Biological Psychiatry: Review CNNI

The in

Aarthi Padmanabhan, Charles J. Lynch, Marie Schaer, and Vinod Menon

ABSTRACT disorder (ASD) is characterized by deficits in social communication and interaction. Since its dis- covery as a major functional brain system, the default mode network (DMN) has been implicated in a number of psychiatric disorders, including ASD. We review converging multimodal evidence for DMN dysfunction in the context of specific components of social cognitive dysfunction in ASD—-referential processing, which is the ability to process social information relative to oneself; and or mentalizing, which is the ability to infer the mental states, such as beliefs, intentions, and emotions, of others. We show that altered functional and structural organization of the DMN and its atypical developmental trajectory are prominent neurobiological features of ASD. We integrate findings on atypical cytoarchitectonic organization and imbalance in excitatory-inhibitory circuits, which alter local and global brain signaling, to scrutinize putative mechanisms underlying DMN dysfunction in ASD. Our synthesis of the extant literature suggests that aberrancies in key nodes of the DMN and their dynamic functional interactions contribute to atypical integration of information about the self in relation to “other” as well as to impairments in the ability to flexibly attend to socially relevant stimuli. We conclude by highlighting open questions for future research. Keywords: Autism, Default mode network, Mentalizing, Self-referential processing, Social, Theory of mind http://dx.doi.org/10.1016/j.bpsc.2017.04.004

AUTISM SPECTRUM DISORDER, SOCIAL DEFICITS, architecture of the DMN, focusing on aspects of social function AND DEFAULT MODE NETWORK that are known to be affected in ASD. These include self- referential processing, which is the ability to process social Autism, derived from the Greek word “auto” meaning “self,” information relative to oneself, and mentalizing or theory of describes a lack of interest in social interactions with the mind, which is the ability to infer the mental states, such as “other.” The term autism was first used by Eugen Bleuler to beliefs, intentions, and emotions, of others. We provide evi- describe adolescents and adults with (1), and it dence for aberrant function of the DMN in ASD as it relates to became the cornerstone for Leo Kanner’s characterization of deficits in these domains of social cognition. We then review infants and young children who showed a lack of interest in functional, structural, cytoarchitectural, and neurophysiological communicating with others and appeared to be “lost in their evidence for neuronal disorganization in key nodes of the DMN own narrow worlds” (2). Although these early descriptions have in ASD. Our synthesis of the extant literature suggests that an been influential in describing the cluster of social impairments altered developmental trajectory of structural and functional now known to characterize autism spectrum disorder (ASD), it organization of the DMN is a prominent neurobiological feature has become increasingly evident that self-related cognitive of ASD. We discuss putative neuronal mechanisms underlying processing is also atypical in affected individuals (3). Under- DMN dysfunction and highlight questions for future research. standing of “self” in the context of “other” is integral to suc- cessful social interactions (4), and it is theorized that individuals with ASD struggle with reciprocal social interaction FUNCTIONAL NEUROANATOMY OF DMN AND ITS largely owing to difficulties in both self-referential cognitive ROLE IN SOCIAL COGNITION processing and inferring the mental states of others (5). Over the past 3 decades, a number of influential studies have Perhaps it is not surprising then that the default mode network consistently demonstrated that a strongly intrinsically inter- (DMN), a core brain system for processing information about connected network of brain structures (11,12), including the the self and other (6,7), has emerged as a key system under- posterior cingulate cortex (PCC), , medial prefrontal lying social dysfunction in ASD (3,6–9). cortex (mPFC), (TPJ), and hippo- Despite the unique functional properties of the DMN and its campus (13,14) (Figure 1), is attenuated during a wide range of links to the ASD phenotype (10), few attempts have been made cognitive tasks (13,15). In parallel, several investigations have to synthesize the extant multimodal literature and provide a also uncovered that these structures, collectively named the unified framework of DMN dysfunction in ASD. We review default mode network, are significantly engaged during tasks converging evidence from multiple scales of brain organization involving social cognitive mental processes that are evaluative that DMN dysfunction is a significant component of social (16,17), including self-referential and autobiographical pro- impairments in ASD. We first describe the functional cessing (13,15,16,18) and mentalizing and theory of mind

ª 2017 Society of Biological Psychiatry. Published by Elsevier Inc. All rights reserved. 1 ISSN: 2451-9022 Biological Psychiatry: Cognitive and - 2017; -:-–- www.sobp.org/BPCNNI Biological Psychiatry: CNNI Default Mode Network in Autism

Figure 1. Functional and structural architecture of the default mode network (DMN) identified using multiple imaging modalities and methods. (A) Architecture of the DMN, identified as regions of “task-induced deactivation” in the seminal meta- analysis by Shulman et al. (135). Data derived from 15 nine studies using [ O]H2O positron emission tomography and reproduced as a surface rendering, as in Buckner et al. (136). (B) DMN topology is readily identifiable using resting-state functional magnetic resonance imaging data and use of an independent component analysis. (C) Core midline DMN nodes, the medial (mPFC) and posterior cingulate cortex (PCC), are structurally connected via a major pathway, the cingulum bundle. Fibers reconstructed using diffu- sion tensor imaging . (D) The strength of structural connections among DMN nodes can be quantified using diffusion imaging. Edge thickness and node size represent connection strength and node degree, respectively. PCC is the most strongly connected node within the DMN. (E) Spring graph illustrates the differing functional connectivity weights between DMN nodes, such that more strongly connected nodes are closer together in space, and these midline hubs are embedded centrally within the network. aMPFC, anteromedial prefrontal cortex; dMPFC, dorsomedial prefrontal cortex; HF1, hippocampal formation; HIP, ; LTC, lateral temporal cortex; pIPL, posterior inferior parietal lobule; PHC, parrahippocampal cortex; Rsp, ; supF, superior frontal gyrus; TempP, temporal pole; TP, temporal pole; TPJ, temporoparietal junction; vMPFC, ventromedial prefrontal cortex. [A, Adapted with permission from Buckner et al. (13); B, adapted with permission from Shirer et al. (137); C, adapted with permission from van den Heuvel et al. (138); D, adapted with permission from Tao et al. (139); E, adapted with permission from Andrews-Hanna et al. (140).]

