborns were recruited from the Paediatric Clinic of the University of Padova, Italy. Six infants were removed from the study for
the following reasons: Three changed state during the test (the
newborn became too tired or started to cry), and three had a strong
position bias (they looked in one direction for more than 80% of
the time). Therefore, the final sample consisted of 12 newborns
(six boys). All of the infants met the screening criteria of a normal
delivery, a birth weight between 2,570 and 3,980 g, and a 5-min
FACE-PERCEPTION SYSTEM AT BIRTH
Apgar score above 8. Their age at the time of testing ranged from
24 to 72 hr. The newborns were tested only if they were awake and
in an alert state (Prechtl & O’Brien, 1982) and after the parents had
given their informed consent.
Stimuli. Grayscale digitized full-frontal images of two human
and two monkey faces (Rhesus Macaques) were prepared using
Adobe Photoshop 7.0 (see Figure 1). The stimuli were taken from
the Psychological Image Collection at Starling (PICS; http://
pics.stir.ac.uk). Two pairs of stimuli comprising a human and a
monkey face were presented side-by-side on the screen. The monkey faces were manipulated by removing the hair from the cheeks
to equalize the monkey and human faces for all low-level variables
(i.e., low spatial frequency, contrast, luminance). The stimuli were
approximately 18.5 cm high ⫻ 13.5 cm wide (with a visual angle
of approximately 34° ⫻ 26°).
Apparatus. The newborn sat on the experimenter’s lap, 30
cm from a 30-in. (76.2-cm) computer screen (2,560 ⫻ 1,600
pixels). The infant’s eyes were aligned with a red flickering LED
at the center of the screen, which was used to attract the infant’s
gaze at the start of each trial and to check that the infant’s sight
was level with the horizontal midline of the screen during the test.
Both stimuli were projected bilaterally on a black screen at a
distance of approximately 8 cm (15°) from a central fixation point.
A video camera, mounted above the central screen, recorded the
infant’s eye movements. To prevent interference from irrelevant
distracters, plain white curtains were drawn on both sides of the
chair on which the experimenter sat with the baby on her lap.
Procedure. The experiment was carried out using a preferential looking procedure. Every trial began with the flashing light
in the center of the screen. As soon as the infant fixated on the
light, one of the experimenters, who was watching the infant’s
Figure 1.
and 2.
Human and monkey face images employed in Experiments 1
3
eyes by means of a video monitoring system, started the trial
sequence by pressing a key on the computer keyboard. This
automatically turned off the central LED and activated the slide
projector, which presented the stimuli on the screen. The stimuli
remained on the screen for as long as the infant fixated on one of
them (infant control procedure). When the infant shifted his or her
gaze from the display for more than 10 s, the observer turned off
the stimuli and the central light turned on automatically. All of the
infants underwent two trials in which the two stimuli were shown
bilaterally, one on the left and one on the right of the central LED.
The left/right position of the stimuli was counterbalanced between
the trials. The session ended when the baby did not look at either
stimulus for more than 10 s. The videotape of the infant’s eye
movements throughout the trial was subsequently analyzed frameby-frame by two coders (an experimenter and a student). Both
coders were unaware of the kind of stimulus presented. The coders
recorded for each stimulus and each position the total fixation time
(i.e., the sum of all fixations) and the number of orienting responses. The mean estimated reliability between the online and
offline coding was r(10) ⫽ .90, p ⬍ .001, N ⫽ 12 (Pearson’s
correlation) for either dependent variable (total fixation time, discrete number of looks, the duration of the longest fixation and the
duration of the first fixation).
Results and Discussion
To determine whether the newborns showed a spontaneous
visual preference for one of the two stimuli, separate two-tailed
dependent samples t tests were performed. As regards the normality of the data distribution, the Kolmogorov-Smirnov test showed
that the data were normally distributed.
