Review Biological

Psychiatry

Autism and the Social Brain: The First-Year

Puzzle
Mayada Elsabbagh and Mark H. Johnson

ABSTRACT
The atypical features of social perception and cognition observed in individuals with a diagnosis of autism have been
explained in two different ways. First, domain-specific accounts are based on the assumption that these end-state
symptoms result from specific impairments within component structures of the social brain network. Second,
domain-general accounts hypothesize that rather than being localized, atypical brain structure and function are
widespread, or hypothesize that the apparent social brain differences are the consequence of adaptations to earlier
occurring widespread changes in brain function. Critical evidence for resolving this basic issue comes from
prospective longitudinal studies of infants at risk for later diagnosis. We highlight selected studies from the newly
emerging literature on infants at familial risk for autism to shed light on this issue. Despite multiple reports of possible
alterations in brain function in the first year of life, overt behavioral symptoms do not emerge until the second year.
Our review reveals only mixed support, within this very early period, for localized deficits in social brain network
systems and instead favors the view that atypical development involving perceptual, attentional, motor, and social
systems precede the emerging autism phenotype.
Keywords: Autism, Brain development, Eye tracking, Imaging, Infant, Risk
http://dx.doi.org/10.1016/j.biopsych.2016.02.019

Autism, a developmental disorder characterized by suboptimal
social and communication behavior, is commonly attributed to
the atypical development of brain networks subserving social
perception and cognition. However, the fundamental question
of why the so-called social brain network is differentially
affected remains unanswered. For this reason, over the last
decade, several groups have begun the prospective study of
infants who are at increased risk for developing autism (1–3).
Later-born siblings of children with autism are more likely to
receive a diagnosis of autism themselves compared with
infants with no family history of autism. Specifically, 20%
of these infants receive an autism diagnosis by 3 years of age,
and a further 20% may have other developmental difficulties or
show subclinical aspects of the phenotype (4). Emerging
results from studies of infant siblings have begun to provide
information about the developmental precursors and risk
processes that lead to the emergence of the autism phenotype
in toddlerhood. There is consensus that during the second
year of life many infants in whom autism is later diagnosed
begin to be differentiated based on their behavior from infants
at risk who do not receive a diagnosis and infants in low-risk
control groups (1–3).
Despite this progress in characterizing the emerging phe-
notype, what has been described as the first-year puzzle (1)
remains unresolved: within the earliest period of postnatal life,
overt behavioral signs of later autism are subtle and variable
and do not obviously map on the profile of atypicality seen
later (5,6). Much of the research on infants at familial risk has
centered on the attempt to identify precursors of the later
diagnosis. To date, many published studies have specifically
focused their investigations on social orienting and perception
in infancy, on the assumption that such tasks will provide
the best predictive markers of later diagnosis. As such, many
studies have adopted the general assumption that domain-
specific deficits in infancy become increasingly compounded
during postnatal development to result in the social impair-
ments observed in diagnosed autism. Infant social brain
network precursors are assumed to also be the origins of
the later emerging social, cognitive, and attention features
observed in this population (7).
In this review, we question the “social-first” assumption
prevalent in the literature on the grounds that the evidence in
support of this view is more mixed than has sometimes been
portrayed, and evidence from infants in the first year is more
consistent with subtle widespread atypicality across multiple
brain systems. Our review of the literature on infants is largely
consistent with evidence from research on adults identifying
neurocognitive differences in perceptual and cognitive sys-
tems in addition to the differences observed in social brain
networks.
SOCIAL-FIRST ASSUMPTION
Although autism is diagnosed on the basis of social and
communication impairment, the phenotype encompasses
much broader differences in perception and cognition. Several

SEE COMMENTARY ON PAGE

94 & 2016 Society of Biological Psychiatry.

Biological Psychiatry July 15, 2016; 80:94–99 www.sobp.org/journal ISSN: 0006-3223

Autism and the Social Brain
accounts, supported by research on children and adults with
autism, have shifted the focus away from social impairment.
For example, the enhanced perceptual functioning account
posits that autism is characterized by locally oriented visual
and auditory perception and enhanced low-level discrimina-
tion (8). More recently, a predictive coding account has been
applied to autism (9). Predictive coding suggests that higher
brain areas attempt to “explain” input from sensory brain areas
and then project these predictions down to lower areas, where
the predicted sensory information is subtracted from the input.