(13,17). Beyond socially relevant functions, the DMN is also DMN nodes, play distinct and interacting roles in monitoring of associated with mind wandering and initiation of spontaneous both the psychological state of self and evaluation of others. thought processes (13) and subjective value judgments (19), but how these functions contribute to social cognition is not known. Given its central importance to the phenomenology of TASK-BASED FUNCTIONAL MAGNETIC RESONANCE ASD, we focus on processes that are more directly relevant to IMAGING STUDIES OF ATYPICAL DMN FUNCTION social function. Analysis of the extant literature in neurotypical A prominent cognitive deficit of ASD is impairment in the ability individuals reveals considerable overlap between core DMN to decode the mental states of self and others, and DMN nodes and brain regions involved in social cognition dysfunction may be a critical neural signature of these deficits (6,17,20–22), most notably the PCC, mPFC, and TPJ (Figure 2). (31–33). The majority of task-based functional magnetic reso- We next summarize some of the known functional roles of nance imaging (fMRI) studies of ASD have been conducted these three brain regions as relevant for the present review. with adults, or mixed groups of adolescents and adults, with As one of the most highly connected regions in the brain ASD relative to age-matched neurotypical individuals. Task- (23), with a high baseline metabolic rate (14), the PCC is related fMRI studies also focus primarily on activation in spe- considered to be a core functional “hub.” The PCC, which is cific nodes of the social brain, including the DMN regions situated between the marginal ramus of the cingulate sulcus described above. and the parieto-occipital sulcus, has been implicated in both Studies of self-referential processing, requiring self-related self-relevant and other-relevant processing, including tasks versus other-related judgments, demonstrate reduced activa- requiring and imagining oneself in tion in the PCC (34) and mPFC (33). One study comparing the the future (20), and evaluating and processing mental states of neural response in the ventral mPFC to self-related versus others (16,24,25). The mPFC encompasses a collection of other-related judgments, showed preferential activation of this strongly interconnected, contiguous regions in the prefrontal region for self-related judgments in neurotypical control adults, cortex, including the medial superior frontal, orbital, and fron- but not in adults with ASD (33). Analysis of multivariate voxel topolar cortices, and anterior portions of the cingulate cortex. patterns further suggests that in adults with ASD, these midline The mPFC is associated with monitoring of both one’s own structures and the PCC in particular are insensitive to semantic mental states and the mental states of others (16,17), which processing of words that connote social interactions. More- are thought to engage the ventral and dorsal subregions, over, machine learning algorithms classified individuals as respectively (13,26). The TPJ is situated between the inferior autistic or control with 97% accuracy from their fMRI neuro- parietal cortex and posterior superior temporal cortex, with cognitive markers in these regions (35). Together, results prominent overlap with the node of the DMN suggest that the PCC and mPFC exhibit aberrant patterns of (27). The TPJ preferentially encodes other-relevant information, self-representation in ASD. including the mental states and beliefs of others. For example, Theory of mind and mentalizing tasks typically involve transcranial magnetic stimulation to the right TPJ has been viewing images, animation, or stories and test participants’ shown to disrupt a participant’s ability to attribute intentions to understanding of others’ intentions or mental states. Overall, others (28) and the ability to distinguish other-relevant from these studies report decreased recruitment of the TPJ and self-relevant information (29). The TPJ has also been linked dorsal mPFC in adults with ASD (36–41). However, other with predicting behaviors of others during social interactions studies report decreases only in relation to sexual dimorphism (30). Thus, the PCC, dorsal and ventral mPFC, and TPJ, core in ASD, with one study showing that male, but not female,

2 Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI Biological Psychiatry: Default Mode Network in Autism CNNI

Figure 2. The default mode network (DMN) over- laps spatially with regions recruited by component processes of social cognition. The result of a meta- analysis using over 1200 functional magnetic reso- nance imaging studies in the Neurosynth database (141), using the search term “default mode” (shown in red). Regions implicated with four additional search terms, (A) “social,” (B) “mentalizing,” (C) “self-referential,” and (D) “theory of mind,” are shown in blue. The overlap between the DMN and regions recruited by these unique aspects of social cognition are shown in yellow.

adults with ASD demonstrate decreased activity in mPFC and approaches. Despite the strong interconnectivity of the DMN TPJ relative to neurotypical adults (42), while a different study andalargebodyofevidencethatASDisadisorderof suggested the opposite pattern, showing that female, but not network-level brain organization, it is surprising that very male, subjects in the ASD group demonstrate decreased little is known regarding functional connectivity of the DMN activation in the mPFC (43). Few studies have examined during social tasks. As reviewed below, there is more functional connectivity during such tasks and report decreased consistent evidence from intrinsic connectivity studies that interhemispheric connectivity of the TPJ with the left lateral the overall functional organization of the DMN is aberrant temporal cortex (44) and between TPJ and mPFC (45) in adults in ASD. with ASD. There have been far fewer task-based fMRI investigations in children with ASD, likely owing to difficulties with task INTRINSIC FUNCTIONAL CONNECTIVITY EVIDENCE compliance and excessive head movement. When asked to FOR ABERRANT DMN either infer the emotional state of an another individual (other- Intrinsic functional connectivity as measured by resting-state task) or judge their own emotional response (self-task), in- fMRI has been widely used to investigate the functional dividuals with ASD (12–31 years of age) showed increased architecture of the DMN (16). Resting-state fMRI has been dorsal mPFC activation with age, whereas neurotypical control used more extensively in studies of ASD than task-related fMRI subjects showed age-related decreases (46). Further evidence owing to the relative ease of acquiring the data, especially in that group differences may be age dependent comes from a developmental cohorts (50). There is strong evidence that the study that used a similar task and found no significant differ- DMN is among the most disrupted functional networks in ASD ences when children and adolescents with ASD were pooled (10,51). Importantly, a number of studies report that disrupted together (47). Lastly, studies using theory of mind tasks report intrinsic DMN organization is associated with social deficits in reduced activation of dorsal mPFC, TPJ, and PCC in children children and adults with ASD (52–56). (48), while other studies report increased activation in ado- Most intrinsic functional connectivity studies in children with lescents with ASD (49) relative to age-matched control sub- ASD report increased within-network connectivity between jects. More studies are needed to clarify the developmental core DMN nodes, while studies in adolescents and adults profile of functional deficits in ASD. report decreased connectivity, and studies in mixed age The extant literature suggests that social impairments in groups report both increases and decreases (57–71). These ASD are linked to atypical function of the DMN. Despite “inconsistencies” likely reflect developmental changes and considerable heterogeneity with respect to age and sex of heterogeneity in connectivity profiles across different nodes of participants and type of task, findings generally point to the DMN (9). For example, we previously showed that children hypoactivation of DMN nodes in ASD, with decreased with ASD demonstrate strong hyperconnectivity between the recruitment of ventral mPFC during processing of self- PCC and lateral and medial but hypo- relevant information and dorsal mPFC and TPJ during pro- connectivity within DMN subregions (65) (Figure 3A). Further- cessing of other-relevant information. A significant limitation more, PCC hyperconnectivity also predicted social is the lack of developmental studies and connectivity-based communication deficits in children with ASD (Figure 3B) (65).

Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI 3 Biological Psychiatry: CNNI Default Mode Network in Autism

Figure 3. Hyperconnectivity with the posterior cingulate cortex (PCC) predicts social communication deficits in children with autism spectrum disorder (ASD). (A) Children with ASD demonstrate whole-brain hyperconnectivity with PCC and retrosplenial cortex (RSC) and hypoconnectivity with precuneus (PreC). *p , .01, **p , .005, ***p , .001. (B) Hyperconnectivity between the PCC and target regions, including the right , left temporal pole, and right lingual gyrus, predicted social impairments as measured by the Autism Diagnostic Observation Schedule (ADOS) social subscale in children withASD.No such significant relationships were demonstrated by the RSC and PreC. *Significant. aLTC, anterolateral temporal cortex; DMPFC, dorsomedial prefrontal cortex; ERc, entorhinal cortex; L, left; LG, lingual gyrus; n.s., not significant; PHG, parahippocampal gyrus; pInsula, posterior insular cortex; PRc, perirhinal cortex; pSTS, posterior superior temporal sulcus; R, right; ROI, region of interest; TD, typical development; TempP, temporal pole. [Adapted with permission from Lynch et al. (65).]

Over the course of typical development, intrinsic functional connectivity also increases with age in neurotypical in- connectivity between DMN nodes increases between child- dividuals (Figure 4C) (76), but not in individuals with ASD hood and adulthood (Figure 4A, B) (72,73). In ASD, there ap- (61,63), suggesting that the development of critical intranet- pears to be no consistent evidence of such developmental work and internetwork connections is altered in ASD. increases in DMN connectivity (63,74,75). This suggests that The extant literature provides converging evidence for hyperconnected DMN links in childhood ASD may become aberrant intrinsic DMN organization and development in ASD. underconnected in adulthood owing to failure to strengthen Overall, the most commonly observed profile in childhood ASD long-range pathways with age in ASD. Finally, cross-network has three features: 1) increased within-network connectivity in

Figure 4. The default mode network transitions from an immature state in childhood to a cohesive network in adulthood. (A) Independent component analysis applied to resting-state functional magnetic resonance imaging data reveals stronger medial prefrontal cortex functional connectivity in a group of adults relative to a group of children 7–9 years of age. (B) Comparison of medial prefrontal cortex functional connectivity strength in children (left) and adults (middle) confirms greater connectivity, espe- cially with posterior cingulate cortex, later in devel- opment (right). (C) Connectivity between the default mode network and other systems of the brain change over development. Spring graph represen- tation of brain network development using resting- state functional magnetic resonance imaging during three periods of development: 9 years (left), 13 years (middle), and 25 years (right). Note that some default mode network nodes (filled red) are isolated from one another in childhood but are more strongly integrated in adulthood. [A, Adapted with permission from Supekar et al. (72); B, adapted with permission from Fair et al. (73); C, adapted with permission from Power et al. (76).]