The results showed no evidence of preference for the human
face for any of the dependent variables when the group of subjects
was considered. Neither the duration of the longest fixation (human face, M ⫽ 13.4 s, SD ⫽ 6.1, vs. monkey face, M ⫽ 12.2 s,
SD ⫽ 6.2), t(11) ⫽ 0.48, ns, nor the duration of the first fixation
on the human face (human face, M ⫽ 5.2 s, SD ⫽ 3.2, vs. monkey
face, M ⫽ 4.1 s, SD ⫽ 3.6), t(11) ⫽ 0.98, ns, reached the level of
significance. Furthermore, the newborns did not look at the human
face for longer (M ⫽ 33.2 s, SD ⫽ 13.9) than the monkey face
(M ⫽ 32.5 s, SD ⫽ 15.1), t(11) ⫽ .09, ns.
We conducted additional analyses on preference scores (percentages) for the human face. The length of time for which each
infant looked at the human face was divided by the total time spent
looking at both test stimuli. The score was then multiplied by 100.
Therefore, only scores significantly above 50% indicate a preference for the human face compared with the monkey face. Preference was not above the chance level of 50% (M ⫽ 50%, SD ⫽
18.1), one-sample t(11) ⫽ 0.12, ns, two-tailed. Six newborns
preferred to look at the human face for more than 55% of their total
fixation time (M ⫽ 65%). The other six newborns preferred to look
at the monkey face for more than 53% of their total fixation time
(M ⫽ 62%). Furthermore, if we consider the number of orienting
responses, the newborns did not orient more frequently to the
human face (M ⫽ 13.2, SD ⫽ 5.5) than the monkey face (M ⫽ 14,
SD ⫽ 5.3), t(11) ⫽ .38, ns. Finally, the correlation between the age
of the newborns (h) and the preference score was not significant
(r ⫽ .06, ns).
DI GIORGIO, LEO, PASCALIS, AND SIMION
4
Heron-Delaney et al. (2011) found a preference for human faces
over monkey faces during the first week of life in an experimental
condition in which hair/fur was not removed. In contrast, our
results demonstrate that the newborns did not show any spontaneous visual preference when the two faces were equated for lowlevel perceptual properties, even though the eyes of the two faces
differed because of the contrast between the sclera and the iris.
One possible interpretation of the lack of visual preference is that
the face-perception system during the first week of life has not
been shaped to be human-specific. However, this null result should
be interpreted with caution, as the newborns may not have been
able to differentiate between the two stimuli. Experiment 2 aimed
to test this hypothesis.
Experiment 2
Experiment 2 tested whether newborns are capable of discriminating, after habituation, a human face from a monkey face,
equated for all low-level perceptual properties.
Method
Participants.
Seventeen healthy and full-term Caucasian
newborns (five boys) from the Paediatric Clinic of the University
of Padova were tested. Three newborns changed their state during
testing and were excluded from statistical analysis. Therefore, the
final sample consisted of 14 newborns. All of the infants met the
screening criteria of a normal delivery, a birth weight between
3,100 and 3,980 g, and a 5-min Apgar score between 8 and 10.
Their age at the time of testing ranged from 24 to 72 hr. The
newborns were tested only if they were awake and in an alert state.
Informed consent was obtained from their parents.
Stimuli. The same grey scale full-frontal images of the human and monkey faces that were used in the previous experiment
were employed in Experiment 2.
Apparatus. The apparatus was the same as in Experiment 1.
Procedure. The experiment was carried out using an infantcontrol habituation procedure (Horowitz, Paden, Bhana, & Self,
1972). The infant was judged to have habituated when, from the
fourth fixation onward, the sum of any three consecutive fixations
was 50% or less of the total of the first three fixations (Slater,
Earle, Morison, & Rose, 1985). Half of the newborns habituated to
the human face, whereas the other half habituated to the monkey
face. During the habituation phase, the same stimuli were presented side-by-side. The stimuli remained on the screen until the
habituation criterion was reached. A bilateral, rather than central,
method of presentation was chosen for two reasons: First, when
newborns look at a centrally presented stimulus, it is difficult for
an observer to decide whether they are actually looking at the
stimulus or simply not moving their eyes from the central position.