This account may explain how brain differences in autism
lead to a bias favoring local over global processing. More
specifically, hypersensitivity in visual or auditory processing
may underlie symptoms such as altered social engagement
and speech delay. These alternative accounts have also
indicated that islands of enhancement, in addition to deficits,
may better characterize some aspects of the phenotype.
Overall, these perspectives suggest that viewing autism in
terms of a focal impairment in the social brain is overly
simplistic. A more complex pattern of interactions between
social and perceptual systems may underlie the pattern of
symptoms or individual differences observed in this condition.
These alternative accounts have received much less attention
from the more recent research area examining early develop-
ment in autism, which has more often continued to pursue the
hypothesis of a social-first deficit. Studying the early emer-
gence of autism signs and symptoms can help explain the
mechanisms underlying their development before symptoms
are amplified and complicated over the course of atypical
development.
There are several good reasons why researchers studying
infants at risk for autism have tended to focus on early markers
of the social brain. First, social brain deficits are assumed to
underlie key diagnostic features of the diagnosed syndrome,
and as such emerging atypical social behaviors (usually
observed in the second year) clearly precede the diagnostic
phenotype (1–3). Second, because of the limited attention
span of infants in their first year, researchers are forced
to make pragmatic decisions about prioritizing particular
domains for tests. Third, a mature literature on the typical
development of the social brain readily allows for the adoption
of paradigms and theoretical ideas to be applied to at-risk
populations.
The social brain is a network of regions implicated in the
processing of social information, including cortical areas
such as orbitofrontal cortex, superior temporal sulcus,
temporoparietal junction, and the fusiform face area as well
as subcortical structures, such as amygdala and pulvinar
(10). Although many aspects of this adult brain specialization
remain to be understood, much progress has been made in
understanding the developmental course of social percep-
tion and orienting. Because moving the eyes is the most
important means of selecting visual input in infants and
adults, eye tracking is being increasingly used. Although eye
movement measures have been used since the 1970s,
the advent of eye-tracking technology has substantially
expanded the utility of these measures across multiple basic
and clinical research fields. Measuring cognitive phenomena
through eye tracking requires a careful experimental design,
and its use to measure “social brain” functions can be
Biological Psychiatry July 15, 2016; 80:94–99 www.sobp.org/journal
Biological
Psychiatry
viewed as a subset of studies using looking behavior to
assess cognition.
Early in postnatal life, the brain has basic orienting mech-
anisms that support rapid attention to salient stimuli, partic-
ularly those of social relevance (11). One such system is based
on a subcortical route and biases the human newborn to
attend toward faces (12). Various studies have allowed further
specification of the key characteristics of faces preferentially
attracting attention (13). These preferences are robust in the
face of manipulation of low-level perceptual and motion
features of the scenes. Although this putative system is based
on simple low-spatial frequency patterns characteristic of
faces, it is sufficient to bias the input to developing cortical
visual areas (14). In response, some of these cortical areas
increase their specialization for processing faces and related
social stimuli, resulting in increasingly selective patterns of
cortical activation with increasing age.
Beginning life with these rudimentary biases, infant social
behavior does not simply mature over development giving
rise to complex and well-developed social and communi-
cation abilities in toddlers. Instead, social orienting reflects
experience-dependent changes across perceptual, cognitive,
and social brain systems. The developmental course of
fixations changes rapidly over development (15–17): newborn
infants tend to fixate mainly around the edge of the face; later,
similar to adults, infants begin to fixate on the internal features
of the face, such as eyes and mouth. Subsequently, they show
a strong preference for the internal features of the face when
they are watching their mother’s face when it displays
communicative expressions, such as maintained eye contact,
smiling, speaking in infant-directed speech, and nodding.
Given this relatively mature literature on typical development,
it is not surprising that hypotheses on the origins of autism
based on the social-first assumption have been so actively
pursued.
ASSESSING THE SOCIAL-FIRST HYPOTHESES
Several groups have theorized that early deficits in social
information processing early in life may interfere with the
emergence of developmental milestones that are critical for
social learning, such as shared attention (7,12,18,19). Accord-
ing to this view, these cascading influences could preclude the
typical development of sociocommunicative skills, eventually
leading to deficits in language acquisition, theory of mind, and
understanding of others. In individuals with a diagnosis of
autism, it may be possible to dissociate differences in social
orienting from differences in social reward systems (7). How-
ever, at the present time, it is impossible to investigate this
putative dissociation in infants ,12 months old, and so we do
not differentiate these hypotheses in the current analysis.