4 Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI Biological Psychiatry: Default Mode Network in Autism CNNI

the DMN, most notably between the PCC and mPFC; adolescents, and adults. Using a metric of cortical folding, or 2) reduced connectivity of DMN nodes with other functional gyrification, we recently demonstrated that in a mixed age systems outside the DMN; and 3) increased local connectivity group of individuals with ASD, male, but not female, subjects within DMN nodes. These patterns appear to take on a more with ASD showed reduced gyrification in the ventral mPFC complex profile with age involving an overall developmental relative to neurotypical control subjects, suggesting a sex- shift from hyperconnectivity in childhood to hypoconnectivity specific locus of vulnerability (Figure 5D) (80). There is also in adolescence and adulthood. Such altered connectivity evidence for shallower sulci in the TPJ of children with ASD between DMN nodes likely underlies disturbances in self- (Figure 5A) (81) and reduced gray matter volume of the right referential and mentalizing processes, whereas reduced TPJ in adults with ASD, which predicted theory of mind deficits intrinsic connectivity between DMN and other functional sys- (Figure 5B) (82). tems may reflect a difficulty in adaptively switching between Studies measuring correlated changes in interregional the DMN and networks involved in monitoring and attending to cortical thickness (83) as a metric of structural network salient stimuli (57). coherence report increases between the PCC and the As reviewed above, there is now considerable evidence for adjoining retrosplenial cortex and the TPJ and/or angular gyrus aberrant functional organization of the DMN in ASD, with the in children with ASD (84). There is also evidence of decreased strongest and most consistent findings to date emerging from covariance between the PCC and the ventromedial PFC in analysis of intrinsic functional connectivity. Importantly, the ASD (79,84). One study reported that both structural coher- converging evidence points to a complex and dynamic ence and functional connectivity between the right TPJ and the neurobiological phenotype of ASD, with both hyper- PFC is reduced in adults with ASD (85). Future studies should connectivity and hypoconnectivity profiles that shift with use multimodal approaches to better characterize links be- development. tween structure and function. Developmental studies report atypical age-related structural change of DMN nodes in ASD, including accelerated thinning STRUCTURAL MRI EVIDENCE FOR ATYPICAL DMN in bilateral PCC between childhood and adulthood (7–39 years IN ASD of age), which correlated with social deficits (86), and decel- Prominent structural aberrations in the DMN across several erated volume reduction with age (7–29 years of age) in the morphological metrics have been reported in ASD, including ventral mPFC and the TPJ (87). Future studies should reconcile cortical thickness and volume and gray matter density. One differences between volumetric versus cortical thickness rates study using voxel-based morphometry reported altered PCC of change, investigate why some regions are impacted differ- gray matter organization in children and adolescents with ASD, ently than others, and characterize the developmental trajec- which was associated with symptom severity (77). Further- tory of these structural changes. A major limitation of this body more, there is also evidence of increased cortical thickness in of literature is that with a few notable exceptions (77,86), most the PCC (78) and ventral mPFC (79) (Figure 5C) in children, studies do not show direct links with behavioral deficits, which

Figure 5. Structural, neurochemical, and cytoarchitectonic disorganization of key default mode network nodes in autism spectrum disorder (ASD). (A) Right temporoparietal junction (TPJ) sul- cus is shallower in children with ASD (red line) compared with neurotypical children (blue line). (B) Relationship between TPJ gray matter volume and ability to assess interactions between two ob- jects in a social motion experiment in adults with ASD (red) relative to neurotypical adults (blue). Note that the relationship between gray matter volume in ASD is significant, such that greater gray matter volume is associated with better performance. (C) Increased cortical thickness of the bilateral medial prefrontal cortex (mPFC) in children with ASD rela- tive to adults and neurotypical children is replicable across sites of the ABIDE data set. (D) Sex differ- ences in gyrification of the mPFC in ASD. Male (M) subjects with ASD have reduced gyrification whereas female (F) subjects have increased gyr- ification relative to neurotypical control subjects. (E) Abnormal laminar patterning in postmortem ASD posterior cingulate cortex (PCC) brain tissue (right) relative to a healthy case (left). Note the general disorganization and poor differentiation between layers IV and V. (F) Reduced 3H-muscimol labeled gamma-aminobutyric acid type A (GABAA) receptor binding density in the PCC in ASD (right) and neurotypical (left) postmortem brain tissue; darker colors indicate greater receptor density. lGI, local gyrification index; NYU, New York University; PITT, University of Pittsburgh; TD, typical development; USM, Utah School of Medicine. [A, Adapted with permission from Dierker et al. (81); B, adapted with permission from David et al. (82); C, adapted with permission from Valk et al. (79); D, adapted with permission from Schaer et al. (80); E, adapted from Oblak et al. (108); F, adapted with permission from Oblak et al. (122).]

Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI 5 Biological Psychiatry: CNNI Default Mode Network in Autism

