Aberrant multivariate mapping between behavioral profiles and cortical morphological brain networks in children with autism spectrum disorder

Yuqing Liao; Ning Pan; Xiaofan Qiu; Jin Jing; Lizi Lin; Xin Wang; Jinhui Wang

doi:10.1038/s41398-026-04233-w

. 2026 Jul 2;16:483. doi: 10.1038/s41398-026-04233-w

Aberrant multivariate mapping between behavioral profiles and cortical morphological brain networks in children with autism spectrum disorder

Yuqing Liao ^1,^#, Ning Pan ^1,^2,^3,^#, Xiaofan Qiu ¹, Jin Jing ⁴, Lizi Lin ^4,^5,^✉, Xin Wang ^1,^2,^3,^✉, Jinhui Wang ^1,^2,^3,^6,^7,^✉

PMCID: PMC13601539 PMID: 42393028

Abstract

Understanding how human behavior relates to brain structure and function is a central goal of neuroscience. Autism spectrum disorder (ASD) is increasingly conceptualized as a disorder of connectome dysfunction; however, how morphological brain networks relate to behavioral profiles in children with ASD remains unclear. In this study, we collected a comprehensive battery of behavioral assessments and structural MRI data from 101 children with ASD and 67 typically developing (TD) children. Individual morphological brain networks were constructed by integrating four cortical features to estimate morphological connectivity (MC). We then applied sparse canonical correlation analysis (sCCA) to identify multivariate brain-behavior associations within each group. Using group-specific MC features, four distinct association modes were identified in TD children, none of which generalized to the ASD group. In ASD children, three modes were detected that captured comparable behavioral dimensions with three TD modes but were supported by largely distinct neural architectures. Within a unified MC framework, an autistic trait mode emerged in both groups; however, it was primarily supported by MC between the default and control networks in TD children, versus MC between the default and dorsal attention networks in ASD children. Notably, the prosociality mode present in TD children was consistently absent in the ASD group, regardless of the MC features used in sCCA. These findings reveal a reorganization of brain–behavior relationships in children with ASD and provide evidence for a neurodiverse neural architecture, in which similar behaviors arise from divergent neurodevelopmental pathways, moving beyond a simple continuum model of typical development.

Subject terms: Autism spectrum disorders, Neuroscience

Introduction

Mapping human behavior onto brain structure and function is a central goal of neuroscience and has received considerable attention in recent years. To date, numerous associations have been documented between diverse brain features and different behavioral traits [1–6]. However, these brain-behavior associations are mostly identified using bivariate methods, which cannot capture the complex, high-dimensional interactions between brain and behavior. Recent studies have therefore adopted multivariate techniques [e.g., canonical correlation analysis (CCA)] to explore how inter-individual differences in behavior are associated with inter-individual differences in brain organization from a more holistic perspective [5]. Despite significant advances in this topic, existing multivariate brain-behavior mapping studies are mainly conducted in healthy populations. Thus, it remains unclear whether and how these associations are altered in pathological conditions.

Autism spectrum disorder (ASD) affects 1–2% of the global population and poses substantial challenges for individuals and families [7, 8]. Behaviorally, social communication deficits and restricted, repetitive behaviors are its core symptoms [9]. Extending beyond these core symptoms, individuals with ASD also often present other behavioral problems, such as reduced empathizing, heightened systemizing tendencies, and increased emotional difficulties [10]. Neurobiologically, ASD is increasingly viewed as a disorder of brain network dysfunction, marked by widespread alterations in structural and functional connectivity [11–17]. To date, several studies have used multivariate approaches to link structural and functional connectivity with cognitive, social and communication profiles in individuals with ASD [17–20]. In addition to these two connectivity types, morphological connectivity (MC), delineating the coordination of local morphological features between brain regions, stands out in recent years due to its high robustness, reproducibility, and test-retest reliability [21]. Moreover, MC has been proven to explain inter-individual variance and predict individual outcomes in multiple behavioral and cognitive domains [22, 23]. With respect to ASD, MC alterations in multiple circuits can distinguish patients from typically developing (TD) individuals and predict social-communication severity [24–28]. While these findings underscore the potential of MC in ASD research, it remains largely unknown how MC is associated with behavioral profiles in individuals with ASD and whether the association differs from that in TD individuals.

To bridge these gaps, we collected a broad range of behavioral measures and structural magnetic resonance imaging (MRI) data from 101 children with ASD and 67 TD children. First, we constructed individual morphological brain networks by integrating 4 morphological features to estimate MC among 200 regions of interest (ROIs). For each group, we then utilized sparse CCA (sCCA) [29] to characterize the association between the behavioral and MC features after a feature selection procedure applied to the morphological brain networks. Further, we examined whether the association identified in the TD group was generalizable to the ASD group using a feature weight projection approach. Finally, we employed a unified feature selection framework to directly compare the MC-behavior association between the ASD and TD children.

Methods and materials

Participants

A total of 190 children were recruited between April 2019 and December 2024 in the Research Center for Child and Adolescent Psychological and Behavioral Development, Sun Yat-sen University. Before enrollment, the study procedures were explained to all participants and their guardians, and written informed consent was obtained. Detailed inclusion and exclusion criteria are provided in Supplement Section 1. All participants completed behavioral assessments (Supplement Section 2) and T1-weighted structural MRI. The average interval between these two procedures was 2.6 months, with the majority of participants completing both procedures within six months. After quality control, 22 children were excluded due to missing data or poor image quality, yielding a final sample of 101 children with ASD and 67 TD children, comparable to previous ASD MRI studies in sample size [26, 30].

MRI data acquisition and preprocessing

Individual T1-weighted structural images were acquired on a Siemens Skyra 3.0 Tesla MRI System with a magnetization-prepared rapid-gradient echo sequence. All images were preprocessed using a standardized pipeline in the CAT12 toolbox to generate four vertex-wise morphological feature maps. More details are provided in Supplement Section 3.

Morphological brain network construction

We utilized the Schaefer atlas [31] to divide the cortical surface into 200 regions of interest (ROIs). These ROIs were categorized into seven sub-networks, including Visual (Vis), Somatomotor (SomMot), Dorsal Attention (DorsAttn), Salience/Ventral Attention (SalVentAttn), Limbic, Control (Cont), and Default networks. For each ROI, a joint probability distribution (PD) was derived from the four morphological maps via multivariate kernel density estimation [32]. MC was defined as the Jensen-Shannon divergence-based similarity (JSDs) between regional PDs (Supplement Section 4). This resulted in a 200 × 200 MC matrix for each participant, for which its upper triangular elements (19,900 MC) were used for subsequent sCCA.

Sparse canonical correlation analysis

For each group, we applied sCCA to identify multivariate associations between MC and behavioral features. sCCA estimates loading vectors for the MC and behavioral feature matrices separately, projecting each matrix onto a latent dimension such that the resulting canonical variates are maximally correlated. Each pair of canonical variates defines one mode of MC-behavior association, and multiple modes can be extracted sequentially to capture distinct, independent patterns of covariation [33] (Supplement Section 5). Before sCCA, a feature selection step was performed to retain 1000 MC features (Supplement Section 6). Statistical significance of each mode was assessed using 10,000 permutations (Supplement Section 7). Model reliability was evaluated using redundancy–reliability (RR) scores based on training–test splits (Supplement Section 8) [34, 35].

Only modes meeting all of the following criteria were retained: (1) statistically significant (P_FDR < 0.05); (2) a median RR score ≥ 0.8 [34]; and (3) an average correlation between MC and behavioral canonical variates in the test sets that reached at least 50% of that observed in the training sets [36]. For each retained mode, we further identified features that made significant contributions using a bootstrapping procedure (Supplement Section 9).

Of note, the MC features entered into the above sCCA differed between the ASD and TD groups. To promote cross-group comparison, we repeated the sCCA in each group using the assembly of the 1000 MC features per group.

Finally, to evaluate the robustness of the results to feature dimensionality, we repeated the analyses using 2000 MC features within each group.

Results

Demographic and behavioral characteristics

There were significant differences between the ASD and TD groups in most demographic and behavioral variables, with the exception of handedness, scores on organize system, and scores on emotional symptom (Table 1).

Table 1.

Demographic and behavioral characteristics of all participants.