Because deficits in social engagement are characteristic
symptoms of autism, the most direct and parsimonious
hypothesis regarding the origins of these symptoms is impair-
ment or absence of rudimentary social orienting biases.
Retrospective studies looking back at the first 2 years of life
(20) show less orienting toward social stimuli and a reduced
response to name calling in the first years of life in infants in
whom autism is later diagnosed compared with infants in
whom developmental delay is later diagnosed. Consistent with
95
Biological
Psychiatry
this finding, experimental studies with children in whom autism
is already diagnosed show that the children orient less to
social stimuli, such as people and voices, compared with
nonsocial stimuli, such as objects and sounds, relative to
children of the same age and developmental level (20). One
possibility is that the increased interest in objects is a
consequence of decreased interest in people.
Researchers have tested the same hypothesis in infants at
risk in whom autism was later diagnosed, using designs
ranging from controlled experiments to less structured obser-
vational studies in different home and laboratory contexts (2).
Contrary to predictions from the social-first hypotheses, in
many studies, infants who went on to develop autism as
toddlers initially showed the typical social responses expected
in the first year of life. These infants were indistinguishable on
a wide range of social orienting measures during the first year:
orienting to faces embedded among distractors (21,22), gaze
to faces and objects (22), shared affect and overall social
responsiveness to an examiner (22,23), parent-reported social
engagement and communication skills (24,25), and quality of
parent-child play interactions (26). In contrast, almost all of
these studies report more frequency and clarity of risk signs in
the second year of life. Taken together, these studies provide,
at best, mixed support for the social-first hypothesis.
Other studies focused on the bias to fixate eyes early on in
life. These studies are motivated by a large body of evidence
suggesting specific and early emerging differences in proc-
essing of eye gaze. Specifically, despite mixed evidence from
older children and adults (15), reduced fixation to eyes has
been proposed as a key feature of the emerging autism
phenotype. The hypothesis that infants who later develop
autism orient differently to the eyes has been tested more
recently using diverse stimuli and observational contexts.
Here, too, in some studies, infants initially exhibit typical
modulation of fixations to different face regions, in particular
to eyes (15,16); they also show little differences in precursors
to joint attention, including frequency of eye contact (22,23)
and reflexive gaze following (27). When trajectories of fixation
to different facial features were examined longitudinally,
several studies showed a decline in looking toward the eyes
by both typically developing infants and infants at risk
(15,16,28). This decline occurs because infants begin to
redistribute their attention away from the eyes and to other
facial features, seeking cues that are relevant for learning and
further development (15,16). However, one of these studies,
which followed infants from 2 months of age, reported a
steeper slope of decline in looking to the eyes in the subgroup
of infants who later developed autism (29). Prominent inter-
individual variability was observed in these early trajectories,
where, paradoxically, some infants in the at-risk group who
later developed autism showed greater fixation to eyes than
control subjects when tested in the first months after birth.
Taken together, these findings suggest that contrary to the
predictions generated by the social-first hypotheses, infants in
whom autism is later diagnosed may begin life with the near-
typical biases to fixate faces and eyes.
Orienting to auditory social stimuli, such as speech or other
vocal sounds, has been less systematically investigated to
date. Because an impairment in such mechanisms was
suggested to underlie later difficulties in language acquisition
96 Biological Psychiatry July 15, 2016; 80:94–99 www.sobp.org/journal
Autism and the Social Brain
(28), it will be important to examine whether infants who later
develop autism prefer speech over nonspeech early in life
similar to typically developing infants (30). Preliminary findings
of risk group differences in the selectivity of responding to
human vocal sounds have been reported, although diagnosis
outcomes are yet available (31).
Despite the lack of clear evidence for frank impairment in
social systems in the first year of life, infants at risk have
provided a window into the full range of developmental
pathways and outcomes observed in this population. Hetero-
geneity in these pathways is increasingly the focus in studies
characterizing older children with a diagnosis of autism, and
its origins remain unexplained (32). For example, in regard to
the wide range of variability in language outcomes in autism,
research on infant siblings has provided unique opportunities
to study the origins of this variability. Merin et al. (17) initially
tested the hypothesis that infants within the first year showed
similar amounts of eye and mouth fixations when an adult was
interacting with the infants but increased fixations of the eyes
relative to mouth in the “still-face” period, during which the
adult suddenly froze, became expressionless, and stopped
interacting with the infants. In the same study, infants at family
risk of autism did not differ as clearly in their scanning, but a
small subgroup of infants at risk looked more to the mouth
relative to the eyes. A follow-up study with a larger group
(including the infants in the previous study) found that more
mouth relative to eye fixations did not relate specifically to
autism but did relate to individual differences in expressive
language as assessed later in toddlerhood (16).