Second, at birth, the photoreceptors in the central fovea are very
immature, resulting in poor vision in the central area of the visual
field (Abramov et al., 1982; Atkinson & Braddick, 1989). The
habituation phase was followed by two preference tests in which a
familiar face and a novel one were presented side-by-side. The
left-right position of the stimuli was reversed from the first to the
second presentation. Looking at one stimulus, generally the novel
one, for a longer period of time indicated discrimination and
recognition. During the preference test phase, the experimenter
recorded the duration of the infant’s fixation on each stimulus by
pressing one of two different buttons, depending on whether the
infant looked toward the right or the left position. The presentation
lasted until the infant had fixated on each stimulus on least once
and a total of 20 s of looking had been accumulated.
Results and Discussion
All of the newborns reached the habituation criterion. A oneway analysis of variance was run to compare the total fixation
times to reach the habituation criterion for the two groups of
subjects that had habituated to the human face (M ⫽ 37.6 s, SD ⫽
6.5, trials M ⫽ 9) or the monkey face (M ⫽ 37.4 s, SD ⫽ 7.7, trials
M ⫽ 8). The results of the comparison were not significant, F(1,
12) ⫽ 0.006, ns.
To test whether the newborns were able to recognize and discriminate between the novel stimulus and the familiar one, a
novelty preference score (percentage) was computed. The time that
each infant spent looking at the novel stimulus during the two test
presentations was divided by the total time spent looking at both
test stimuli during the two presentations, and subsequently converted into a percentage score. Hence, only scores that were
significantly above 50% indicated a preference for the novel
stimulus. The mean novelty preference score was 66.9% (SD ⫽
8.8), which differed significantly from the chance level of 50%,
t(13) ⫽ 20.9, p ⬍ .01 (Cohen’s d ⫽ 1.3; Cohen, 1988). When they
were habituated to a monkey face, the newborns exhibited a
novelty preference for the human face (69%), whereas when they
were habituated to a human face, they displayed a preference for
the monkey face (58%). The results of Experiment 2 disprove the
assumption that the lack of preference found in Experiment 1 is
due to the newborns’ inability to discriminate between the two
stimuli. The newborns were shown to be sensitive to differences
between the two stimuli and therefore able to discriminate and
recognize the familiarized stimulus.
Experiment 3
The purpose of Experiment 3 was to investigate whether the
newborns’ preference for monkey faces was altered by their orientation, as has been observed in previous studies using human
faces (Macchi Cassia et al., 2004; Turati, Macchi Cassia, Simion,
& Leo, 2006). If monkey faces are perceived as human faces, then
a preference for an upright monkey face over an inverted monkey
face should be found in newborns.
Method
Participants. The participants were 12 full-term healthy Caucasian newborns (eight boys), recruited from the maternity ward of
the Paediatric Clinic of the University of Padova. One other
newborn was excluded from the final sample because his state
changed during testing. All of the infants met the screening criteria
of a normal delivery, a birth weight between 2,620 and 3,700 g,
and a 5-min Apgar score between 8 and 10. Their age at the time
of testing ranged from 24 to 72 hr. The method used to select the
newborns was identical to that which was used in Experiments 1
and 2.
FACE-PERCEPTION SYSTEM AT BIRTH
5
Stimuli. The same grayscale full-frontal monkey face images
used in Experiments 1 and 2 were employed (see Figure 2). An
upright monkey face was contrasted with the same monkey face
that had been inverted.
Apparatus and procedure.
The apparatus and procedure
were identical to those described in Experiment 1.
Results and Discussion
To test the newborns’ preferences, one dependent-samples t test
(two-tailed) was performed for each dependent variable (duration
of the longest fixation and first fixation, total fixation time and
number of orientations).
The duration of the longest fixation on the upright monkey face
(M ⫽ 15.5 s, SD ⫽ 8.2) was significantly different from the
duration of the longest fixation on the inverted monkey face (M ⫽
10.6 s, SD ⫽ 4.1), t(11) ⫽ 2.3, p ⬍ .05 (Cohen’s d ⫽ 0.70).