	TD (n = 67)	ASD (n = 101)	χ²/t	P
Sex (Male/Female)	40/27	87/14	15.260	<0.001^a
Age (years)	8.84 ± 1.87	5.56 ± 2.42	9.900	<0.001^b
Maternal age (years)	39.3 ± 4.71	34.40 ± 4.69	−6.621	<0.001^b
Paternal age (years)	42.22 ± 5.79	37.03 ± 5.36	−5.955	<0.001^b
Family income (Low/High)	21/46	68/33	20.936	<0.001^a
Maternal education (Low/High)	4/63	34/67	17.649	<0.001^a
Paternal education (Low/High)	10/57	32/69	6.033	0.014^a
Handedness (Left/Right/Mixed)	5/55/7	8/69/24	4.916	0.086^a
Full-scale intelligence quotient	109.54 ± 10.89	66.92 ± 27.42	14.041	<0.001^b
Social Responsiveness Scale
Social awareness	7.72 ± 2.74	11.71 ± 3.02	−8.708	<0.001^b
Social cognition	9.73 ± 3.96	20.33 ± 5.75	−14.148	<0.001^b
Social communication	10.99 ± 5.93	35.54 ± 12.21	−17.362	<0.001^b
Social motivation	6.39 ± 3.64	15.67 ± 6.43	−11.929	<0.001^b
Autistic mannerism	5.13 ± 3.37	19.45 ± 8.74	−14.875	<0.001^b
Empathy Quotient
Cognitive empathy	10.61 ± 3.72	3.20 ± 2.45	14.390	<0.001^b
Social skill	9.22 ± 3.06	7.62 ± 2.84	3.446	<0.001^b
Affective empathy	6.16 ± 3.03	2.26 ± 2.38	8.897	<0.001^b
Systemizing Quotient
Technical system	5.63 ± 2.89	2.76 ± 2.45	6.901	<0.001^b
Abstract system	8.01 ± 2.36	4.80 ± 2.68	7.969	<0.001^b
Organize system	2.88 ± 2.02	2.49 ± 1.74	1.352	0.178^b
Collectible system	3.45 ± 1.57	2.00 ± 1.50	6.021	<0.001^b
Strengths and Difficulties Questionnaire
Hyperactivity/inattention	4.81 ± 2.45	6.77 ± 2.11	−5.547	<0.001^b
Emotional symptom	2.30 ± 1.92	2.50 ± 1.82	−0.706	0.481^b
Prosocial behavior	7.30 ± 2.07	3.47 ± 2.54	10.720	<0.001^b
Peer relationship problem	2.15 ± 1.55	5.49 ± 1.76	−12.613	<0.001^b
Conduct problem	1.96 ± 1.48	2.50 ± 1.75	−2.076	0.039^b

Open in a new tab

Values are represented as mean ± SD for continuous variables. Low income is defined as less than 8000 RMB per month per person while high income is defined as more than 8000 RMB per month per person. Low education includes primary, secondary, and high school while high education refers to university and above. TD, typical developing; ASD, autism spectrum disorder.

^aP-values were obtained by chi-square tests.

^bP-values were obtained by two-sample t-tests (with Levene’s test for equality of variances).

MC-behavior associations in the TD group

Using the optimized combination of penalty parameters (Fig. S1a), four modes were identified in the TD group, for each of which the MC canonical variate was significantly correlated with the behavioral canonical variate (Fig. 1). We labeled these modes according to the behavioral features that were significantly involved in them (i.e., having large loadings or contributions). Specifically, the first mode (r = 0.918, P_FDR < 0.001; median RR score = 0.959; train/test proportion = 90.1%) was mainly involved in social-communicative difficulties and autistic mannerism and was termed autistic trait mode (Fig. 2a); the second mode (r = 0.876, P_FDR = 0.003; median RR score = 0.927; train/test proportion = 82.5%) was primarily associated with affective empathy, prosocial behavior, and emotional symptom and was termed prosociality mode (Fig. 2b); the third mode (r = 0.872, P_FDR = 0.014; median RR score = 0.838; train/test proportion = 70.3%) was predominantly related to collectible, organize, and technical systems and was termed systemizing mode (Fig. 2c); and the fourth mode (r = 0.870, P_FDR = 0.006; median RR score = 0.924; train/test proportion = 75.9%) was mainly characterized by conduct problem, hyperactivity/inattention, and peer relationship problem and was termed behavior problem mode (Fig. 2d).

Fig. 1 — a Four modes are identified in the TD group that meet the criteria detailed in our methods. **b-e** Scatter plots of the correlation between MC and behavioral canonical variates for each mode highlighted in panel a. Each insert displays the null distribution of the correlation between MC and behavioral canonical variates, with the dashed line indicating the actual correlation. MC, morphological connectivity.

Fig. 2 — **a-d** Each mode is related to distinct behavioral features, with behavioral features significantly contributing to the mode indicated by asterisks. **e-h** The MC canonical variate corresponding to each mode in the TD group exhibited significant correlations with multiple behavioral features among the TD children that were largely comparable with those significantly contributing to the modes. However, these correlations failed to be reproduced in the ASD group when projecting the TD-derived MC loadings to their corresponding MC features in the ASD group for calculating the MC canonical variates. Significant correlations are indicated by filled dots. TD typically developing, ASD autism spectrum disorder, SA social awareness, SCOG social cognition, SCOM social communication, SM social motivation, SA autistic mannerisms, EQCOG empathy quotient cognitive empathy, EQSOC empathy quotient social skill, EQAFF empathy quotient affective empathy, SQTEC systemizing quotient technical systems, SQABS systemizing quotient abstract systems, SQORG systemizing quotient organize systems, SQCOL systemizing quotient collectible systems, HA hyperactivity /inattention, ES emotional symptom, PB prosocial behavior, PP peer relationship problem, CP conduct problem.

With respect to the MC features significantly contributing to these modes (n = 322, 345, 284, and 293, respectively), we calculated their cumulative absolute loadings at both the sub-network and regional levels. We found that the autistic trait mode was mainly associated with the MC between the Default and the Vis, DorsAttn, and SomMot networks and the MC linking several parietal and occipital regions, such as the right superior parietal gyrus and right lingual gyrus (Fig. S2a, b); the prosociality mode was primarily related to the MC across the Default, DorsAttn, and SomMot networks and the MC linking some frontal and parietal regions, such as the right frontal eye field and left precuneus (Fig. S2c, d); the systemizing mode was mainly driven by the MC between the DorsAttn network and SomMot, Cont, and Default networks and the MC linking a few parietal and frontal regions, such as the left postcentral gyrus, right superior parietal gyrus and right frontal eye field (Fig. S2e, f); and the behavior problem mode was predominantly dominated by the MC between the Cont and Vis, DorsAttn and Default, and DorsAttn and SomMot networks and the MC linking some parietal and temporal regions, such as the right superior parietal gyrus and left bank of the superior temporal sulcus (Fig. S2g, h).

Cross-group projection

To explore whether the MC-behavior associations derived from the TD group are generalizable to the ASD, we conducted cross-group projection analyses. Specifically, for each mode in the TD group, we correlated the MC canonical variate (i.e., the product of MC features and their corresponding loadings) with each behavioral feature among the TD children. Significant correlations were observed for multiple behavioral features that were largely comparable with those implicated in the modes identified by sCCA (Fig. 2e-h). However, when we projected the TD-derived MC loadings to their corresponding MC features in the ASD group for calculating the MC canonical variates, these correlations failed to be reproduced in the ASD group (Fig. 2e-h), indicating that the MC-behavior associations are altered in children with ASD.

MC-behavior associations in the ASD group

In the ASD group, three modes were identified (Fig. 3) using the optimized combination of penalty parameters (Fig. S1b). For each of the three modes, we calculated the Spearman correlation, Dice coefficient, and Cosine similarity between its absolute behavioral loadings and the absolute behavioral loadings of each mode derived from the TD group. We found that the first mode (r = 0.858, P_FDR < 0.001; median RR score = 0.946; train/test proportion = 87.0%) exhibited the strongest correspondence with the autistic trait mode of the TD group (Spearman correlation = 0.560; Dice coefficient = 0.706; Cosine similarity = 0.779); the second mode (r = 0.836, P_FDR = 0.023; median RR score = 0.822; train/test proportion = 55.1%) exhibited the strongest correspondence with the behavior problem mode of the TD group (Spearman correlation = 0.570; Dice coefficient = 0.769; Cosine similarity = 0.737); and the third mode (r = 0.830, P_FDR = 0.002; median RR score = 0.853; train/test proportion = 60.0%) exhibited the strongest correspondence with the systemizing mode of the TD group (Spearman correlation = 0.473; Dice coefficient = 0.632; Cosine similarity = 0.746). Therefore, the three modes were labeled as the autistic trait mode, systemizing mode, and behavior problem mode, respectively. Indeed, these labels aligned well with the dominant behavioral features contributing to the three modes (Fig. S3).