Another study with an independent cohort replicated and
extended these findings (15). In complex scenes requiring
more endogenous control of attention, increased scanning of
the mouth relative to the eyes in the first year of life was
associated with superior expressive language in toddlerhood.
As in the previous study, this relationship held across both
typically and atypically developing groups (i.e., infants in
whom autism or other developmental conditions were later
diagnosed). Moreover, the association is highly sensitive to
specific viewing contexts (complex scenes where there are
competing cues) and to a narrow developmental window
within the first year (15). As such, these studies suggest
continuity between the developmental mechanisms observed
in autism with the mechanisms found in typical development.
Studying visual fixations early in life is only one approach
for testing social-first hypotheses. Other hypotheses are
based on the idea of localizable deficits in structures within
the cortical social brain network. By this view, behavioral
symptoms emerge on a developmental timetable consistent
with the maturation of the regions concerned. Given chal-
lenges in imaging brains of infants, there is a much smaller
literature on functional brain development. Although, as
discussed earlier, young infants who later develop autism do
not show clear evidence of atypical overt behavior in these
areas, initial evidence from near-infrared spectroscopy studies
indicates that infants with familial risk show attenuated
responses to social stimuli within the first year of life in
posterior temporal lobe regions (31). Infants who later receive
an autism diagnosis also show atypical neural responses to
eye gaze shifts within the first year of life recorded over the
temporal lobes (33).

Autism and the Social Brain
It is likely that these atypical early responses originating
from regions of the cortical social brain contribute to the later
emerging social behavioral difficulties. However, as discussed
subsequently, it is also possible that superior temporal sulcus
and other nexus (or critical hub) regions of cortical social brain
are particularly vulnerable to generalized differences in the
efficiency or quality of neural processing across large swathes
of cortex.
FROM SOCIAL-FIRST HYPOTHESES TO
WIDESPREAD ATYPICALITY
More recent accounts of several psychiatric conditions have
shifted the focus from localizable deficits affecting isolated
structures or networks to the atypical functioning of large-
scale networks and whole-brain connectivity. As such, it is not
enough to account for sparing or impairment in response to
specific socially relevant information such as faces or eyes,
but instead we need to consider how information is integrated
across multiple brain regions and in response to a wide range
of stimuli and task demands.
Comprehensive reviews of the broad range of early markers
identified in infants at risk within the first year of life raise
substantial challenges to the social-first hypotheses (1,2,6).
These include the presence of head lag when infants are
pulled to sit, reflecting delayed motor maturation (34); higher
level of perceptual sensitivity (24); production of fewer middle
consonant types and fewer late consonant types (35); and
slower attention disengagement, reflecting generally poor
attentional flexibility [rather than specific to socially relevant
information (36) and overall lower activity level (23)]. The wide
range of these atypicalities appears more consistent with
domain-general differences during the first year.
In the light of these domain-general and widespread
effects, how can we account for differentially severe conse-
quences for social behavior in autism? Theoretical accounts
suggest that delayed or atypical specialization of function and
associated microstructure can occur as a result of a change in
the fidelity of neural processing in cortical circuits during
development and will likely result in delayed or atypical
patterns of specialization (37). However, mild impairment that
reduces the fidelity of processing only slightly may have
significant consequences primarily for brain regions associ-
ated with complex, dynamic, and less predictable stimuli—
features that typically characterize the social world (5).
It has been proposed that regions such as the fusiform face
area, superior temporal sulcus, and temporoparietal junction
become recruited into the cortical social brain network during
development precisely because their patterns of primary
inputs and outputs make them suitable as a “nexus” region
for integrating across sensory channels with high temporal
precision (38). For example, Carter and Huettel (38) argued
that the temporoparietal junction is frequently active in social
cognitive tasks because it is located at the anatomic con-
vergence of several different critical pathways of information
processing, and the overlap and integration of these different
streams lead to novel secondary computations, such as
decision making within a social context. It is not surprising
that evidence shows that these regions can be differentially
impaired in adults with autism (39).