future studies should incorporate. Nevertheless, together, MRI NEUROPHYSIOLOGICAL BASIS OF DMN studies point to structural abnormalities of the DMN in ASD DYSFUNCTION IN ASD that appear to persist throughout development. A prominent neurobiological theory postulates that ASD is characterized by an excitation/inhibition (E/I) imbalance in ATYPICAL WHITE MATTER PATHWAYS ASSOCIATED local neural circuits that subserve sensory, social, and WITH DMN IN ASD affective processes (109–113). E/I imbalance in ASD is A number of diffusion tensor imaging studies report atypical thought to alter local and global brain signaling and white matter connectivity in ASD (88). White matter tracts contribute to atypical fluctuations in regional fMRI signals along the cingulum bundle connect the mPFC and PCC (67,114), leading to difficulties with modulating flexible and (Figure 1C), and the majority of studies report decreased goal-directed behaviors. Thus, the DMN has emerged as a fractional anisotropy, a metric of white matter integrity, in the key target for E/I investigations using optogenetic studies in cingulum in children and adolescents with ASD (89–93), rodents, and there is evidence that elevated E/I balance in although some studies report increases (94,95). Tracts con- the rodent mPFC is associated with impaired social func- necting the PCC and TPJ have been challenging to measure tion, an effect that was improved by increasing inhibitory with conventional diffusion tensor imaging owing to crossing function (115). fibers in this region (96); however, some studies have identified One mechanism for E/I imbalance is atypical expression of reduced fractional anisotropy in tracts adjacent to the TPJ in inhibitory interneurons, modulated by the neurotransmitter children and adolescents with ASD (91,97). Over the course of gamma-aminobutyric acid (GABA). During early embryonic typical development, there is evidence of increases in frac- development, GABAergic interneurons are involved with tional anisotropy within the cingulum from childhood to regulating cell migration, differentiation, and synapse formation adulthood (98,99),reflecting a strengthening of white matter (116). Rodent models suggest that decreasing GABA trans- tracts between the PCC and mPFC over development. mission in the mPFC decreases sociability (117), consistent Evidence from cross-sectional developmental studies suggest with our view that E/I imbalance in the DMN can contribute to that decreased fractional anisotropy in the cingulum in ASD impaired social function in autism. may be most prominent in childhood (100,101), with a decel- There is evidence of altered signaling in inhibitory path- erated rate of fractional anisotropy increases with age in young ways that are modulated by GABAergic interneurons in ASD children with ASD (102) as well as mixed-age groups of chil- (118–120). Significant reductions in GABA expression have dren and adults with ASD (103,104) compared with age- been found using magnetic resonance spectroscopy (121) as matched control subjects. well as in postmortem brain tissue of individuals with ASD The extant literature points to decreased white matter (Figure 5F) (122). The colocalized pattern of deficits in cell integrity in tracts linking DMN structures in ASD, most notably migration and GABA signaling in the cortex of individuals within the cingulum bundle. Furthermore, the maturation of with ASD suggests that these deficits occur early in cortical these tracts from childhood to adulthood may be atypical in development, potentially leading to persistent E/I imbalances ASD. How immature development of DMN tracts contributes throughout development (123–125) and altered brain con- to impairments in their functional connectivity and conse- nectivity (9,126,127). Crucially, these types of developmental quently leads to social deficits in ASD is currently not known aberrancies in local brain circuits are likely to influence and is a critical avenue for future research. global circuit function, especially if they occur in widely connected hubs such as the PCC (67,112). Studies using ATYPICAL CELLULAR ORGANIZATION OF DMN IN magnetic resonance spectroscopy in combination with ASD: EVIDENCE FROM POSTMORTEM STUDIES resting-state and task-based fMRI are needed to probe GABA-related and E/I imbalance–related DMN dysfunction Although MRI studies provide critical information regarding in ASD. aberrant brain function and structure at macroscopic levels, Research on the effects of GABA on social function have led histological analysis of human postmortem brain tissue is to the development of specific drugs that potentiate GABA necessary to elucidate the cellular processes that may go awry transmission as treatment for ASD symptoms (128). For in ASD (105). Neuronal migration deficits in early brain devel- example, the GABA agonist arbaclofen has been used to opment may be a critical component of the pathophysiology of B improve social function in patients with fragile X syndrome, a ASD (106,107). Recent postmortem studies of brain tissue genetic condition that often manifests with ASD symptoms demonstrate altered laminar patterns and increased density of (129). In a mouse model of autism, administration of benzo- white matter neurons in superficial layers of PCC, but not in the diazepines, which enhances GABAergic signaling, improved fusiform gyrus (Figure 5E) (108). This selective disruption sug- social and cognitive deficits (130). Whether altering DMN gests that neuronal migration from the germinal zone to the function using locally delivered GABA agonists such as cortical plate, which occurs between 16 and 20 weeks of B arbaclofen can improve social function is an important topic for gestation, is affected in ASD (108). Additional studies with future research. larger samples and in other brain regions are required to confirm whether cell migration deficits in ASD are specific to the PCC. If confirmed, it would provide strong evidence that early DMN DYSFUNCTION IN THE CONTEXT OF SALIENCE insults to the cellular organization of the PCC adversely impact NETWORK ABERRATIONS IN ASD brain development and contribute to the ASD phenotype by Complex social behaviors involve cognitive and perceptual virtue of focused and early disruption to a core brain hub. processes that are supported by interactions between

6 Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI Biological Psychiatry: Default Mode Network in Autism CNNI

large-scale brain systems, including the DMN (131). Notably, substantial evidence that structural and functional disruptions the DMN interfaces with two other major networks in the brain, to key nodes of the DMN, their connectivity with each other, the (SN) and the central executive network and atypical patterns of connectivity with other brain regions (131). Thus, the consequences of DMN dysfunction likely play an important role in the symptomatology of ASD. Our manifest also in interactions of the DMN with other brain synthesis of extant findings suggests that atypical integration systems (10,131). of information about self and other in the DMN underlies A triple network model of psychopathology (131) posits that social deficits in ASD. atypical interactions with the SN might contribute to DMN Research on how the DMN changes from childhood to dysfunction in ASD and contribute to a lack of engagement adulthood is providing fundamental new insights not only into with socially relevant stimuli. The SN is a large-scale neuro- the ontogeny of complex social functions in typically devel- cognitive network anchored in the anterior insula and anterior oping individuals, but also into components of social cognition cingulate cortex, and it plays a critical role in detecting salient that can go awry in ASD. It is clear that research on the DMN stimuli and orienting attentional resources to them in an and its dysfunctional organization have laid the groundwork for adaptive and goal-relevant manner (132). As such, reduced a more sophisticated and principled approach to investigations activation of the PCC and mPFC of the DMN may arise from of the complex, but interrelated, social and behavioral prob- weak mapping of salient socially relevant cues with self- lems that define autism. Despite an explosive amount of relevance and subjective value necessary for effective social research on individual brain regions and their interconnectivity, function. Consistent with this view, recent evidence suggests progress in our understanding of how the brain interactively that children with ASD exhibit intrinsic functional hyper- processes socially relevant stimuli remains a major challenge connectivity within the SN, central executive network, and for the neuroscience of ASD. Analysis of brain circuit dynamics DMN (66) and other neural systems (67). This within-network offers new possibilities in this regard. The ability to flexibly intrinsic hyperconnectivity in ASD can result in “network switch between different modes of thought and reference isolation,” limiting dynamic interactions between brain net- frames is a critical feature of adaptive social function, and works that are necessary for complex social behaviors impairments in complex social processes in ASD are likely (57,133). Consistent with this view, a recent study demon- dependent on interactions between multiple large-scale brain strated that intrinsic connectivity between the DMN, SN, and systems. Progress in the field will depend on better charac- central executive network predicted longitudinal improvements terization of DMN dysfunction in ASD and aberrant dynamic in adaptive behaviors in ASD (134), highlighting the signifi- context-specific interactions with other brain networks, such cance of internetwork interactions in the pathophysiology of as the salience network. Box 1 lists some questions for future ASD (131). research. The findings reviewed here highlight the necessity for studying heterogeneous patterns of functional and structural CONCLUSIONS brain connectivity and their interrelationships at different We have sought to provide a comprehensive review of evi- stages of development. Longitudinal studies leveraging dence pointing to aberrant DMN function in the context of quantitative multimodal systems neuroscience approaches will mental processes that contribute to social deficits in ASD. play a crucial role moving forward. This includes novel These processes most prominently include mentalizing, the- computational tools, individualized atlases, wiring diagrams, ory of mind, and self-referential processing, which have and methodological developments that will propel systematic consistently been shown to activate the core nodes of the ways of thinking about atypical brain organization in ASD. DMN in neurotypical individuals. We have linked neural Knowledge of how functional interactions between the DMN models of DMN function with specific phenotypic features of and other brain systems mature over time and how they social dysfunction in ASD. We used this framing to delineate change mental representations of socially relevant information how DMN function is disrupted in children, adolescents, and will be fundamental to the study of the neurobiological basis of adults with ASD. As we have reviewed, there is now social interaction deficits in ASD.

Box 1. Questions for Future Research When do structural aberrations of the DMN emerge during fetal development and infancy? What is the relationship between structural and functional aberrancies in the DMN, and how do they change with age? How does E/I imbalance impact DMN circuits and function in ASD? How does DMN dysfunction impact social information processing in children with ASD? How do intrinsic hyperconnectivity and hypoconnectivity of specific DMN pathways impact aberrant functional responses during social information processing tasks? What is the relationship between heterogeneity in the clinical profile of ASD and heterogeneity in DMN function? How does DMN organization differ in individuals with low-functioning ASD compared with individuals with high-functioning ASD? Are there sex differences in DMN function, and could these sex differences explain the sex differences in prevalence rates and phenotypic presentation of the disorder? How does impaired judgment of social stimulus value and social distance contribute to social cognitive dysfunction in ASD?

ASD, autism spectrum disorder; DMN, default mode network; E/I, excitation/inhibition.

Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI 7 Biological Psychiatry: CNNI Default Mode Network in Autism

ACKNOWLEDGMENTS AND DISCLOSURES 16. Gusnard DA, Raichle ME, Raichle ME (2001): Searching for a base- line: Functional imaging and the resting . Nat Rev This work was supported by the National Institutes of Health (Grant Nos. Neurosci 2:685–694. HD059205, MH084164, and NS086085 to VM), Singer Foundation (to VM), 17. Schilbach L, Eickhoff SB, Rotarska-Jagiela A, Fink GR, Vogeley K Simons Foundation (to VM), Stanford Child Health Research Initiative and (2008): Minds at rest? Social cognition as the default mode of the Autism Science Foundation Post Doctoral Fellowships (to AP), and cognizing and its putative relationship to the “default system” of the Swiss National Science Foundation Fellowship Grant No. 158831 (to MS). brain. Conscious Cogn 17:457–467. The authors report no biomedical financial interests or potential conflicts 18. Andrews-Hanna JR, Reidler JS, Huang C, Buckner RL (2010): Evi- of interest. dence for the default network’s role in spontaneous cognition. J Neurophysiol 104:322–335. 19. Acikalin MY, Gorgolewski KJ, Poldrack RA (2017): A coordinate- ARTICLE INFORMATION based meta-analysis of overlaps in regional specialization and From the Department of Psychiatry and Behavioral Sciences (AP, VM), functional connectivity across subjective value and default mode Program in Neuroscience (VM), and Department of and Neuro- networks. Front Neurosci 11:1. logical Sciences (VM), Stanford University School of Medicine, Stanford, 20. Spreng RN, Mar RA, Kim AS (2009): The common neural basis of California; Department of (CJL), Georgetown University, autobiographical memory, prospection, navigation, theory of mind, Washington, DC; and Department of Psychiatry (MS), University of Geneva, and the default mode: A quantitative meta-analysis. J Cogn Neurosci Geneva, Switzerland. 21:489–510. AP and CJL contributed equally to this work. 21. Laird AR, Fox PM, Eickhoff SB, Turner JA, Ray KL, McKay DR, et al. Address correspondence to Vinod Menon, Ph.D., 401 Quarry Road, (2011): Behavioral interpretations of intrinsic connectivity networks. Department of Psychiatry and Behavioral Sciences, Stanford University, J Cogn Neurosci 23:4022–4037. Stanford, CA 94304-5719; E-mail: [email protected]. 22. Smith SM, Fox PT, Miller KL, Glahn DC, Fox PM, Mackay CE, et al. Received Dec 1, 2016; revised and accepted Apr 7, 2017. (2009): Correspondence of the brain’s functional architecture during activation and rest. Proc Natl Acad Sci U S A 106:13040–13045. 23. Leech R, Sharp DJ (2014): The role of the posterior cingulate cortex in cognition and disease. Brain 137(pt 1):12–32. REFERENCES 24. Schiller D, Freeman JB, Mitchell JP, Uleman JS, Phelps EA (2009): 1. Bleuler E (1951): Autistic thinking. In: Organization and Pathology of A neural mechanism of first impressions. Nat Neurosci 12:508–514. Thought: Selected Sources. New York: Columbia University Press, 25. Kuzmanovic B, Bente G, von Cramon DY, Schilbach L, 399–437. Tittgemeyer M, Vogeley K (2012): Imaging first impressions: distinct 2. Kanner L (1943): Autistic disturbances of affective contact. Nervous neural processing of verbal and nonverbal social information. Neu- Child 2:217–250. roimage 60:179–188. 3. Uddin LQ (2011): The self in autism: An emerging view from neuro- 26. Amodio DM, Frith CD (2006): Meeting of minds: The medial frontal imaging. Neurocase 17:201–208. cortex and social cognition. Nat Rev Neurosci 7:268–277. 4. Banaji MR, Prentice DA (1994): The self in social contexts. Annu Rev 27. Uddin LQ, Supekar K, Amin H, Rykhlevskaia E, Nguyen DA, Psychol 45:297–332. Greicius MD, et al. (2010): Dissociable connectivity within human 5. Baron-Cohen S, Leslie AM, Frith U (1985): Does the autistic child angular gyrus and intraparietal sulcus: Evidence from functional and have a “theory of mind”? Cognition 21:37–46. structural connectivity. Cerebral Cortex 20:2636–2646. 6. Mars RB, Neubert FX, Noonan MP, Sallet J, Toni I, Rushworth MF 28. Costa A, Torriero S, Oliveri M, Caltagirone C (2008): Prefrontal and (2012): On the relationship between the “default mode network” and temporo-parietal involvement in taking others’ perspective: TMS the “social brain”. Front Hum Neurosci 6:189. evidence. Behav Neurol 19:71–74. 7. Molnar-Szakacs I, Uddin LQ (2013): Self-processing and the default 29. Heinisch C, Kruger MC, Brune M (2012): Repetitive transcranial mode network: Interactions with the mirror neuron system. Front magnetic stimulation over the temporoparietal junction influences Hum Neurosci 7:571. distinction of self from famous but not unfamiliar others. Behav 8. Li W, Mai X, Liu C (2014): The default mode network and social un- Neurosci 126:792–796. derstanding of others: What do brain connectivity studies tell us. 30. Carter RM, et al. (2012): A distinct role of the temporal-parietal Front Hum Neurosci 8:74. junction in predicting socially guided decisions. Science 337: 9. Uddin LQ, Supekar K, Menon V (2013): Reconceptualizing functional 109–111. brain connectivity in autism from a developmental perspective. Front 31. Happe F, Cook JL, Bird G (2017): The structure of social cognition: Hum Neurosci 7:458. In(ter)dependence of sociocognitive processes. Annu Rev Psychol 10. Moseley RL, Ypma RJ, Holt RJ, Floris D, Chura LR, Spencer MD, 68:243–267. et al. (2015): Whole-brain functional hypoconnectivity as an endo- 32. Murdaugh DL, Shinkareva SV, Deshpande HR, Wang J, Pennick MR, phenotype of autism in adolescents. Neuroimage Clin 9:140–152. Kana RK (2012): Differential deactivation during mentalizing and 11. Hagmann P, Cammoun L, Gigandet X, Meuli R, Honey CJ, classification of autism based on default mode network connectivity. Wedeen VJ, et al. (2008): Mapping the structural core of human PloS One 7:e50064. cerebral cortex. PLoS Biol 6:e159. 33. Lombardo MV, Chakrabarti B, Bullmore ET, Sadek SA, Pasco G, 12. Buckner RL, Sepulcre J, Talukdar T, Krienen FM, Liu H, Hedden T, Wheelwright SJ, et al. (2010): Atypical neural self-representation in et al. (2009): Cortical hubs revealed by intrinsic functional connec- autism. Brain 133(pt 2):611–624. tivity: Mapping, assessment of stability, and relation to Alzheimer’s 34. Morita T, Kosaka H, Saito DN, Ishitobi M, Munesue T, Itakura S, et al. disease. J Neurosci 29:1860–1873. (2012): Emotional responses associated with self-face processing in 13. Buckner RL, Andrews-Hanna JR, Schacter DL (2008): The brain’s individuals with autism spectrum disorders: An fMRI study. Soc default network: Anatomy, function, and relevance to disease. Ann N Neurosci 7:223–239. Y Acad Sci 1124:1–38. 35. Just MA, Cherkassky VL, Buchweitz A, Keller TA, Mitchell TM (2014): 14. Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Identifying autism from neural representations of social interactions: Shulman GL (2001): A default mode of brain function. Proc Natl Acad Neurocognitive markers of autism. PLoS One 9:e113879. Sci U S A 98:676–682. 36. Murdaugh DL, Nadendla KD, Kana RK (2014): Differential role of 15. Greicius MD, Menon V (2004): Default-mode activity during a passive temporoparietal junction and medial prefrontal cortex in causal sensory task: Uncoupled from deactivation but impacting activation. inference in autism: An independent component analysis. Neurosci J Cogn Neurosci 16:1484–1492. Lett 568:50–55.