Also, we calculated the cumulative absolute loadings of the MC features that significantly contributed to these modes (n = 495, 264, and 239, respectively) at both the sub-network and regional levels (Fig. S4). We found that the distributions of the cumulative absolute loadings exhibited strong positive correlations with those derived from the corresponding modes in the TD group at the sub-network (autistic trait mode: rho = 0.738, P < 0.001; systemizing mode: rho = 0.598, P < 0.001; behavior problem mode: rho = 0.657, P < 0.001) but not regional (autistic trait mode: rho = 0.004, P = 0.953; systemizing mode: rho = 0.029, P = 0.679; behavior problem mode: rho = −0.003, P = 0.965) levels. At the regional level, the autistic trait mode was mainly associated with MC linking several occipital and frontal regions, such as the right lingual gyrus and frontal eye field (Fig. S4b); the behavior problem mode was predominantly associated with MC linking some parietal and temporal regions, such as the right paracentral lobule, right superior parietal gyrus, and left middle temporal gyrus (Fig. S4d); and the systemizing mode was primarily related to MC linking a few temporal and frontal regions, such as the right insula and precentral gyrus (Fig. S4f).

MC-behavior associations in a unified framework of MC features

The above sCCA modes were derived from different sets of MC features between the ASD and TD groups, which limits direct between-group comparisons. Thus, we re-ran sCCA for each group by using the same MC features (i.e., the assembly of the 1000 top-ranked MC features per group). Using the optimized combination of penalty parameters (Fig. S5), two modes were identified in the TD group (Fig. S6) while one in the ASD group (Fig. S7). According to the behavioral measures that were significantly involved in these modes, the first mode in the TD group (r = 0.927, P_FDR < 0.001, median RR score = 0.938, train/test proportion = 90.3%) was mainly involved in social communicative difficulties and autistic mannerism (Fig. S8a) and was termed autistic trait mode and the second mode (r = 0.887, P_FDR = 0.023, median RR score = 0.924, train/test proportion = 72.3%) was primarily associated with affective empathy and prosocial behavior (Fig. S8b) and was termed prosociality mode. For the mode identified in the ASD group (r = 0.873, P_FDR = 0.002, median RR score = 0.914, train/test proportion = 78.0%), it was mainly related to social-communicative difficulties and autistic mannerism (Fig. S8c) and exhibited the strongest correspondence with the autistic trait mode of the TD group in the absolute behavioral loadings (Spearman correlation = 0.559; Dice coefficient = 0.737; Cosine similarity = 0.814) and thus was termed autistic trait mode. The naming of these modes was further supported by examining their correspondence in the absolute behavioral loadings with those modes identified earlier based on the 1000 MC features in the TD group.

With respect to the MC features significantly contributing to these modes, the distributions of their cumulative absolute loadings (Fig. S9) exhibited strong positive correlations with those reported earlier from the corresponding modes in their respective groups at both the sub-network (autistic trait mode_TD: rho = 0.934, P < 0.001; prosociality mode_TD: rho = 0.843, P < 0.001; autistic trait mode_ASD: rho = 0.983, P < 0.001) and regional (autistic trait mode_TD: rho = 0.762, P < 0.001; prosociality mode_TD: rho = 0.805, P < 0.001; autistic trait mode_ASD: rho = 0.803, P < 0.001) levels.

Focusing further on the common autistic trait mode, we found that it was mainly driven by 868 edges in the TD group (Fig. S10a) and 718 edges in the ASD group (Fig. S10b). Across these two sets of edges, only 117 were shared, resulting in a low Dice coefficient of 0.147 (Fig. 4a). These overlapping edges were mainly involved in the Default, Vis, and DorsAttn networks (Fig. 4b) and linked to regions such as the left inferior temporal gyrus and right superior parietal gyrus (Fig. 4c). For the 751 edges unique to the TD group, they were primarily distributed between the Default network and the Vis, SomMot, and Cont networks (Fig. 5a) and concentrated in the left precuneus, right superior parietal gyrus, and right lingual gyrus (Fig. 5b), while for the 601 edges unique to the ASD group, they were mainly distributed between the Default network and the Vis, SomMot, and DorsAttn networks (Fig. 5c) and concentrated in the right lingual gyrus, supramarginal gyrus, and precentral gyrus (Fig. 5d).

Fig. 4 — Only 117 edges are shared between the TD and ASD groups a, which are mainly involved in the default, visual, and dorsal attention networks b and linked to regions such as the left inferior temporal gyrus and right superior parietal gyrus c. TD, typically developing; ASD, autism spectrum disorder; Vis, visual network; SomMot, somatomotor network; DorsAttn, dorsal attention network; SalVentAttn, salience ventral attention network; Cont, control network; Limbic, limbic network; Default, default mode network.

Fig. 5 — a, b For the 751 edges unique to the TD group, they are primarily distributed between the default network and the visual, somatomotor, and control networks and concentrated in the left precuneus, right superior parietal gyrus, and right lingual gyrus. c, d For the 601 edges unique to the ASD group, they are mainly distributed between the default network and the visual, somatomotor, and dorsal attention networks and concentrated in the right lingual gyrus, supramarginal gyrus, and precentral gyrus. Vis visual network, SomMot somatomotor network, DorsAttn dorsal attention network, SalVentAttn salience ventral attention network, Cont control network, Limbic limbic network, Default default mode network.

Between-group differences in MC features

For the common autistic trait mode identified above, we further examined between-group differences in the MC of these shared and unique edges using a permutation testing procedure (10,000 times). No significant differences were observed in any edges (P > 0.05, FDR corrected). These results suggest that the manner in which morphological brain networks relate to behavior is altered in children with ASD, despite the absence of group differences in MC per se.

Validation results

When all sCCA procedures mentioned above were repeated using 2000 MC features per group, the identified modes (Fig. S11) were largely comparable with those reported earlier in terms of the distribution of behavioral and MC feature loadings (Tables S1 and S2).

Discussion

This study employed sCCA to investigate how behavioral profiles map onto MC in children with ASD and TD children. We found that the MC-behavior association modes identified in the TD children did not generalize to the ASD group, indicating a reorganization of brain-behavior relationships in this condition. Although both groups exhibited comparable behavioral dimensions, the neural architectures supporting these dimensions were largely distinct. In particular, the autistic trait mode was primarily characterized by MC between the Default and Cont networks in the TD children, but between the Default and DorsAttn networks in the children with ASD. Notably, the prosociality mode emerged in the TD children but was absent in those with ASD. Collectively, these findings suggest similar behavioral constructs may be instantiated by divergent neural mechanisms in ASD, offering novel insights into the heterogeneity of social-behavioral phenotypes in ASD.

MC-behavior association modes in TD children

In TD children, four distinct MC-behavior association modes were identified, corresponding to autistic trait, prosociality, systemizing, and behavior problem. All modes involved MC across multiple subnetworks, indicating that these behaviors are supported by distributed neural circuits. Particularly, the autistic trait mode was mainly driven by MC between the Default and several other systems, consistent with its pivotal role in social behavior and autistic traits in TD children [37]. Moreover, the systemizing mode was mainly related to MC between the DorsAttn and several other systems, consistent with prior evidence linking this network to systemizing tendencies in children and young adults [38, 39]. Regionally, key contributors to the autistic trait mode were the right superior parietal and lingual gyri, which jointly support social information processing [40–43]. For the prosociality mode, the major regional contributions were from the right frontal eye field and left precuneus, regions supporting executive control [44–47] and self-reflection [48]. These results are consistent with prior findings linking prosocial behavior to cortical thickness in theory-of-mind and executive regions [43]. With regard to the systemizing mode, the key regional contributions emerged in the left postcentral gyrus and right superior parietal gyrus, associated with sensorimotor [49] and attentional processes [50] essential for systemizing. Finally, the most contributive regions to the behavior problem mode were located in the right superior parietal gyrus and left bank of the superior temporal sulcus, consistent with prior evidence linking superior temporal gyrus morphology to conduct problems [51, 52]. Together, these findings highlight the neural substrates of behavior in TD children from an integrative perspective of morphological brain networks.