Biological Psychiatry July 15, 2016; 80:94–99 www.sobp.org/journal
Biological
Psychiatry
At the present time, the largest body of work supporting
claims on the nature of the social brain network comes from
studying adults. Much less is understood regarding the infant
brain, but an emerging literature on structural imaging with
infants at risk for autism may begin to address this knowledge
gap. Findings to date have not lent support for specific
differences in the social brain network. For example, one
study reported enlarged total cerebral volumes and excess
extra-axial fluid within the first year of life in infants who later
developed autism. By the second year, the same infants still
had larger total cerebral volumes but typical levels of extra-
axial fluid (40). Another study with an independent cohort
reported that the corpus callosum more specifically may be
enlarged in infants who develop autism within the first year of
life but that differences are no longer apparent by the age of 2
years (41). In contrast, functional analysis in the same cohort
showed that posterior brain network structures, on the whole,
show reduced efficiency, including regions involved in proc-
essing auditory and visual stimuli, language, and nonlinguistic
social stimuli (42).
Methods other than magnetic resonance imaging in infants
at risk lend further support to altered overall network con-
nectivity in the first year of life. First, converging measures of
overall brain connectivity patterns have documented differ-
ences in infants at risk as a group within the first year, albeit
without these differences being examined directly in relation to
a later autism diagnosis. These differences include increased
central and temporal gamma oscillatory activity (43) and
altered trajectory of multiscale entropy derived from electro-
encephalography (44). By the start of the second year, these
differences become characteristic of the emerging phenotype,
where infants who later develop autism show elevated phase-
lagged alpha-frequency electroencephalography connectivity
(45). However, these connectivity differences are not unique to
social brain structures but appear to reflect altered connec-
tivity across large swathes of cortex.
A few behavioral studies have examined how the social
brain networks become deployed for more cognitively
demanding tasks that require high-resolution temporal inte-
gration. One example of this pattern mentioned earlier is the
high degree of modulation of fixations to different face regions
as a function of complex versus simple social scenes;
performance in the complex scene only was associated with
language outcomes (15). Another study found few differences
in face scanning across multiple conditions except when the
actress started speaking (46). The authors suggested that
speech introduces additional task demands, which dispropor-
tionally disrupts attention to the inner features of the face in
the group who later develop autism (46). Infants who later
developed autism in this study also looked overall less at the
scenes as a whole, suggesting that they interacted differently
with the task and raising the possibility that this performance
difference is attributed to multiple social and nonsocial brain
networks.
Finally, despite being indistinguishable in orienting toward,
or scanning of, socially relevant information, studies have
suggested that infants who later develop autism may rely
on alternative different neural mechanisms to achieve this
performance. For example, one study found similar patterns of
visual fixations to dynamic gaze cues (eyes looking toward
97
Biological
Psychiatry
and away from the infant), but an atypical event-related
potential response to the same gaze cues characterized the
subgroup of infants who developed autism as toddlers (33).
Although infants who later developed autism were indistin-
guishable in face-sensitive components correlated with initial
visual response and analysis (P100 and N290), they were
atypical in components reflecting integration of top-down and
bottom-up information (33). Direct measurement of brain
structure and function early in life suggests widespread
atypicality of emerging neural networks in autism that differ-
entially affects brain computations necessary for an under-
standing of the social world.
SOLVING THE FIRST-YEAR PUZZLE: FUTURE
DIRECTIONS
The findings reviewed here provide support for a potential
resolution to the first-year puzzle (1); widespread and diffuse
effects on brain connectivity have partial and mild effects on
multiple systems during the first year, but their effects are only
clearly revealed when more complex multisensory integration
and high temporal resolution processing becomes important
for age-appropriate typical behaviors. Key nexus regions of
the social brain that depend on high temporal resolution
integration may be particularly vulnerable to the mildly
degraded signal processing. From this point, a child may
adapt his or her behavior to his or her own processing
capabilities, by focusing attention and interests on domains
that are more predictable than the increasingly complex social
environment (5).
Three research directions may support further testing of
alternatives to the social-first hypothesis. First, it could be
argued that infant sibling research to date has remained
descriptive of the natural history of prodromal and emerging
autism in infancy. Moving forward, the substantial body of
research on the typical development of the social brain can be
put to use with more targeted hypothesis to clarify underlying
mechanisms. For example, a few studies have directly inves-
tigated early visual (47) and auditory (48) mechanisms in
infants at risk for autism, providing rare evidence for enhance-
ment of processing as potentially underlying later social out-
comes in toddlerhood (47).