Moreover, the newborns looked at the upright monkey face for
significantly longer (M ⫽ 38.7 s, SD ⫽ 15.9) than at the same
monkey face that was inverted (M ⫽ 25.5 s, SD ⫽ 7.3), t(11) ⫽
2.5, p ⬍ .05 (Cohen’s d ⫽ 0.70; see Figure 3). In addition,
preference scores for the upright monkey face were above the
chance level of 50% (M ⫽ 59%, SD ⫽ 12), t(11) ⫽ 2.5, p ⬍ .05
(Cohen’s d ⫽ 0.70).
Furthermore, the t test that was performed on the number of
orientations revealed that the newborns did not orient more frequently to the upright monkey face (M ⫽ 12, SD ⫽ 2.7), rather
than the inverted monkey face (M ⫽ 10, SD ⫽ 2.1), t(11) ⫽ 1.8,
ns. An examination of the data for individual infants found that
eight out of the 12 newborns looked at the upright monkey face for
longer (binomial tests, ns). Moreover, eight out of the 12 newborns
oriented more frequently toward the upright monkey face (binomial tests, ns). The duration of the first fixation did not attain
statistical significance (ns). Finally, as in Experiment 1, the correlation between the age of the newborns (h) and the preference
score was not significant (r ⫽ .02, ns).
The results of Experiment 3 show that, as with human faces, the
manipulation of the orientation of a monkey face affects newborns’ visual preferences. This outcome demonstrates that, as for
human faces, the newborns showed a preference for the upright
monkey face over the same face that had been inverted. The
upright orientation effect has already been demonstrated with
face-like patterns (Johnson & Morton, 1991; Valenza et al., 1996)
and with photographs of real faces (Macchi Cassia et al., 2004).
Figure 2.
Stimuli employed in Experiment 3.
Figure 3. Total fixation time on upright monkey face versus inverted
monkey face in Experiment 3. ⴱ p ⬍ .05.
The presence of the orientation effect with a primate face can be
interpreted as showing that newborns react in the same way to
human and monkey faces.
General Discussion
The present study aimed to address the question of the speciesspecificity of the human face-perception system within the first
few days of life. Starting from a previous study in which a
preference for human faces over monkey faces was demonstrated
(Heron-Delaney et al., 2011), we investigated whether this preference might be explained by a difference in the low-level perceptual properties of the stimuli. For these reasons, in the present
study, the human faces and monkey faces were equated for lowlevel perceptual properties.
The results of Experiment 1 demonstrate that the newborns did
not show any visual preference for a human face over a face that
is human-like (primate face), corroborating the idea that newborns’
face-perception system is not human-specific during the first week
of life. The lack of preference must be interpreted cautiously,
however, as it may be due to the newborns’ inability to differentiate between the two stimuli that were presented. Experiment 2
disproves this interpretation, because it demonstrates that the newborns were able to discriminate between the two stimuli employed
in the preference test in Experiment 1. Finally, Experiment 3
demonstrated that, as with human faces (Macchi Cassia et al.,
2004), the manipulation of the orientation of a monkey face affects
newborns’ visual preferences. The preference for the upright monkey face is interpreted as demonstrating that newborns react to
nonhuman primate faces in the same way as previously demonstrated for human faces.
This preference for the upright monkey face can be interpreted
in two ways. The first refers to the hypothesis that one critical
perceptual property that elicits a face preference in newborns is the
presence of a greater number of elements in the upper part of the
configuration (i.e., up-down asymmetry; Simion, Valenza, Macchi
Cassia, Turati, & Umiltà, 2002). Therefore, newborns may prefer
the upright monkey face because of the presence of more elements
(i.e., the eyes) in the upper part. The second interpretation refers to
the presence of first-order configural information present in the
6
DI GIORGIO, LEO, PASCALIS, AND SIMION
monkey face as well as in the human face (Simion et al., 2002).
Indeed, recent results using Mooney face patterns (in which features are formed of patches of intense light and shadow only and
require closure) support the role of first-order configural information. The results demonstrate that when a Mooney face is contrasted with a Mooney stimulus, such as a butterfly equated for the
number of elements in the upper part of the configuration, newborns prefer to look at holistic patterns that are closer to upright
faces (Leo & Simion, 2009).