Altered MC-behavior associations in children with ASD

Cross-group projection showed that the TD-derived modes did not generalize to the ASD group, implying fundamentally distinct neural mechanisms underlying behavioral profiles in patients. This aligns with prior findings of divergent brain-behavior associations in ASD. For instance, Elton et al. [37] reported opposite correlations between salience–occipital connectivity and Social Responsiveness Scale scores in children with and without ASD [37]. Similarly, Pan et al. [30] reported negative correlations between empathizing-systemizing difference score and amygdala/occipital morphology in ASD, but not in TD individuals [30]. It should be noted that ASD is characterized by substantial heterogeneity across both behavioral and neurobiological domains [19, 53]. Such heterogeneity may attenuate the stability of group-level MC-behavior associations and thereby limit cross-group generalization, offering a complementary explanation for why the TD-derived modes failed to generalize to the ASD group.

Applying sCCA to the ASD group identified three modes that broadly corresponded to three of the four modes identified in TD children (i.e., autistic trait, behavior problem, and systemizing) in terms of the behavioral dimensions. Both groups showed similar MC loading patterns between and within networks, supporting the view of ASD as a continuum rather than a discrete category [54, 55]. However, marked regional differences in MC loadings were observed. Specifically, the autistic trait mode in the ASD group was mainly associated with the right lingual gyrus and frontal eye field, areas consistently implicated in visual processing [56], social cognition [57], and repetitive behaviors [58]. For the behavior problem mode, the most contributive regions were the right paracentral lobule, right superior parietal gyrus, and left middle temporal gyrus. The middle temporal gyrus often shows altered connectivity in ASD and is linked to poorer theory-of-mind performance and behavioral problems [59–61]. Finally, the systemizing mode was primarily related to the right insula and precentral gyrus, which may suggest individuals with ASD may depend on embodied cognition, using sensory salience and motor simulation to form understanding [62, 63]. The dissociation between subnetwork-level convergence and region-level divergence in MC loadings may reflect neural degeneracy, wherein different neural substrates support comparable functions [64–66]. Thus, the similar distribution of MC loadings at the sub-network level across groups for a given behavioral dimension likely reflects shared high-level computational demands [18, 67], while the regional differences may arise from the atypical neurodevelopmental trajectory in ASD, characterized by early alterations in synaptic pruning and connectivity [68]. The reorganization of MC-behavior associations underscores the value of identifying connectivity patterns linked to specific behavioral domains. Defining such domain-relevant neural signatures may help refine mechanistic models of ASD and, in turn, inform more targeted interventions. From a translational perspective, these signatures may also support individualized prediction by identifying children who are more likely to benefit from interventions aimed at improving particular behavioral traits.

Divergent neural substrates of the autistic trait mode between children with ASD and TD children

To enable direct quantitative comparisons, we re-ran the sCCA within a unified feature framework. The autistic trait mode was consistently identified in both groups; however, the corresponding MC showed limited overlap between the two groups. TD-specific MC underlying this mode was primarily expressed between the Default and the Cont networks, whereas the ASD-specific MC was predominantly observed between the Default and the DorsAttn networks. This indicates that although both groups engage the Default network, its large-scale integration differs fundamentally between groups. The Cont network suppresses task-irrelevant activities to facilitate goal-directed cognition [69], and its atypical connectivity with the Default network has been consistently linked to ASD symptoms, particularly deficits in inhibitory control, social communication, and repetitive behaviors [70–76]. Thus, the less engagement of the Cont network in ASD may reflect impaired inhibitory control in patients [77, 78]. This speculation is supported by regional findings that the superior parietal gyrus, central to inhibitory control in neurotypical populations [77, 79], was less prominent in relation to the autistic trait mode in ASD. Conversely, the greater engagement of MC between the Default and DorsAttn networks in the autistic trait mode may represent a compensatory strategy to enhance attentional monitoring and mitigate inhibitory deficits in ASD given previous finding of hyper-connectivity between these two networks in patients [80]. Given the cross-sectional nature of the present study, this interpretation remains speculative and will need to be tested in future longitudinal or interventional studies. In addition, greater engagement was observed in the supramarginal and precentral gyrus in the ASD children, suggesting a shift toward sensorimotor processing. This is in line with previous finding that reduced inhibition may be linked to enhanced perceptual capacity in ASD [81].

Absence of the prosociality mode in children with ASD

A key finding was the absence of a prosociality mode in children with ASD. In TD children, this mode encompassed affective empathy and prosocial behavior. Its absence suggests that prosocial traits lack a consistent association with morphological brain networks in ASD, reflecting greater heterogeneity in the neural basis of prosociality. Supporting this view, a prior study reported that ASD-specific functional connectivity related to prosocial behavior showed poor reproducibility across datasets [82]. Similarly, mixed findings were also found for neural correlates of affective empathy, which is closely linked to prosocial behavior in ASD [83]. For example, one study reported an intact affective empathy network in ASD [84], whereas others found altered activity in core nodes of this network, such as the anterior cingulate [85] and anterior insula [86]. These findings, along with our own results, highlight the importance of moving beyond one-size-fits-all approaches when targeting prosocial behavior in children with ASD. In this context, children with ASD may be stratified into neurobiologically informed subgroups based on their MC-behavior association patterns, which could help prioritize intervention targets and support more tailored treatment selection.

Limitations

Several limitations of this study should be acknowledged. First, the ASD group included significantly fewer female participants, consistent with the well-established male predominance in ASD epidemiology [8, 87]. However, given the known sex-related differences in ASD neurobiology [88–90], future studies are needed to clarify potential sex-specific patterns of MC-behavior associations. Second, we adopted a multidimensional probability distribution approach to integrate different morphological features for network construction. Recent studies have shown that morphological networks generated using different feature-integration strategies differ in the stability of their cognitive associations, test–retest reliability, and sensitivity to age [91, 92]. Therefore, future studies should systematically evaluate the reproducibility of our findings across different feature-integration strategies. Third, accumulating evidence indicates that ASD is associated with substantial heterogeneity across both behavioral and neurobiological domains [19, 53]. Thus, future studies with larger samples are important to examine MC-behavior associations across ASD subgroups defined by functional level, intellectual ability, or neurobiological features. Fourth, the sample in this study spanned a broad developmental window (2–12 years), a period marked by substantial age-related changes in both brain morphology and behavior. Accordingly, MC-behavior associations may vary across developmental stages, an issue that warrants systematic investigation in future work. Finally, our analyses focused on MC; incorporating functional and structural connectivity in future work will enable a more comprehensive, multimodal understanding of how brain networks relate to behavior in ASD.

Conclusion

In conclusion, this study mapped multivariate associations between behavioral profiles and MC in children with and without ASD. We found that although both groups exhibited similar behavioral dimensions, the neural architectures supporting these dimensions were largely distinct. This suggests that TD individuals may express behavioral traits as part of normative neurodevelopmental variation. In ASD, however, similar behaviors likely arise from atypical neurodevelopmental pathways. Thus, our findings challenge the notion that ASD simply reflects a quantitative extreme of typical traits and demonstrate the neurodiverse architecture of the autistic brain.

Supplementary information

Supplementary information^{(3.6MB, docx)}

Acknowledgements

The authors thank all participants and their families for their participation in this study.

Author contributions

JHW designed the study, supervised the overall conduct of the study, and reviewed and revised the manuscript. NP contributed to data acquisition and reviewed and revised the manuscript. JJ contributed to data acquisition. LZL contributed to data acquisition and reviewed and revised the manuscript. XW contributed to data acquisition and reviewed and revised the manuscript. YQL analyzed the data and drafted the manuscript. XFQ contributed to data analysis.

Funding

This work was supported by Brain Science and Brain-like Intelligence Technology - National Science and Technology Major project (No. 2021ZD0200500), National Natural Science Foundation of China (Nos. 82472092, 52400247, and 82574037), China Postdoctoral Science Foundation (No. 2024M750966), Guangdong Basic and Applied Basic Research Foundation (Nos. 2025A1515010685, 2023A1515110220, and 2022B1515130007), Research Center for Brain Cognition and Human Development, Guangdong, China (No. 2024B0303390003), Guangdong University Featured Innovation Program Project (No. 2024WTSCX052), Guangdong Philosophy and Social Science Foundation (No. GD25YGG24), and Scientific Research Innovation Project of Graduate School of South China Normal University (No. 2025KYLX015).

Data availability

All data that support the findings of this study are available from the corresponding author upon reasonable request. Due to the nature of this research, participants of this study did not agree for their data to be shared publicly.