Second, a key issue for this alternative perspective will be
to dissociate the later behavioral symptoms of autism that
reflect an ongoing pathology of brain processing from behav-
ioral symptoms that are adaptive responses. For example,
some infant brain structure and function atypicalities correlate
best with later measures of repetitive behaviors (45), suggest-
ing that other behavioral features of autism may reflect later
adaptations as an atypical brain seeks to fit the environmental
niche of human development (5). Analyses of emerging
structural and functional connectivity will be critical for this
endeavor (49). Our review indicates that relying on a model of
focal deficit to find predictors of autism may not be sufficient
to capture the full range of developmental pathways and
outcomes observed in this population.
Finally, another approach to testing key issues about the
social-first hypothesis is to target social or other brain systems
through early intervention before the onset of diagnostic
symptoms (50). For example, one prodromal intervention
98 Biological Psychiatry July 15, 2016; 80:94–99 www.sobp.org/journal
Autism and the Social Brain
has centered on potentially boosting early emerging social
skills (51), and similar approaches to attention and executive
systems are possible (52). With increasing refinement, such
models may begin to target specific developmental mecha-
nisms, highlighting the extent to which they are involved in
later emerging symptoms.
ACKNOWLEDGMENTS AND DISCLOSURES
EM acknowledges salary support from the Bourgeois Foundation and Brain
Canada and the Azrieli Foundation Grant No. BC_Azrieli_MIRI_3388. MHJ
acknowledges funding from the United Kingdom Medical Research Council,
MQ: Transforming Mental Health PsyIMPACT Programme, and Innovative
Medicines Initiative joint undertaking under Agreement No. 11530, resour-
ces of which are composed of financial contributions from the European
Commission Seventh Framework Programme Grant No. FP7/2009‐2013
and European Federation of Pharmaceutical Industries and Associations
companies in kind contribution.
ARTICLE INFORMATION
From the Department of Psychiatry (ME), McGill University, Montreal,
Quebec, Canada; and Centre for Brain and Cognitive Development (MHJ),
Birkbeck, University of London, London, United Kingdom.
Address correspondence to Mayada Elsabbagh, Ph.D., Department of
Psychiatry, Faculty of Medicine, McGill University, 1033 Pine Avenue West,
Montreal, Quebec H3A 1A1, Canada; E-mail: mayada.elsabbagh@mcgill.ca.
Received Mar 21, 2015; revised Feb 5, 2016; accepted Feb 18, 2016.
REFERENCES
1. Elsabbagh M, Johnson MH (2010): Getting answers from babies about
autism. Trends Cogn Sci 14:81–87.
2. Jones EJH, Gliga T, Bedford R, Charman T, Johnson MH (2014):
Developmental pathways to autism: A review of prospective studies of
infants at risk. Neurosci Biobehav Rev 39:1–33.
3. Falck-Ytter T, Bölte S, Gredebäck G (2013): Eye tracking in early
autism research. J Neurodev Disord 5:28.
4. Ozonoff S, Young GS, Carter A, Messinger D, Yirmiya N, Zwaigen-
baum L, et al. (2011): Recurrence risk for autism spectrum disorders: A
Baby Siblings Research Consortium study. Pediatrics 128:e488–e495.
5. Johnson MH, Jones EJH, Gliga T (2015): Brain adaptation and
alternative developmental trajectories. Dev Psychopathol 27:
425–442.
6. Gliga T, Jones EJ, Johnson MH (2014): Low noise in autism: Cause or
consequence? Autism 19:369–370.
7. Chevallier C, Kohls G, Troiani V, Brodkin ES, Schultz RT (2012): The
social motivation theory of autism. Trends Cogn Sci 16:231–239.
8. Mottron L, Dawson M, Soulieres I, Hubert B, Burack J (2006):
Enhanced perceptual functioning in autism: An update, and eight
principles of autistic perception. J Autism Dev Disord 36:27–43.
9. Lawson RP, Rees G, Friston C (2014): An aberrant precision account
of autism. Front Hum Neurosci 8:302.
10. Adolphs R (2009): The social brain: neural basis of social knowledge.
Annu Rev Psychol 60:693–716.
11. Klein JT, Shepherd SV, Platt ML (2009): Social attention and the brain.
Curr Biol 19:R958–R962.