8 Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI Biological Psychiatry: Default Mode Network in Autism CNNI

37. Aoki Y, Yahata N, Watanabe T, Takano Y, Kawakubo Y, Kuwabara H, 56. Lynch CJ, Uddin LQ, Supekar K, Khouzam A, Phillips J, Menon V et al. (2014): Oxytocin improves behavioural and neural deficits in (2013): Default mode network in childhood autism: Posteromedial inferring others’ social emotions in autism. Brain 137(pt 11): cortex heterogeneity and relationship with social deficits. Biol Psy- 3073–3086. chiatry 74:212–219. 38. Kana RK, Patriquin MA, Black BS, Channell MM, Wicker B (2016): 57. Kennedy DP, Courchesne E (2008): The intrinsic functional organi- Altered medial frontal and superior temporal response to implicit zation of the brain is altered in autism. Neuroimage 39:1877–1885. processing of emotions in autism. Autism Res 9:55–66. 58. Cherkassky VL, Kana RK, Keller TA, Just MA (2006): Functional 39. von dem Hagen EA, Stoyanova RS, Rowe JB, Baron-Cohen S, connectivity in a baseline resting-state network in autism. Neuro- Calder AJ (2014): Direct gaze elicits atypical activation of the theory- report 17:1687–1690. of-mind network in autism spectrum conditions. Cerebral Cortex 59. Monk CS, Peltier SJ, Wiggins JL, Weng SJ, Carrasco M, Risi S, et al. 24:1485–1492. (2009): Abnormalities of intrinsic functional connectivity in autism 40. Assaf M, Hyatt CJ, Wong CG, Johnson MR, Schultz RT, Hendler T, spectrum disorders. Neuroimage 47:764–772. et al. (2013): Mentalizing and motivation neural function during social 60. Weng SJ, Wiggins JL, Peltier SJ, Carrasco M, Risi S, Lord C, et al. interactions in autism spectrum disorders. Neuroimage Clin 3: (2010): Alterations of resting state functional connectivity in the 321–331. default network in adolescents with autism spectrum disorders. Brain 41. Pantelis PC, Byrge L, Tyszka JM, Adolphs R, Kennedy DP (2015): Res 1313:202–214. A specific hypoactivation of right temporo-parietal junction/posterior 61. Wiggins JL, Peltier SJ, Ashinoff S, Weng SJ, Carrasco M, Welsh RC, superior temporal sulcus in response to socially awkward situations et al. (2011): Using a self-organizing map algorithm to detect age- in autism. Soc Cogn Affect Neurosci 10:1348–1356. related changes in functional connectivity during rest in autism 42. Kirkovski M, Enticott PG, Hughes ME, Rossell SL, Fitzgerald PB spectrum disorders. Brain Res 1380:187–197. (2016): Atypical neural activity in males but not females with autism 62. Assaf M, Jagannathan K, Calhoun VD, Miller L, Stevens MC, Sahl R, spectrum disorder. J Autism Dev Disord 46:954–963. et al. (2010): Abnormal functional connectivity of default mode 43. Holt RJ, Chura LR, Lai MC, Suckling J, von dem Hagen E, Calder AJ, sub-networks in autism spectrum disorder patients. Neuroimage et al. (2014): “Reading the mind in the eyes”: An fMRI study of ado- 53:247–256. lescents with autism and their siblings. Psychol Med 44:3215–3227. 63. Washington SD, Gordon EM, Brar J, Warburton S, Sawyer AT, 44. Mason RA, Williams DL, Kana RK, Minshew N, Just MA (2008): Wolfe A, et al. (2014): Dysmaturation of the default mode network in Theory of Mind disruption and recruitment of the right hemisphere autism. Hum Brain Mapp 35:1284–1296. during narrative comprehension in autism. Neuropsychologia 64. Jung M, Kosaka H, Saito DN, Ishitobi M, Morita T, Inohara K, et al. 46:269–280. (2014): Default mode network in young male adults with autism 45. Kana RK, Keller TA, Cherkassky VL, Minshew NJ, Just MA (2009): spectrum disorder: relationship with autism spectrum traits. Mol Atypical frontal-posterior synchronization of Theory of Mind regions Autism 5:35. in autism during mental state attribution. Soc Neurosci 4:135–152. 65. Lynch CJ, Uddin LQ, Supekar K, Khouzam A, Phillips J, Menon V 46. Schulte-Ruther M, Greimel E, Piefke M, Kamp-Becker I, (2013): Default mode network in childhood autism: Posteromedial Remschmidt H, Fink GR, et al. (2014): Age-dependent changes in the cortex heterogeneity and relationship with social deficits. Biol Psy- neural substrates of empathy in autism spectrum disorder. Soc Cogn chiatry 74:212–219. Affect Neurosci 9:1118–1126. 66. Uddin LQ, Supekar K, Lynch CJ, Khouzam A, Phillips J, Feinstein C, 47. Vander Wyk BC, Hoffman F, Pelphrey KA (2014): Equivalent neural et al. (2013): Salience network-based classification and prediction of responses in children and adolescents with and without autism symptom severity in children with autism. JAMA Psychiatry 70: during judgments of affect. Dev Cogn Neurosci 8:121–130. 869–879. 48. Kana RK, Maximo JO, Williams DL, Keller TA, Schipul SE, 67. Supekar K, Uddin LQ, Khouzam A, Phillips J, Gaillard WD, Cherkassky VL, et al. (2015): Aberrant functioning of the theory-of- Kenworthy LE, et al. (2013): Brain hyperconnectivity in children with mind network in children and adolescents with autism. Mol Autism autism and its links to social deficits. Cell Rep 5:738–747. 6:59. 68. Funakoshi Y, Harada M, Otsuka H, Mori K, Ito H, Iwanaga T (2016): 49. White SJ, Frith U, Rellecke J, Al-Noor Z, Gilbert SJ (2014): Autistic Default mode network abnormalities in children with autism spectrum adolescents show atypical activation of the brain’s mentalizing sys- disorder detected by resting-state functional magnetic resonance tem even without a prior history of mentalizing problems. Neuro- imaging. J Med Invest 63:204–208. psychologia 56:17–25. 69. Fishman I, Keown CL, Lincoln AJ, Pineda JA, Müller RA (2014): 50. Hull JV, Jacokes ZJ, Torgerson CM, Irimia A, Van Horn JD (2017): Atypical cross talk between mentalizing and mirror neuron networks Resting-state functional connectivity in autism spectrum disorders: A in autism spectrum disorder. JAMA Psychiatry 71:751–760. review. Front Psychiatry 7:205. 70. Cheng W, Rolls ET, Gu H, Zhang J, Feng J (2015): Autism: 51. Glerean E, Pan RK, Salmi J, Kujala R, Lahnakoski JM, Roine U, et al. Reduced connectivity between cortical areas involved in face (2016): Reorganization of functionally connected brain subnetworks expression, theory of mind, and the sense of self. Brain 138(pt 5): in high-functioning autism. Hum Brain Mapp 37:1066–1079. 1382–1393. 52. Yerys BE, Gordon EM, Abrams DN, Satterthwaite TD, Weinblatt R, 71. Olivito G, Clausi S, Laghi F, Tedesco AM, Baiocco R, Jankowski KF, et al. (2015): Default mode network segregation and Mastropasqua C, et al. (2017): Resting-state functional connectivity social deficits in autism spectrum disorder: Evidence from non- changes between dentate nucleus and cortical social brain regions in medicated children. Neuroimage Clin 9:223–232. autism spectrum disorders. Cerebellum 16:283–292. 53. Chen H, Duan X, Liu F, Lu F, Ma X, Zhang Y, et al. (2016): Multivariate 72. Supekar K, Uddin LQ, Prater K, Amin H, Greicius MD, Menon V classification of autism spectrum disorder using frequency-specific (2010): Development of functional and structural connectivity within resting-state functional connectivity-A multi-center study. Prog the default mode network in young children. Neuroimage 52: Neuropsychopharmacol Biol Psychiatry 64:1–9. 290–301. 54. Nielsen JA, Zielinski BA, Fletcher PT, Alexander AL, Lange N, 73. Fair DA, Cohen AL, Dosenbach NU, Church JA, Miezin FM, Bigler ED, et al. (2013): Multisite functional connectivity MRI classi- Barch DM, et al. (2008): The maturing architecture of the brain’s fication of autism: ABIDE results. Front Hum Neurosci 7:599. default network. Proc Natl Acad Sci U S A 105:4028–4032. 55. Ypma RJ, Moseley RL, Holt RJ, Rughooputh N, Floris DL, Chura LR, 74. Wiggins JL, Peltier SJ, Bedoyan JK, Carrasco M, Welsh RC, et al. (2016): Default mode hypoconnectivity underlies a sex-related Martin DM, et al. (2012): The impact of serotonin transporter geno- autism spectrum. Biol Psychiatry Cogn Neurosci Neuroimaging type on default network connectivity in children and adolescents with 1:364–371. autism spectrum disorders. Neuroimage Clin 2:17–24.

Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI 9 Biological Psychiatry: CNNI Default Mode Network in Autism

75. Doyle-Thomas KA, Lee W, Foster NE, Tryfon A, Ouimet T, Hyde KL, 96. Greicius MD, Supekar K, Menon V, Dougherty RF (2009): Resting- et al. (2015): Atypical functional brain connectivity during rest in state functional connectivity reflects structural connectivity in the autism spectrum disorders. Ann Neurol 77:866–876. default mode network. Cereb Cortex 19:72–78. 76. Power JD, Fair DA, Schlaggar BL, Petersen SE (2010): The devel- 97. Barnea-Goraly N, Kwon H, Menon V, Eliez S, Lotspeich L, Reiss AL opment of human functional brain networks. Neuron 67:735–748. (2004): White matter structure in autism: preliminary evidence from 77. Uddin LQ, Menon V, Young CB, Ryali S, Chen T, Khouzam A, et al. diffusion tensor imaging. Biol Psychiatry 55:323–326. (2011): Multivariate searchlight classification of structural magnetic 98. Simmonds DJ, Hallquist MN, Asato M, Luna B (2014): Developmental resonance imaging in children and adolescents with autism. Biol stages and sex differences of white matter and behavioral develop- Psychiatry 70:833–841. ment through adolescence: A longitudinal diffusion tensor imaging 78. Haar S, Berman S, Behrmann M, Dinstein I (2016): Anatomical (DTI) study. Neuroimage 92:356–368. abnormalities in autism? Cereb Cortex 26:1440–1452. 99. Lebel C, Gee M, Camicioli R, Wieler M, Martin W, Beaulieu C (2012): 79. Valk SL, Di Martino A, Milham MP, Bernhardt BC (2015): Multicenter Diffusion tensor imaging of white matter tract evolution over the mapping of structural network alterations in autism. Hum Brain Mapp lifespan. Neuroimage 60:340–352. 36:2364–2373. 100. Ameis SH, Catani M (2015): Altered white matter connectivity as a 80. Schaer M, Kochalka J, Padmanabhan A, Supekar K, Menon V (2015): neural substrate for social impairment in autism spectrum disorder. Sex differences in cortical volume and gyrification in autism. Mol Cortex 62:158–181. Autism 6:42. 101. Ameis SH, Fan J, Rockel C, Voineskos AN, Lobaugh NJ, Soorya L, 81. Dierker DL, Feczko E, Pruett JR Jr, Petersen SE, Schlaggar BL, et al. (2011): Impaired structural connectivity of socio-emotional cir- Constantino JN, et al. (2015): Analysis of cortical shape in children cuits in autism spectrum disorders: A diffusion tensor imaging study. with simplex autism. Cereb Cortex 25:1042–1051. PloS One 6:e28044. 82. David N, Schultz J, Milne E, Schunke O, Schöttle D, Münchau A, et al. 102. Ouyang M, Cheng H, Mishra V, Gong G, Mosconi MW, Sweeney J, (2014): Right temporoparietal gray matter predicts accuracy of social et al. (2016): Atypical age-dependent effects of autism on white perception in the autism spectrum. J Autism Dev Disord 44: matter microstructure in children of 2-7 years. Hum Brain Mapp 1433–1446. 37:819–832. 83. Raznahan A, Lerch JP, Lee N, Greenstein D, Wallace GL, 103. Ikuta T, Shafritz KM, Bregman J, Peters BD, Gruner P, Stockman M, et al. (2011): Patterns of coordinated anatomical Malhotra AK, et al. (2014): Abnormal cingulum bundle development change in human cortical development: A longitudinal neuroimaging in autism: A probabilistic tractography study. Psychiatry Res 221: study of maturational coupling. Neuron 72:873–884. 63–68. 84. Zielinski BA, Anderson JS, Froehlich AL, Prigge MB, Nielsen JA, 104. Shukla DK, Keehn B, Muller RA (2011): Tract-specific analyses of Cooperrider JR, et al. (2012): scMRI reveals large-scale brain network diffusion tensor imaging show widespread white matter compro- abnormalities in autism. PloS One 7:e49172. mise in autism spectrum disorder. J Child Psychol Psychiatry 52: 85. Mueller S, Keeser D, Samson AC, Kirsch V, Blautzik J, Grothe M, 286–295. et al. (2013): Convergent findings of altered functional and structural 105. Schumann CM, Nordahl CW (2011): Bridging the gap between MRI brain connectivity in individuals with high functioning autism: A and postmortem research in autism. Brain Res 1380:175–186. multimodal MRI study. PloS One 8:e67329. 106. Casanova EL, Casanova MF (2014): Genetics studies indicate that 86. Doyle-Thomas KA, Duerden EG, Taylor MJ, Lerch JP, Soorya LV, neural induction and early neuronal maturation are disturbed in Wang AT, et al. (2013): Effects of age and symptomatology on autism. Front Cell Neurosci 8:397. cortical thickness in autism spectrum disorders. Res Autism Spectr 107. Reiner O, Karzbrun E, Kshirsagar A, Kaibuchi K (2016): Regulation of Disord 7:141–150. neuronal migration, an emerging topic in autism spectrum disorders. 87. Lin HY, Ni HC, Lai MC, Tseng WY, Gau SS (2015): Regional brain J Neurochem 136:440–456. volume differences between males with and without autism spectrum 108. Oblak AL, Rosene DL, Kemper TL, Bauman ML, Blatt GJ (2011): disorder are highly age-dependent. Mol Autism 6:29. Altered posterior cingulate cortical cyctoarchitecture, but normal 88. Ismail MM, Keynton RS, Mostapha MM, ElTanboly AH, density of neurons and interneurons in the posterior cingulate cortex Casanova MF, Gimel’farb GL, et al. (2016): Studying autism spectrum and fusiform gyrus in autism. Autism Res 4:200–211. disorder with structural and diffusion magnetic resonance imaging: A 109. Testa-Silva G, et al. (2012): Hyperconnectivity and slow synapses survey. Front Hum Neurosci 10:211. during early development of medial prefrontal cortex in a mouse 89. Walker L, Gozzi M, Lenroot R, Thurm A, Behseta B, Swedo S, et al. model for mental retardation and autism. Cereb Cortex 22: (2012): Diffusion tensor imaging in young children with autism: Bio- 1333–1342. logical effects and potential confounds. Biol Psychiatry 72:1043–1051. 110. Vattikuti S, Chow CC (2010): A computational model for cerebral 90. Barnea-Goraly N, Lotspeich LJ, Reiss AL (2010): Similar white matter cortical dysfunction in autism spectrum disorders. Biol Psychiatry aberrations in children with autism and their unaffected siblings: A 67:672–678. diffusion tensor imaging study using tract-based spatial statistics. 111. Rubenstein JL (2010): Three hypotheses for developmental defects Arch Gen Psychiatry 67:1052–1060. that may underlie some forms of autism spectrum disorder. Curr Opin 91. Noriuchi M, Kikuchi Y, Yoshiura T, Kira R, Shigeto H, Hara T, et al. Neurol 23:118–123. (2010): Altered white matter fractional anisotropy and social impairment 112. Rubenstein JL, Merzenich MM (2003): Model of autism: Increased in children with autism spectrum disorder. Brain Res 1362:141–149. ratio of excitation/inhibition in key neural systems. Genes Brain 92. Jou RJ, Jackowski AP, Papademetris X, Rajeevan N, Staib LH, Behav 2:255–267. Volkmar FR (2011): Diffusion tensor imaging in autism spectrum 113. Zikopoulos B, Barbas H (2013): Altered neural connectivity in excit- disorders: Preliminary evidence of abnormal neural connectivity. Aust atory and inhibitory cortical circuits in autism. Front Hum Neurosci N Z J psychiatry 45:153–162. 7:609. 93. Catani M, Dell’Acqua F, Budisavljevic S, Howells H, Thiebaut de 114. Leveille C, Barbeau EB, Bolduc C, Limoges E, Berthiaume C, Schotten M, Froudist-Walsh S, et al. (2016): Frontal networks in Chevrier E, et al. (2010): Enhanced connectivity between adults with autism spectrum disorder. Brain 139(pt 2):616–630. and other regions of the brain in autism: A REM EEG coherence 94. Billeci L, Calderoni S, Tosetti M, Catani M, Muratori F (2012): White study. Autism Res 3:280–285. matter connectivity in children with autism spectrum disorders: A 115. Yizhar O, Fenno LE, Prigge M, Schneider F, Davidson TJ, O’Shea DJ, tract-based spatial statistics study. BMC Neurol 12:148. et al. (2011): Neocortical excitation/inhibition balance in information 95. Weinstein M, Ben-Sira L, Levy Y, Zachor DA, Ben Itzhak E, Artzi M, processing and social dysfunction. Nature 477:171–178. et al. (2011): Abnormal white matter integrity in young children with 116. Le Magueresse C, Monyer H (2013): GABAergic interneurons shape autism. Hum Brain Mapp 32:534–543. the functional maturation of the cortex. Neuron 77:388–405.