Code availability

The code for morphological network construction was publicly available at https://github.com/Junle-1995/Morphological-Networks-ASD. The code for sparse canonical correlation analysis was publicly available at https://github.com/cedricx/sCCA/tree/master/sCCA/code/final.

Ethics declarations

The study was approved by the Institutional Review Board of the Institute for Brain Research and Rehabilitation, South China Normal University (SCNU-BRR-2023-029). Informed consent was obtained from all participants and their parents, and all methods were performed in accordance with the relevant guidelines and regulations.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Yuqing Liao, Ning Pan.

Contributor Information

Lizi Lin, Email: linlz@mail.sysu.edu.cn.

Xin Wang, Email: wxin3@scnu.edu.cn.

Jinhui Wang, Email: jinhui.wang.1982@m.scnu.edu.cn.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41398-026-04233-w.

References

1.Foster NEV, Zatorre RJ. Cortical structure predicts success in performing musical transformation judgments. NeuroImage. 2010;53:26–36. [DOI] [PubMed] [Google Scholar]
2.Rice K, Redcay E. Spontaneous mentalizing captures variability in the cortical thickness of social brain regions. Soc Cogn Affect Neurosci. 2015;10:327–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Kanai R, Feilden T, Firth C, Rees G. Political orientations are correlated with brain structure in young adults. Curr Biol. 2011;21:677–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Nostro AD, Müller VI, Reid AT, Eickhoff SB. Correlations between personality and brain structure: a crucial role of gender. Cereb Cortex. 2016;26:3458–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Genon S, Eickhoff SB, Kharabian S. Linking interindividual variability in brain structure to behaviour. Nat Rev Neurosci. 2022;23:307–18. [DOI] [PubMed] [Google Scholar]
6.Marek S, Tervo-Clemmens B, Calabro FJ, Montez DF, Kay BP, Hatoum AS, et al. Reproducible brain-wide association studies require thousands of individuals. Nature. 2022;603:654–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Lord C, Elsabbagh M, Baird G, Veenstra-Vanderweele J. Autism spectrum disorder. The Lancet. 2018;392:508–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Maenner MJ, Warren Z, Williams AR, Amoakohene E, Bakian AV, Bilder DA, et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years — autism and developmental disabilities monitoring network, 11 sites, United States, 2020. MMWR Surveill Summ. 2023;72:1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Hirota T, King BH. Autism spectrum disorder: a review. JAMA. 2023;329:157–68. [DOI] [PubMed] [Google Scholar]
10.Lombardo MV, Lai M-C, Baron-Cohen S. Big data approaches to decomposing heterogeneity across the autism spectrum. Mol Psychiatry. 2019;24:1435–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Jou RJ, Jackowski AP, Papademetris X, Rajeevan N, Staib LH, Volkmar FR. Diffusion tensor imaging in autism spectrum disorders: preliminary evidence of abnormal neural connectivity. Aust N Z J Psychiatry. 2011;45:153–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Nair A, Treiber JM, Shukla DK, Shih P, Müller R-A. Impaired thalamocortical connectivity in autism spectrum disorder: a study of functional and anatomical connectivity. Brain. 2013;136:1942–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Woodward ND, Giraldo-Chica M, Rogers B, Cascio CJ. Thalamocortical dysconnectivity in autism spectrum disorder: an analysis of the autism brain imaging data exchange. Biol Psychiatry Cogn Neurosci Neuroimaging. 2017;2:76–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Pereira AM, Campos BM, Coan AC, Pegoraro LF, De Rezende TJR, Obeso I, et al. Differences in cortical structure and functional MRI connectivity in high functioning autism. Front Neurol. 2018;9:539. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Nunes AS, Peatfield N, Vakorin V, Doesburg SM. Idiosyncratic organization of cortical networks in autism spectrum disorder. NeuroImage. 2019;190:182–90. [DOI] [PubMed] [Google Scholar]
16.Hong S-J, Vos De Wael R, Bethlehem RAI, Lariviere S, Paquola C, Valk SL, et al. Atypical functional connectome hierarchy in autism. Nat Commun. 2019;10:1022. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Jang Y. Structural connectome alterations between individuals with autism and neurotypical controls using feature representation learning. Behav Brain Funct. 2024;20:2. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Easson AK, Fatima Z, McIntosh AR. Functional connectivity-based subtypes of individuals with and without autism spectrum disorder. Netw Neurosci. 2019;3:344–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Buch AM, Vértes PE, Seidlitz J, Kim SH, Grosenick L, Liston C. Molecular and network-level mechanisms explaining individual differences in autism spectrum disorder. Nat Neurosci. 2023;26:650–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Tong X, Xie H, Fonzo GA, Zhao K, Satterthwaite TD, Carlisle NB, et al. Symptom dimensions of resting-state electroencephalographic functional connectivity in autism. Nat Ment Health. 2024;2:287–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Wang J, He Y. Toward individualized connectomes of brain morphology. Trends Neurosci. 2023;46:789–802. [DOI] [PubMed] [Google Scholar]
22.Li Z, Li J, Wang N, Lv Y, Zou Q, Wang J. Single-subject cortical morphological brain networks: Phenotypic associations and neurobiological substrates. NeuroImage. 2023;283:120434. [DOI] [PubMed] [Google Scholar]
23.Lin Q, Jin S, Yin G, Li J, Asgher U, Qiu S, et al. Cortical morphological networks differ between gyri and sulci. Neurosci Bull. 2025;41:46–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Soussia M, Rekik I High-order connectomic manifold learning for autistic brain state identification. In: Wu G, Laurienti P, Bonilha L, Munsell BC (eds). Connectomics in NeuroImaging. Cham: Springer International Publishing; 2017, pp. 51–59.
25.Leming MJ, Baron-Cohen S, Suckling J. Single-participant structural similarity matrices lead to greater accuracy in classification of participants than function in autism in MRI. Mol Autism. 2021;12:63. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.He C, Cortes JM, Kang X, Cao J, Chen H, Guo X, et al. Individual-based morphological brain network organization and its association with autistic symptoms in young children with autism spectrum disorder. Hum Brain Mapp. 2021;42:3282–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Gao J, Chen M, Li Y, Gao Y, Li Y, Cai S, et al. Multisite autism spectrum disorder classification using convolutional neural network classifier and individual morphological brain networks. Front Neurosci. 2021;14:629630. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Yi T, Wei W, Ma D, Wu Y, Cai Q, Jin K, et al. Individual brain morphological connectome indicator based on jensen–shannon divergence similarity estimation for autism spectrum disorder identification. Front Neurosci. 2022;16:952067. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Witten DM, Tibshirani R, Hastie T. A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis. Biostatistics. 2009;10:515–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Pan N, Lin L-Z, Wang X, Shi L, Xu X-Y, Jin Y-Y, et al. Brain structure underlying the empathizing–systemizing difference in children with autism spectrum disorder. World J Pediatr. 2023;19:782–92. [DOI] [PubMed] [Google Scholar]
31.Schaefer A, Kong R, Gordon EM, Laumann TO, Zuo X-N, Holmes AJ, et al. Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI. Cereb Cortex. 2018;28:3095–114. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Wang Y, Li J, Jin S, Wang J, Lv Y, Zou Q, et al. Mapping morphological cortical networks with joint probability distributions from multiple morphological features. NeuroImage. 2024;296:120673. [DOI] [PubMed] [Google Scholar]
33.Xia CH, Ma Z, Ciric R, Gu S, Betzel RF, Kaczkurkin AN, et al. Linked dimensions of psychopathology and connectivity in functional brain networks. Nat Commun. 2018;9:3003. [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Moser DA, Doucet GE, Lee WH, Rasgon A, Krinsky H, Leibu E, et al. Multivariate associations among behavioral, clinical, and multimodal imaging phenotypes in patients with psychosis. JAMA Psychiatry. 2018;75:386. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Modabbernia A, Reichenberg A, Ing A, Moser DA, Doucet GE, Artiges E, et al. Linked patterns of biological and environmental covariation with brain structure in adolescence: a population-based longitudinal study. Mol Psychiatry. 2021;26:4905–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Li Q, Zhao Y, Hu Y, Liu Y, Wang Y, Zhang Q, et al. Linked patterns of symptoms and cognitive covariation with functional brain controllability in major depressive disorder. eBioMedicine. 2024;106:105255. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Elton A, Di Martino A, Hazlett HC, Gao W. Neural connectivity evidence for a categorical-dimensional hybrid model of autism spectrum disorder. Biol Psychiatry. 2016;80:120–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Sassa Y, Taki Y, Takeuchi H, Hashizume H, Asano M, Asano K, et al. The correlation between brain gray matter volume and empathizing and systemizing quotients in healthy children. NeuroImage. 2012;60:2035–41. [DOI] [PubMed] [Google Scholar]
39.Takeuchi H, Taki Y, Sassa Y, Hashizume H, Sekiguchi A, Fukushima A, et al. Regional gray matter volume is associated with empathizing and systemizing in young adults. PLoS ONE. 2014;9:e84782. [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Nomi JS, Scherfeld D, Friederichs S, Schäfer R, Franz M, Wittsack H-J, et al. On the neural networks of empathy: a principal component analysis of an fMRI study. Behav Brain Funct. 2008;4:41. [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Palejwala AH, Dadario NB, Young IM, O’Connor K, Briggs RG, Conner AK, et al. Anatomy and white matter connections of the lingual gyrus and cuneus. World Neurosurg. 2021;151:426–37. [DOI] [PubMed] [Google Scholar]
42.Wang J, Yang Y, Fan L, Xu J, Li C, Liu Y, et al. Convergent functional architecture of the superior parietal lobule unraveled with multimodal neuroimaging approaches: parcellation of superior parietal lobule. Hum Brain Mapp. 2015;36:238–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Pretzsch CM, Schäfer T, Lombardo MV, Warrier V, Mann C, Bletsch A, et al. Neurobiological correlates of change in adaptive behavior in autism. Am J Psychiatry. 2022;179:336–49. [DOI] [PubMed] [Google Scholar]
44.Husain M, Parton A, Hodgson TL, Mort D, Rees G. Self-control during response conflict by human supplementary eye field. Nat Neurosci. 2003;6:117–8. [DOI] [PubMed] [Google Scholar]