12. Johnson MH (2005): Subcortical face processing. Nat Rev Neurosci 6:
766–774.
13. Farroni T, Menon E, Johnson MH (2006): Factors influencing new-
borns’ preference for faces with eye contact. J Exp Child Psychol 95:
298–308.
14. Johnson MH (2013): The paradox of the emerging social brain. In:
Banaji MR, Gelman SA, editors. Navigating the Social World: What
Infants, Children, and Other Species Can Teach Us. Oxford: Oxford
University Press, 6–10.
15. Elsabbagh M, Bedford R, Senju A, Charman T, Pickles A, Johnson MH
(2013): What you see is what you get: Contextual modulation of face

Autism and the Social Brain
scanning in typical and atypical development. Soc Cogn Affect
Neurosci 9:538–543.
16. Young GS, Merin N, Rogers SJ, Ozonoff S (2009): Gaze behavior and
affect at 6 months: predicting clinical outcomes and language
development in typically developing infants and infants at risk for
autism. Dev Sci 12:798–814.
17. Merin N, Young GS, Ozonoff S, Rogers SJ (2006): Visual fixation
patterns during reciprocal social interaction distinguish a subgroup
of 6-month-old infants at-risk for autism from comparison infants.
J Autism Dev Disord 37:108–121.
18. Schultz RT (2005): Developmental deficits in social perception in
autism: The role of the amygdala and fusiform face area. Int J Dev
Neurosci 23:125–141.
19. Dawson G (2008): Early behavioral intervention, brain plasticity, and
the prevention of autism spectrum disorder. Dev Psychopathol 20:
775–803.
20. Elsabbagh M, Johnson MH (2007): Infancy and autism: Progress,
prospects, and challenges. Prog Brain Res 164:355–383.
21. Elsabbagh M, Gliga T, Pickles A, Hudry K, Charman T, Johnson MH
(2013): The development of face orienting mechanisms in infants at-
risk for autism. Behav Brain Res 251:147–154.
22. Ozonoff S, Iosif A-M, Baguio F, Cook IC, Hill MM, Hutman T, et al.
(2010): A prospective study of the emergence of early behavioral signs
of autism. J Am Acad Child Adolesc Psychiatry 49:256–266; e252.
23. Zwaigenbaum L, Bryson S, Rogers T, Roberts W, Brian J, Szatmari P
(2005): Behavioral manifestations of autism in the first year of life. Int J
Dev Neurosci 23:143–152.
24. Clifford SM, Hudry K, Elsabbagh M, Charman T, Johnson MH (2012):
Temperament in the first 2 years of life in infants at high-risk for autism
spectrum disorders. J Autism Dev Disord 43:673–686.
25. Hudry K, Chandler S, Bedford R, Pasco G, Gliga T, Elsabbagh M, et al.
(2013): Early language profiles in infants at high-risk for autism
spectrum disorders. J Autism Dev Disord 44:154–167.
26. Wan MW, Green J, Elsabbagh M, Johnson M, Charman T, Plummer F
(2012): Quality of interaction between at-risk infants and caregiver at
12-15 months is associated with 3-year autism outcome. J Child
Psychol Psychiatry 54:763–771.
27. Bedford R, Elsabbagh M, Gliga T, Pickles A, Senju A, Charman T,
et al. (2012): Precursors to social and communication difficulties in
infants at-risk for autism: Gaze following and attentional engagement.
J Autism Dev Disord 42:2208–2218.
28. Kuhl PK, Coffey-Corina S, Padden D, Dawson G (2005): Links
between social and linguistic processing of speech in preschool
children with autism: Behavioral and electrophysiological measures.
Dev Sci 8:F1–F12.
29. Jones W, Klin A (2013): Attention to eyes is present but in decline in 2–
6-month-old infants later diagnosed with autism. Nature 504:427–431.
30. Vouloumanos A, Werker JF (2007): Listening to language at birth:
Evidence for a bias for speech in neonates. Dev Sci 10:159–164.
31. Lloyd-Fox S, Blasi A, Elwell CE, Charman T, Murphy D, Johnson MH
(2013): Reduced neural sensitivity to social stimuli in infants at risk for
autism. Proc R Soc Lond B Biol Sci 280:20123026.
32. Szatmari P, Georgiades S, Duku E, Bennett TA, Bryson S, Fombonne
E, et al. (2015): Developmental trajectories of symptom severity and
adaptive functioning in an inception cohort of preschool children with
autism spectrum disorder. JAMA Psychiatry 72:276.