10 Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI Biological Psychiatry: Default Mode Network in Autism CNNI

117. Paine TA, Swedlow N, Swetschinski L (2016): Decreasing GABA 3-subunits ameliorates behavioral deficits in a mouse model of function within the medial prefrontal cortex or basolateral amygdala autism. Neuron 81:1282–1289. decreases sociability. Behav Brain Res 317:542–552. 131. Menon V (2011): Large-scale brain networks and psychopathology: 118. Baroncelli L, et al. (2011): Brain plasticity and disease: A matter of A unifying triple network model. Trends Cogn Sci 15:483–506. inhibition. Neural Plast 2011:286073. 132. Seeley WW, Menon V, Schatzberg AF, Keller J, Glover GH, Kenna H, 119. Pizzarelli R, Cherubini E (2011): Alterations of GABAergic signaling in et al. (2007): Dissociable intrinsic connectivity networks for salience autism spectrum disorders. Neural Plast 2011:297153. processing and executive control. J Neurosci 27:2349–2356. 120. Thatcher RW, et al. (2009): Autism and EEG phase reset: Deficient 133. Uddin LQ, Supekar K, Lynch CJ, Cheng KM, Odriozola P, Barth ME, GABA mediated inhibition in thalamo-cortical circuits. Dev Neuro- et al. (2015): Brain state differentiation and behavioral inflexibility in psychol 34:780–800. autism. Cereb Cortex 25:4740–4747. 121. Robertson CE, Ratai EM, Kanwisher N (2016): Reduced GABAergic 134. Plitt M, Barnes KA, Wallace GL, Kenworthy L, Martin A, et al. (2015): action in the autistic brain. Curr Biol 26:80–85. Resting-state functional connectivity predicts longitudinal change in 122. Oblak AL, Gibbs TT, Blatt GJ (2011): Reduced GABAA receptors and autistic traits and adaptive functioning in autism. Proc Natl Acad Sci benzodiazepine binding sites in the posterior cingulate cortex and U S A 112:E6699–E6706. fusiform gyrus in autism. Brain Res 1380:218–228. 135. Shulman GL, Fiez JA, Corbetta M, Buckner RL, Miezin FM, Raichle 123. Schroer RJ, Phelan MC, Michaelis RC, Crawford EC, Skinner SA, ME, Petersen SE (1997): Common blood flow changes across vi- Cuccaro M, et al. (1998): Autism and maternally derived aberrations sual tasks: II. Decreases in cerebral cortex. J Cogn Neurosci 9: of chromosome 15q. Am J Med Genet 76:327–336. 648–663. 124. Fatemi SH, Folsom TD, Reutiman TJ, Thuras PD (2009): Expression 136. Buckner RL, Andrews-Hanna JR, Schacter DL (2008): The brain’s of GABA(B) receptors is altered in brains of subjects with autism. default network: Anatomy, function, and relevance to disease. Cerebellum 8:64–69. Ann N Y Acad Sci 1124:1–38. 125. Fatemi SH, Reutiman TJ, Folsom TD, Thuras PD (2009): GABA(A) 137. Shirer WR, Ryali S, Rykhlevskaia E, Menon V, Greicius MD, et al. receptor downregulation in brains of subjects with autism. J Autism (2012): Decoding subject-driven cognitive states with whole-brain Dev Disord 39:223–230. connectivity patterns. Cerebral Cortex 22:158–165. 126. Menon V (2013): Developmental pathways to functional brain net- 138. van den Heuvel M, Mandl R, Luigjes J, Hulshoff Pol H (2008): works: Emerging principles. Trends Cogn Sci 17:627–640. Microstructural organization of the cingulum tract and the 127. Guo X, Chen H, Long Z, Duan X, Zhang Y, Chen H (2017): Atypical level of default mode functional connectivity. J Neurosci developmental trajectory of local spontaneous brain activity in autism 28:10844–10851. spectrum disorder. Sci Rep 7:39822. 139. Tao Y, Liu B, Zhang X, Li J, Qin W, Yu C, et al. (2015): The Structural 128. Cellot G, Cherubini E (2014): GABAergic signaling as therapeutic connectivity pattern of the default mode network and its association target for autism spectrum disorders. Front Pediatr 2:70. with memory and anxiety. Front Neuroanat 9:152. 129. Berry-Kravis EM, Hessl D, Rathmell B, Zarevics P, Cherubini M, Wal- 140. Andrews-Hanna JR, Reidler JS, Sepulcre J, Poulin R, Buckner RL ton-Bowen K, et al. (2012): Effects of STX209 (arbaclofen) on neuro- (2010): Functional-anatomic fractionation of the brain’s default behavioral function in children and adults with fragile X syndrome: A network. Neuron 65:550–562. randomized, controlled, phase 2 trial. Sci Transl Med 4:152ra127. 141. Yarkoni T, Poldrack RA, Nichols TE, Van Essen DC, Wager TD (2011): 130. Han S, Tai C, Jones CJ, Scheuer T, Catterall WA (2014): Enhance- Large-scale automated synthesis of human ment of inhibitory neurotransmission by GABAA receptors having a2, data. Nat Methods 8:665–670.

Biological Psychiatry: Cognitive Neuroscience and Neuroimaging - 2017; -:-–- www.sobp.org/BPCNNI 11