There is prominent evidence in support of the proposal that at
birth, the face-perception system is broad and unspecified and that
it is shaped by the faces that are experienced in the visual environment in the first few months of life (de Schonen & Mathivet,
1989; Nelson, 2001; Sugita, 2008). However, as Nelson (2003)
pointed out, it is not clear what kind of experience is necessary and
when and for how long this experience needs to occur. The
experiments in this study were designed to disentangle these
issues. The findings of Experiment 1 are consistent with a large
proportion of the literature on face-perception at birth in both
animals (Sugita, 2008) and humans (Kelly et al., 2005; Quinn et
al., 2008), which reveals clear evidence of a basic, coarsely tuned
face-perception system in primates as well as humans at birth.
Newborns do not show any visual preference for faces from their
own or other ethnic groups (Kelly et al., 2005), and they do not
respond differentially to the gender of the faces that are presented
(Quinn et al., 2008).
The lack of a preference between a monkey face and a human
face contradicts previous results that have demonstrated a preference for human faces over monkey faces in newborns (HeronDelaney et al., 2011). However, in the study in question, the
monkey and human faces were not equated for low-level perceptual properties. A possible alternative interpretation could be based
on the sensory hypothesis (Banks & Salapatek, 1981), as the two
stimuli differed in terms of the amount of effective energy in the
two patterns. Consequently, a more parsimonious interpretation is
that newborns’ preference for human faces may be due to the
physical and perceptual characteristics of the pictures used.
Importantly, the only perceptual characteristic that differentiates
the two stimuli in the present study is the presence in the human
face alone of the correct contrast polarity within the eyes (i.e., a
black pupil surrounded by a white sclera). The human species is
the only primate species with a white sclera and a dark iris
(Kobayashi & Kohshima, 1997). Although the presence of this
perceptual characteristic might explain newborns’ discrimination
in Experiment 2, it fails to explain the absence of a preference for
the same stimuli in Experiment 1. In addition, the results of
Experiment 2, which show that newborns are able to discriminate
between human and monkey faces, extend our knowledge of the
face-perception system and are consistent with the literature. Indeed, previous studies have demonstrated that newborns are able to
discriminate between faces from their own ethnic group (Pascalis
& de Schonen, 1994) on the basis of both the inner features and the
outer contours (Turati et al., 2006).
Overall, our results show that newborns do not respond differently to human and monkey faces, showing that the human faceperception system is rather broad during the first week of life. The
presence of a preference for the upright monkey face in Experiment 3 corroborates our conclusion that newborns react in the
same way to human and monkey faces. Nonhuman primates
clearly share a similar face arrangement to humans that can be
expected to be processed in a similar way. Campbell et al. (1997)
explored this hypothesis by testing adult human subjects’ categorical perceptions of three categories of various computer-morphed
faces: human-monkey, monkey-cow, and human-cow faces. The
authors showed that the perceptual discrimination boundaries for
monkey-human morphs were less distinct than for monkey-cow or
human-cow faces. Therefore, in adults, a single category formed
the basis for the discrimination of both human and monkey faces,
suggesting that similar mechanisms are involved in processing
these two categories of stimuli (Campbell et al., 1997).
The present study, in addition to extending our knowledge of the
face-perception system at birth, opens up some intriguing questions. Do newborns process monkey faces as human faces? Do
human faces and monkey faces belong to the same category? To
what degree is newborns’ face representation “primate”-like? In
effect, if a primate face representation exists during the first few
days of life, at birth, we should observe a systematic preference for
human and nonhuman primate faces over faces belonging to other
species (e.g., dog, cow, bird). Starting from a recent study that
showed that 4- to 6-month-old infants, but not 9- to 11-month-old
infants, were able to discriminate nonprimate faces (i.e., sheep
faces) from primate ones (Simpson, Varga, Frick, & Fragaszy,
2011), it could be interesting to investigate whether infants’ faceperception system is broadly attuned at the beginning not only to
primate faces but to nonprimate faces as well.