Biological Psychiatry July 15, 2016; 80:94–99 www.sobp.org/journal
Biological
Psychiatry
33. Elsabbagh M, Mercure E, Hudry K, Chandler S, Pasco G, Charman T,
et al. (2012): Infant neural sensitivity to dynamic eye gaze is
associated with later emerging autism. Curr Biol 22:338–342.
34. Flanagan JE, Landa R, Bhat A, Bauman M (2012): Head lag in infants
at risk for autism: A preliminary study. Am J Occup Ther 66:577–585.
35. Paul R, Fuerst Y, Ramsay G, Chawarska K, Klin A (2010): Out of the
mouths of babes: Vocal production in infant siblings of children with
ASD. J Child Psychol Psychiatry 52:588–598.
36. Elsabbagh M, Fernandes J, Jane Webb S, Dawson G, Charman T,
Johnson MH (2013): Disengagement of visual attention in infancy is
associated with emerging autism in toddlerhood. Biol Psychiatry 74:
189–194.
37. Rubenstein JLR, Merzenich MM (2003): Model of autism: Increased ratio of
excitation/inhibition in key neural systems. Genes Brain Behav 2:255–267.
38. Carter RM, Huettel SA (2013): A nexus model of the temporal–parietal
junction. Trends Cogn Sci 17:328–336.
39. Lombardo MV, Chakrabarti B, Bullmore ET, Baron-Cohen S (2011):
Specialization of right temporo-parietal junction for mentalizing and its
relation to social impairments in autism. Neuroimage 56:1832–1838.
40. Shen MD, Nordahl CW, Young GS, Wootton-Gorges SL, Lee A, Liston
SE, et al. (2013): Early brain enlargement and elevated extra-axial fluid in
infants who develop autism spectrum disorder. Brain 136:2825–2835.
41. Wolff JJ, Gerig G, Lewis JD, Soda T, Styner MA, Vachet C, et al.
(2015): Altered corpus callosum morphology associated with autism
over the first 2 years of life. Brain 138:2046–2058.
42. Lewis JD, Evans AC, Pruett JR, Botteron K, Zwaigenbaum L, Estes A,
et al. (2014): Network inefficiencies in autism spectrum disorder at 24
months. Transl Psychiatry 4:e388.
43. Bosl W, Tierney A, Tager-Flusberg H, Nelson C (2011): EEG complex-
ity as a biomarker for autism spectrum disorder risk. BMC Med 9:18.
44. Elsabbagh M, Volein A, Csibra G, Holmboe K, Garwood H, Tucker L,
et al. (2009): Neural correlates of eye gaze processing in the infant
broader autism phenotype. Biol Psychiatry 65:31–38.
45. Orekhova EV, Elsabbagh M, Jones EJH, Dawson G, Charman T,
Johnson MH (2014): EEG hyper-connectivity in high-risk infants is
associated with later autism. J Neurodev Disord 6:40.
46. Shic F, Macari S, Chawarska K (2014): Speech disturbs face scanning
in 6-month-old infants who develop autism spectrum disorder. Biol
Psychiatry 75:231–237.
47. Gliga T, Bedford R, Charman T, Johnson MH; BASIS Team (2015):
Enhanced visual search in infancy predicts emerging autism symp-
toms. Curr Biol 25:1727–1730.
48. Guiraud JA, Kushnerenko E, Tomalski P, Davies K, Ribeiro H, Johnson
MH; BASIS Team (2011): Differential habituation to repeated sounds
in infants at high risk for autism. Neuroreport 22:845–849.
49. Vértes PE, Bullmore ET (2014): Annual research review: Growth connec-
tomics—the organization and reorganization of brain networks during nor-
mal and abnormal development. J Child Psychol Psychiatry 56:299–320.
50. Green J, Dunn G (2008): Using intervention trials in developmental
psychiatry to illuminate basic science. Br J Psychiatry 192:323–325.
51. Green J, Charman T, Pickles A, Wan MW, Elsabbagh M, Slonims V,
et al. (2015): Parent-mediated intervention versus no intervention for
infants at high risk of autism: A parallel, single-blind, randomised trial.
Lancet Psychiatry 2:133–140.
52. Wass S, Porayska-Pomsta K, Johnson MH (2011): Training attentional
control in infancy. Curr Biol 21:1543–1547.
99