45.Armstrong KM, Chang MH, Moore T. Selection and maintenance of spatial information by frontal eye field neurons. J Neurosci. 2009;29:15621–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Bedini M, Olivetti E, Avesani P, Baldauf D. Accurate localization and coactivation profiles of the frontal eye field and inferior frontal junction: an ALE and MACM fMRI meta-analysis. Brain Struct Funct. 2023;228:997–1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Song F, Dong X, Zhao J, Wang J, Sang X, He X, et al. Causal role of the frontal eye field in attention-induced ocular dominance plasticity. eLife. 2024;12:e93213. [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Dadario NB, Sughrue ME. The functional role of the precuneus. Brain. 2023;146:3598–607. [DOI] [PubMed] [Google Scholar]
49.Alahmadi AAS. Investigating the sub-regions of the superior parietal cortex using functional magnetic resonance imaging connectivity. Insights Imaging. 2021;12:47. [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Baron-Cohen S, Lombardo MV. Autism and talent: the cognitive and neural basis of systemizing. Dialogues Clin Neurosci. 2017;19:345–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Rogers JC, De Brito SA. Cortical and subcortical gray matter volume in youths with conduct problems: a meta-analysis. JAMA Psychiatry. 2016;73:64–72. [DOI] [PubMed] [Google Scholar]
52.Gao Y, Jiang Y, Ming Q, Zhang J, Ma R, Wu Q, et al. Neuroanatomical changes associated with conduct disorder in boys: influence of childhood maltreatment. Eur Child Adolesc Psychiatry. 2022;31:601–13. [DOI] [PubMed] [Google Scholar]
53.Guo X, Zhai G, Liu J, Cao Y, Zhang X, Cui D, et al. Inter-individual heterogeneity of functional brain networks in children with autism spectrum disorder. Mol Autism. 2022;13:52. [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Constantino JN, Todd RD. Autistic traits in the general population: a twin study. Arch Gen Psychiatry. 2003;60:524–30. [DOI] [PubMed] [Google Scholar]
55.Rashid B, Blanken LME, Muetzel RL, Miller R, Damaraju E, Arbabshirani MR, et al. Connectivity dynamics in typical development and its relationship to autistic traits and autism spectrum disorder. Hum Brain Mapp. 2018;39:3127–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Degré-Pelletier J, Danis É, Thérien VD, Bernhardt B, Barbeau EB, Soulières I. Differential neural correlates underlying visuospatial versus semantic reasoning in autistic children. Cereb Cortex. 2024;34:19–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Itahashi T, Yamada T, Watanabe H, Nakamura M, Ohta H, Kanai C, et al. Alterations of local spontaneous brain activity and connectivity in adults with high-functioning autism spectrum disorder. Mol Autism. 2015;6:30. [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Agam Y, Joseph RM, Barton JJS, Manoach DS. Reduced cognitive control of response inhibition by the anterior cingulate cortex in autism spectrum disorders. NeuroImage. 2010;52:336–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Carter Leno V, Chandler S, White P, Yorke I, Charman T, Jones CRG, et al. Associations between theory of mind and conduct problems in autistic and nonautistic youth. Autism Res. 2021;14:276–88. [DOI] [PubMed] [Google Scholar]
60.Cheng W, Rolls ET, Gu H, Zhang J, Feng J. Autism: reduced connectivity between cortical areas involved in face expression, theory of mind, and the sense of self. Brain. 2015;138:1382–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Xu J, Wang C, Xu Z, Li T, Chen F, Chen K, et al. Specific functional connectivity patterns of middle temporal gyrus subregions in children and adults with autism spectrum disorder. Autism Res. 2020;13:410–22. [DOI] [PubMed] [Google Scholar]
62.Cooke DF, Graziano MSA. Sensorimotor integration in the precentral gyrus: polysensory neurons and defensive movements. J Neurophysiol. 2004;91:1648–60. [DOI] [PubMed] [Google Scholar]
63.Gogolla N. The insular cortex. Curr Biol. 2017;27:580–6. [DOI] [PubMed] [Google Scholar]
64.Tononi G. Consciousness and complexity. Science. 1998;282:1846–51. [DOI] [PubMed] [Google Scholar]
65.Edelman GM, Gally JA. Degeneracy and complexity in biological systems. Proc Natl Acad Sci. 2001;98:13763–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Albantakis L, Bernard C, Brenner N, Marder E, Narayanan R. The Brain’s best kept secret is its degenerate structure. J Neurosci. 2024;44:e1339242024. [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Lake EMR, Finn ES, Noble SM, Vanderwal T, Shen X, Rosenberg MD, et al. The functional brain organization of an individual allows prediction of measures of social abilities transdiagnostically in autism and Attention-Deficit/Hyperactivity disorder. Biol Psychiatry. 2019;86:315–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Geschwind DH, Levitt P. Autism spectrum disorders: developmental disconnection syndromes. Curr Opin Neurobiol. 2007;17:103–11. [DOI] [PubMed] [Google Scholar]
69.Thomas Yeo BT, Krienen FM, Sepulcre J, Sabuncu MR, Lashkari D, Hollinshead M, et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J Neurophysiol. 2011;106:1125–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
70.Bishop DVM, Norbury CF. Executive functions in children with communication impairments, in relation to autistic symptomatology: 2: response inhibition. Autism. 2005;9:29–43. [DOI] [PubMed] [Google Scholar]
71.Mosconi MW, Kay M, D’Cruz A-M, Seidenfeld A, Guter S, Stanford LD, et al. Impaired inhibitory control is associated with higher-order repetitive behaviors in autism spectrum disorders. Psychol Med. 2009;39:1559–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Geurts HM, Van Den Bergh SFWM, Ruzzano L. Prepotent response inhibition and interference control in autism spectrum disorders: two meta-analyses. Autism Res. 2014;7:407–20. [DOI] [PubMed] [Google Scholar]
73.Schmitt LM, White SP, Cook EH, Sweeney JA, Mosconi MW. Cognitive mechanisms of inhibitory control deficits in autism spectrum disorder. J Child Psychol Psychiatry. 2018;59:586–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Voorhies W, Dajani DR, Vij SG, Shankar S, Turan TO, Uddin LQ. Aberrant functional connectivity of inhibitory control networks in children with autism spectrum disorder. Autism Res. 2018;11:1468–78. [DOI] [PubMed] [Google Scholar]
75.De Lacy N, Doherty D, King BH, Rachakonda S, Calhoun VD. Disruption to control network function correlates with altered dynamic connectivity in the wider autism spectrum. NeuroImage Clin. 2017;15:513–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Wang L, Qin Y, Yang S, Jin D, Zhu Y, Li X, et al. Posterior default mode network is associated with the social performance in male children with autism spectrum disorder: a dynamic causal modeling analysis based on triple-network model. Hum Brain Mapp. 2024;45:e26750. [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Durston S, Thomas KM, Yang Y, Uluğ AM, Zimmerman RD, Casey BJ. A neural basis for the development of inhibitory control. Dev Sci. 2002;5:9–16. [Google Scholar]
78.Schmitz N, Rubia K, Daly E, Smith A, Williams S, Murphy DGM. Neural correlates of executive function in autistic spectrum disorders. Biol Psychiatry. 2006;59:7–16. [DOI] [PubMed] [Google Scholar]
79.Levy BJ, Wagner AD. Cognitive control and right ventrolateral prefrontal cortex: reflexive reorienting, motor inhibition, and action updating. Ann N Y Acad Sci. 2011;1224:40–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
80.Norman LJ, Sudre G, Bouyssi-Kobar M, Jiao M, Gligorovic S, Jean J, et al. Cross-sectional mega-analysis of resting-state alterations associated with autism and attention-deficit/hyperactivity disorder in children and adolescents. Nat Ment Health. 2025;3:709–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
81.Adams NC, Jarrold C. Inhibition in autism: children with autism have difficulty inhibiting irrelevant distractors but not prepotent responses. J Autism Dev Disord. 2012;42:1052–63. [DOI] [PubMed] [Google Scholar]
82.Voldsbekk I, Kjelkenes R, Frogner ER, Westlye LT, Alnæs D. Testing the sensitivity of diagnosis-derived patterns in functional brain networks to symptom burden in a Norwegian youth sample. Hum Brain Mapp. 2024;45:e26631. [DOI] [PMC free article] [PubMed] [Google Scholar]
83.Wang X, Auyeung B, Pan N, Lin L-Z, Chen Q, Chen J-J, et al. Empathy, theory of mind, and prosocial behaviors in autistic children. Front Psychiatry. 2022;13:844578. [DOI] [PMC free article] [PubMed] [Google Scholar]
84.Guo X, Xu C, Chen J, Wu Z, Hou S, Wei Z. Disrupted cognitive and affective empathy network interactions in autistic children viewing social animation. Soc Cogn Affect Neurosci. 2024;19:nsae028. [DOI] [PMC free article] [PubMed] [Google Scholar]
85.Balsters JH, Apps MAJ, Bolis D, Lehner R, Gallagher L, Wenderoth N. Disrupted prediction errors index social deficits in autism spectrum disorder. Brain. 2017;140:235–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
86.Gu X, Eilam-Stock T, Zhou T, Anagnostou E, Kolevzon A, Soorya L, et al. Autonomic and brain responses associated with empathy deficits in autism spectrum disorder. Hum Brain Mapp. 2015;36:3323–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
87.Loomes R, Hull L, Mandy WPL. What is the male-to-female ratio in autism spectrum disorder? A systematic review and meta-analysis. J Am Acad Child Adolesc Psychiatry. 2017;56:466–74. [DOI] [PubMed] [Google Scholar]
88.Lee JK, Amaral DG, Solomon M, Rogers SJ, Ozonoff S, Nordahl CW. Sex differences in the amygdala resting-state connectome of children with autism spectrum disorder. Biol Psychiatry Cogn Neurosci Neuroimaging. 2020;5:320–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
89.Walsh MJM, Wallace GL, Gallegos SM, Braden BB. Brain-based sex differences in autism spectrum disorder across the lifespan: a systematic review of structural MRI, fMRI, and DTI findings. NeuroImage Clin. 2021;31:102719. [DOI] [PMC free article] [PubMed] [Google Scholar]
90.Zielinski BA, Andrews DS, Lee JK, Solomon M, Rogers SJ, Heath B, et al. Sex-dependent structure of socioemotional salience, executive control, and default mode networks in preschool-aged children with autism. NeuroImage. 2022;257:119252. [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Han S, Jia H, Vilaseca G, Vilaró N, Duan F, Sun Z, et al. Revealing structural brain-cognition relationships in children: a comparison of morphometric similarity and INverse divergence networks. Neuroinformatics. 2026;24:4. [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Gao J, Hu Y. MIND versus MSN: a systematic evaluation of test–retest reliability and age sensitivity for T1-weighted structural similarity networks. Netw Neurosci. 2026. [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary information^{(3.6MB, docx)}