A second series of questions that remains open concerns the role
of eyes. It has been suggested that eyes are important in determining newborns’ orientation toward faces (Acerra et al., 2002; Batki,
Baron-Cohen, Wheelwright, Connellan, & Ahluwalia, 2000; Farroni, Csibra, Simion, & Johnson, 2002). The results of the present
study seem to show that the contrast between the sclera and the
iris, which is present in human eyes, does not determine any
preference. Future studies should investigate what visual properties render human eyes so attractive from birth.
Overall, data are in line with the hypothesis that the faceperception system is not human-specific after a few days of life
and demonstrates that a few hours of visual experience of human
faces is not enough to render the system human-specific. The
results corroborate the existing literature on face-perception and
provide evidence of the presence of a face-perception system that
is not human-specific at birth but that biases newborns’ visual
attention toward human and nonhuman primate faces. Because of
visual experience of a species-specific environment, this faceperception system becomes tuned to human faces during development (Nelson, 2001; Pascalis, de Haan, & Nelson, 2002; Pascalis
& Kelly, 2009). This conclusion is supported by studies demonstrating that 3 months’ worth of visual experience is sufficient not
only to induce a gender (Quinn, Yahr, Kuhn, Slater, & Pascalis,
2002), own-species (Heron-Delaney et al., 2011), and own-race
preference (Kelly et al., 2005) but also an own-race effect in face
processing (Sangrigoli & de Schonen, 2004).
References
Abramov, I., Gordon, J., Hendrickson, A., Hainline, L., Dobson, V., &
LaBossiere, E. (1982). The retina of the newborn human infant. Science,
217, 265–267. doi:10.1126/science.6178160
Acerra, F., Burnod, I., & de Schonen, S. (2002). Modelling aspects of face
FACE-PERCEPTION SYSTEM AT BIRTH
processing in early infancy. Developmental Science, 5, 98 –117. doi:
10.1111/1467-7687.00215
Atkinson, J., & Braddick, O. (1989). Development of basic visual functions. In A. Slater & G. Bremner (Eds.), Infant development (pp. 7– 41).
London, England: Erlbaum.
Banks, M. S., & Salapatek, P. (1981). Infant pattern vision: A new
approach based on the contrast sensitivity function. Journal of Experimental Child Psychology, 31, 1– 45. doi:10.1016/0022-0965(81)90002-3
Batki, A., Baron-Cohen, S., Wheelwright, S., Connellan, J., & Ahluwalia,
J. (2000). Is there an innate gaze module? Evidence from human neonates. Infant Behavior & Development, 23, 223–229. doi:10.1016/
S0163-6383(01)00037-6
Bushnell, I. W. R., Sai, F., & Mullin, J. T. (1989). Neonatal recognition of
the mother’s face. British Journal of Developmental Psychology, 7,
3–15. doi:10.1111/j.2044-835X.1989.tb00784.x
Campbell, R., Pascalis, O., Coleman, M., Wallace, S. B., & Benson, P. J.
(1997). Are faces of different species perceived categorically by human
observers? Proceedings of the Royal Society B: Biological Sciences,
264, 1429 –1434. doi:10.1098/rspb.1997.0199
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed). Hillsdale, NJ: Erlbaum.
de Schonen, S. (1989). Some reflections on brain specialization in faceness
and physiognomy processing. In A. Young & H. D. Ellis (Eds.), Handbook of research on face processing (pp. 379 –389). Amsterdam, the
Netherlands: North Holland.
de Schonen, S., Mancini, J., & Leigeois, F. (1998). About functional
cortical specialization: The development of face recognition. In F.
Simion & G. Butterworth (Eds.), The development of sensory, motor and
cognitive capacities in early infancy (pp. 103–120). Hove, England:
Psychology Press.
de Schonen, S., & Mathivet, E. (1989). First come, first served: A scenario
about the development of hemispheric specialization in face recognition
during infancy. European Bulletin of Cognitive Psychology, 9, 3– 44.
Fantz, R. L. (1961). The origin of form perception. Scientific American,
204, 66 –73. doi:10.1038/scientificamerican0561– 66
Farroni, T., Csibra, G., Simion, F., & Johnson, M. H. (2002). Eye contact
detection in humans from birth. Proceedings of the National Academy of
Sciences of the United States of America, 99, 9602–9605. doi:10.1073/
pnas.152159999
Field, T. M., Cohen, D., Garcia, R., & Greenberg, R. (1984). Mothe–
stranger face discrimination by the newborn. Infant Behavior & Development, 7, 19 –25. doi:10.1016/S0163-6383(84)80019-3
Geldart, S., Maurer, D., & Carney, K. (1999). Effects of eye size on adults’
aesthetic rating of faces and 5-month-olds’ looking times. Perception,
28, 361–374. doi:10.1068/p2885
Goren, C. C., Sarty, M., & Wu, P. Y. K. (1975). Visual following and
pattern discrimination of face-like stimuli by newborn infants. Pediatrics, 56, 544 –549.
Heron-Delaney, M., Wirth, S., & Pascalis, O. (2011). Infants’ knowledge
of their own species. Philosophical Transactions of the Royal Society B:
Biological Sciences, 366, 1753–1763. doi:10.1098/rstb.2010.0371
Horowitz, F. D., Paden, L., Bhama, K., & Self, P. (1972). An infant-control
procedure for studying infant visual fixation. Developmental Psychology, 7, 90. doi:10.1037/h0032855
Johnson, M. H., & Morton, J. (1991). Biology and cognitive development:
The case of face recognition. Oxford, England: Blackwell.
Kelly, D. J., Slater, A. M., Lee, K., Gibson, A., Smith, M., Ge, L., &
Pascalis, O. (2005). Three-month-olds, but not newborns, prefer ownrace faces. Developmental Science, 8, F31–F36. doi:10.1111/j.14677687.2005.0434a.x
Kobayashi, H., & Kohshima, S. (1997). Unique morphology of the human
eye. Nature, 387, 767–768. doi:10.1038/42842
Leo, I., & Simion, F. (2009). Newborns’ Mooney-face-perception. Infancy,
14, 641– 653. doi:10.1080/15250000903264047
7
Macchi Cassia, V., Turati, C., & Simion, F. (2004). Can a nonspecific bias
toward top-heavy patterns explain newborns’ face preference? Psychological Science, 15, 379 –383. doi:10.1111/j.0956-7976.2004.00688.x
Maurer, D., & Barrera, M. (1981). Infants’ perception of natural and
distorted arrangements of a schematic face. Child Development, 52,
196 –202. doi:10.2307/1129230
Meltzoff, A. N., & Moore, M. K. (1977). Imitation of facial and manual
gestures by human neonates. Science, 198, 75–78. doi:10.1126/
science.198.4312.75
Nelson, C. A. (2001). The development and neural bases of face recognition. Infant and Child Development, 10, 3–18. doi:10.1002/icd.239
Nelson, C. A. (2003). The developmental of face recognition reflects an
experience-expectant and experience-dependent process. In O. Pascalis
& A. Slater (Eds.), The development of face processing in infancy and
early childhood: Current perspectives (pp. 79 – 88). New York, NY:
Nova Science.
Pascalis, O., de Haan, M., & Nelson, C. A. (2002). Is face processing
species-specific during the first year of life? Science, 296, 1321–1323.
doi:10.1126/science.1070223
Pascalis, O., & de Schonen, S. (1994). Recognition memory in 3- to
4-day-old human neonates. NeuroReport, 5, 1721–1724. doi:10.1097/
00001756-199409080-00008
Pascalis, O., de Schonen, S., Morton, J., Deruelle, C., & Fabre-Grenet, M.
(1995). Mother’s face recognition by neonates: A replication and an
extension. Infant Behavior & Development, 18, 79 – 85. doi:10.1016/
0163-6383(95)90009-8
Pascalis, O., & Kelly, D. J. (2009). The origins of face processing in
humans: Philogeny and ontogeny. Perspectives on Psychological Science, 4, 200 –209. doi:10.1111/j.1745-6924.2009.01119.x
Prechtl, H., & O’Brien, M. J. (1982). Behavioral states of the full term
newborn: The emergence of a concept. In P. Stratton (Ed.), Psychobiology of the human newborn (pp. 53–73). New York, NY: Wiley.