Data Availability Statement

[CR1] 1.Foster NEV, Zatorre RJ. Cortical structure predicts success in performing musical transformation judgments. NeuroImage. 2010;53:26–36. [DOI] [PubMed] [Google Scholar]

[CR2] 2.Rice K, Redcay E. Spontaneous mentalizing captures variability in the cortical thickness of social brain regions. Soc Cogn Affect Neurosci. 2015;10:327–34. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR3] 3.Kanai R, Feilden T, Firth C, Rees G. Political orientations are correlated with brain structure in young adults. Curr Biol. 2011;21:677–80. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Nostro AD, Müller VI, Reid AT, Eickhoff SB. Correlations between personality and brain structure: a crucial role of gender. Cereb Cortex. 2016;26:3458–70. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Genon S, Eickhoff SB, Kharabian S. Linking interindividual variability in brain structure to behaviour. Nat Rev Neurosci. 2022;23:307–18. [DOI] [PubMed] [Google Scholar]

[CR6] 6.Marek S, Tervo-Clemmens B, Calabro FJ, Montez DF, Kay BP, Hatoum AS, et al. Reproducible brain-wide association studies require thousands of individuals. Nature. 2022;603:654–60. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR7] 7.Lord C, Elsabbagh M, Baird G, Veenstra-Vanderweele J. Autism spectrum disorder. The Lancet. 2018;392:508–20. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Maenner MJ, Warren Z, Williams AR, Amoakohene E, Bakian AV, Bilder DA, et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years — autism and developmental disabilities monitoring network, 11 sites, United States, 2020. MMWR Surveill Summ. 2023;72:1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Hirota T, King BH. Autism spectrum disorder: a review. JAMA. 2023;329:157–68. [DOI] [PubMed] [Google Scholar]

[CR10] 10.Lombardo MV, Lai M-C, Baron-Cohen S. Big data approaches to decomposing heterogeneity across the autism spectrum. Mol Psychiatry. 2019;24:1435–50. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Jou RJ, Jackowski AP, Papademetris X, Rajeevan N, Staib LH, Volkmar FR. Diffusion tensor imaging in autism spectrum disorders: preliminary evidence of abnormal neural connectivity. Aust N Z J Psychiatry. 2011;45:153–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Nair A, Treiber JM, Shukla DK, Shih P, Müller R-A. Impaired thalamocortical connectivity in autism spectrum disorder: a study of functional and anatomical connectivity. Brain. 2013;136:1942–55. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.Woodward ND, Giraldo-Chica M, Rogers B, Cascio CJ. Thalamocortical dysconnectivity in autism spectrum disorder: an analysis of the autism brain imaging data exchange. Biol Psychiatry Cogn Neurosci Neuroimaging. 2017;2:76–84. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Pereira AM, Campos BM, Coan AC, Pegoraro LF, De Rezende TJR, Obeso I, et al. Differences in cortical structure and functional MRI connectivity in high functioning autism. Front Neurol. 2018;9:539. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Nunes AS, Peatfield N, Vakorin V, Doesburg SM. Idiosyncratic organization of cortical networks in autism spectrum disorder. NeuroImage. 2019;190:182–90. [DOI] [PubMed] [Google Scholar]

[CR16] 16.Hong S-J, Vos De Wael R, Bethlehem RAI, Lariviere S, Paquola C, Valk SL, et al. Atypical functional connectome hierarchy in autism. Nat Commun. 2019;10:1022. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR17] 17.Jang Y. Structural connectome alterations between individuals with autism and neurotypical controls using feature representation learning. Behav Brain Funct. 2024;20:2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Easson AK, Fatima Z, McIntosh AR. Functional connectivity-based subtypes of individuals with and without autism spectrum disorder. Netw Neurosci. 2019;3:344–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Buch AM, Vértes PE, Seidlitz J, Kim SH, Grosenick L, Liston C. Molecular and network-level mechanisms explaining individual differences in autism spectrum disorder. Nat Neurosci. 2023;26:650–63. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Tong X, Xie H, Fonzo GA, Zhao K, Satterthwaite TD, Carlisle NB, et al. Symptom dimensions of resting-state electroencephalographic functional connectivity in autism. Nat Ment Health. 2024;2:287–98. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Wang J, He Y. Toward individualized connectomes of brain morphology. Trends Neurosci. 2023;46:789–802. [DOI] [PubMed] [Google Scholar]

[CR22] 22.Li Z, Li J, Wang N, Lv Y, Zou Q, Wang J. Single-subject cortical morphological brain networks: Phenotypic associations and neurobiological substrates. NeuroImage. 2023;283:120434. [DOI] [PubMed] [Google Scholar]

[CR23] 23.Lin Q, Jin S, Yin G, Li J, Asgher U, Qiu S, et al. Cortical morphological networks differ between gyri and sulci. Neurosci Bull. 2025;41:46–60. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Soussia M, Rekik I High-order connectomic manifold learning for autistic brain state identification. In: Wu G, Laurienti P, Bonilha L, Munsell BC (eds). Connectomics in NeuroImaging. Cham: Springer International Publishing; 2017, pp. 51–59.