Quinn, P. C., & Slater, A. (2003). Face-perception at birth and beyond. In
O. Pascalis & A. Slater (Eds.), The development of face processing in
infancy and early childhood: Current perspectives (pp. 3–11). Huntington, NY: Nova Science.
Quinn, P. C., Uttley, L., Lee, K., Gibson, A., Smith, M., Slater, A., &
Pascalis, O. (2008). Infant preference for female faces occurs for samebut not other-race faces. Journal of Neuropsychology, 2, 15–26. doi:
10.1348/174866407X231029
Quinn, P. C., Yahr, J., Kuhn, A., Slater, A. M., & Pascalis, O. (2002).
Representation of the gender of human faces by infants: A preference for
female. Perception, 31, 1109 –1121. doi:10.1068/p3331
Sai, F. (2005). The role of the mother’s voice in developing mother’s face
preference: Evidence for intermodal perception at birth. Infant and Child
Development, 14, 29 –50. doi:10.1002/icd.376
Sangrigoli, S., & de Schonen, S. (2004). Recognition of own-race and other-race
faces by three-month-old infants. Journal of Child Psychology and Psychiatry,
45, 1219–1227. doi:10.1111/j.1469-7610.2004.00319.x
Scott, L. S., Pascalis, O., & Nelson, C. A. (2007). A domain-general theory of the
development of perceptual discrimination. Current Directions in Psychological
Science, 16, 197–201. doi:10.1111/j.1467-8721.2007.00503.x
Simion, F., Valenza, E., Macchi Cassia, V., Turati, C., & Umiltà, C. A.
(2002). Newborns’ preference for up-down asymmetrical configurations. Developmental Science, 5, 427– 434. doi:10.1111/14677687.00237
Simpson, E. A., Varga, K., Frick, J. E., & Fragaszy, D. (2011). Infants
experience perceptual narrowing for non-primate faces. Infancy, 16,
318 –328. doi:10.1111/j.1532-7078.2010.00052.x
Slater, A., Earle, D. C., Morison, V., & Rose, D. (1985). Pattern preferences at birth and their interaction with habituation-induced novelty
preferences. Journal of Experimental Child Psychology, 39, 37–54.
doi:10.1016/0022-0965(85)90028-1
Slater, A., & Kirby, R. (1998). Innate and learned perceptual abilities in the
8
DI GIORGIO, LEO, PASCALIS, AND SIMION
newborn infant. Experimental Brain Research, 123, 90 –94. doi:10.1007/
s002210050548
Slater, A., Quinn, P. C., Kelly, D. J., Lee, K., Longmore, C. A., McDonald,
P. R., & Pascalis, O. (2010). The shaping of the face space in early
infancy: Becoming a native face processor. Child Development Perspectives, 4, 205–211. doi:10.1111/j.1750-8606.2010.00147.x
Slater, A., von der Schulenburg, C., Brown, E., Badenoch, M., Butterworth, G., Parsons, S., & Samuels, C. (1998). Newborn infants prefer
attractive faces. Infant Behavior & Development, 21, 345–354. doi:
10.1016/S0163-6383(98)90011-X
Sugita, Y. (2008). Face-perception in monkeys reared with no exposure to
faces. Proceedings of the National Academy of Sciences of the United
States of America, 105, 394 –398. doi:10.1073/pnas.0706079105
Turati, C., Macchi Cassia, V., Simion, F., & Leo, I. (2006). Newborns’ face
recognition: Role of inner and outer facial features. Child Development,
77, 297–311. doi:10.1111/j.1467-8624.2006.00871.x
Valenza, E., Simion, F., Macchi Cassia, V., & Umiltà, C. (1996). Face
preference at birth. Journal of Experimental Psychology: Human
Perception and Performance, 22, 892–903. doi:10.1037/00961523.22.4.892
Walton, G. E., & Bower, T. G. R. (1993). Newborns form “prototypes” in
less than 1 minute. Psychological Science, 4, 203–205. doi:10.1111/
j.1467-9280.1993.tb00488.x
Received January 7, 2011
Revision received September 30, 2011
Accepted October 14, 2011 䡲