[CR25] 25.Leming MJ, Baron-Cohen S, Suckling J. Single-participant structural similarity matrices lead to greater accuracy in classification of participants than function in autism in MRI. Mol Autism. 2021;12:63. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.He C, Cortes JM, Kang X, Cao J, Chen H, Guo X, et al. Individual-based morphological brain network organization and its association with autistic symptoms in young children with autism spectrum disorder. Hum Brain Mapp. 2021;42:3282–94. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Gao J, Chen M, Li Y, Gao Y, Li Y, Cai S, et al. Multisite autism spectrum disorder classification using convolutional neural network classifier and individual morphological brain networks. Front Neurosci. 2021;14:629630. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR28] 28.Yi T, Wei W, Ma D, Wu Y, Cai Q, Jin K, et al. Individual brain morphological connectome indicator based on jensen–shannon divergence similarity estimation for autism spectrum disorder identification. Front Neurosci. 2022;16:952067. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Witten DM, Tibshirani R, Hastie T. A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis. Biostatistics. 2009;10:515–34. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Pan N, Lin L-Z, Wang X, Shi L, Xu X-Y, Jin Y-Y, et al. Brain structure underlying the empathizing–systemizing difference in children with autism spectrum disorder. World J Pediatr. 2023;19:782–92. [DOI] [PubMed] [Google Scholar]

[CR31] 31.Schaefer A, Kong R, Gordon EM, Laumann TO, Zuo X-N, Holmes AJ, et al. Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI. Cereb Cortex. 2018;28:3095–114. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Wang Y, Li J, Jin S, Wang J, Lv Y, Zou Q, et al. Mapping morphological cortical networks with joint probability distributions from multiple morphological features. NeuroImage. 2024;296:120673. [DOI] [PubMed] [Google Scholar]

[CR33] 33.Xia CH, Ma Z, Ciric R, Gu S, Betzel RF, Kaczkurkin AN, et al. Linked dimensions of psychopathology and connectivity in functional brain networks. Nat Commun. 2018;9:3003. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR34] 34.Moser DA, Doucet GE, Lee WH, Rasgon A, Krinsky H, Leibu E, et al. Multivariate associations among behavioral, clinical, and multimodal imaging phenotypes in patients with psychosis. JAMA Psychiatry. 2018;75:386. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Modabbernia A, Reichenberg A, Ing A, Moser DA, Doucet GE, Artiges E, et al. Linked patterns of biological and environmental covariation with brain structure in adolescence: a population-based longitudinal study. Mol Psychiatry. 2021;26:4905–18. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR36] 36.Li Q, Zhao Y, Hu Y, Liu Y, Wang Y, Zhang Q, et al. Linked patterns of symptoms and cognitive covariation with functional brain controllability in major depressive disorder. eBioMedicine. 2024;106:105255. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Elton A, Di Martino A, Hazlett HC, Gao W. Neural connectivity evidence for a categorical-dimensional hybrid model of autism spectrum disorder. Biol Psychiatry. 2016;80:120–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Sassa Y, Taki Y, Takeuchi H, Hashizume H, Asano M, Asano K, et al. The correlation between brain gray matter volume and empathizing and systemizing quotients in healthy children. NeuroImage. 2012;60:2035–41. [DOI] [PubMed] [Google Scholar]

[CR39] 39.Takeuchi H, Taki Y, Sassa Y, Hashizume H, Sekiguchi A, Fukushima A, et al. Regional gray matter volume is associated with empathizing and systemizing in young adults. PLoS ONE. 2014;9:e84782. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR40] 40.Nomi JS, Scherfeld D, Friederichs S, Schäfer R, Franz M, Wittsack H-J, et al. On the neural networks of empathy: a principal component analysis of an fMRI study. Behav Brain Funct. 2008;4:41. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Palejwala AH, Dadario NB, Young IM, O’Connor K, Briggs RG, Conner AK, et al. Anatomy and white matter connections of the lingual gyrus and cuneus. World Neurosurg. 2021;151:426–37. [DOI] [PubMed] [Google Scholar]

[CR42] 42.Wang J, Yang Y, Fan L, Xu J, Li C, Liu Y, et al. Convergent functional architecture of the superior parietal lobule unraveled with multimodal neuroimaging approaches: parcellation of superior parietal lobule. Hum Brain Mapp. 2015;36:238–57. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Pretzsch CM, Schäfer T, Lombardo MV, Warrier V, Mann C, Bletsch A, et al. Neurobiological correlates of change in adaptive behavior in autism. Am J Psychiatry. 2022;179:336–49. [DOI] [PubMed] [Google Scholar]

[CR44] 44.Husain M, Parton A, Hodgson TL, Mort D, Rees G. Self-control during response conflict by human supplementary eye field. Nat Neurosci. 2003;6:117–8. [DOI] [PubMed] [Google Scholar]

[CR45] 45.Armstrong KM, Chang MH, Moore T. Selection and maintenance of spatial information by frontal eye field neurons. J Neurosci. 2009;29:15621–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Bedini M, Olivetti E, Avesani P, Baldauf D. Accurate localization and coactivation profiles of the frontal eye field and inferior frontal junction: an ALE and MACM fMRI meta-analysis. Brain Struct Funct. 2023;228:997–1017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR47] 47.Song F, Dong X, Zhao J, Wang J, Sang X, He X, et al. Causal role of the frontal eye field in attention-induced ocular dominance plasticity. eLife. 2024;12:e93213. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Dadario NB, Sughrue ME. The functional role of the precuneus. Brain. 2023;146:3598–607. [DOI] [PubMed] [Google Scholar]

[CR49] 49.Alahmadi AAS. Investigating the sub-regions of the superior parietal cortex using functional magnetic resonance imaging connectivity. Insights Imaging. 2021;12:47. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.Baron-Cohen S, Lombardo MV. Autism and talent: the cognitive and neural basis of systemizing. Dialogues Clin Neurosci. 2017;19:345–53. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR51] 51.Rogers JC, De Brito SA. Cortical and subcortical gray matter volume in youths with conduct problems: a meta-analysis. JAMA Psychiatry. 2016;73:64–72. [DOI] [PubMed] [Google Scholar]

[CR52] 52.Gao Y, Jiang Y, Ming Q, Zhang J, Ma R, Wu Q, et al. Neuroanatomical changes associated with conduct disorder in boys: influence of childhood maltreatment. Eur Child Adolesc Psychiatry. 2022;31:601–13. [DOI] [PubMed] [Google Scholar]

[CR53] 53.Guo X, Zhai G, Liu J, Cao Y, Zhang X, Cui D, et al. Inter-individual heterogeneity of functional brain networks in children with autism spectrum disorder. Mol Autism. 2022;13:52. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Constantino JN, Todd RD. Autistic traits in the general population: a twin study. Arch Gen Psychiatry. 2003;60:524–30. [DOI] [PubMed] [Google Scholar]

[CR55] 55.Rashid B, Blanken LME, Muetzel RL, Miller R, Damaraju E, Arbabshirani MR, et al. Connectivity dynamics in typical development and its relationship to autistic traits and autism spectrum disorder. Hum Brain Mapp. 2018;39:3127–42. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR56] 56.Degré-Pelletier J, Danis É, Thérien VD, Bernhardt B, Barbeau EB, Soulières I. Differential neural correlates underlying visuospatial versus semantic reasoning in autistic children. Cereb Cortex. 2024;34:19–29. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR57] 57.Itahashi T, Yamada T, Watanabe H, Nakamura M, Ohta H, Kanai C, et al. Alterations of local spontaneous brain activity and connectivity in adults with high-functioning autism spectrum disorder. Mol Autism. 2015;6:30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Agam Y, Joseph RM, Barton JJS, Manoach DS. Reduced cognitive control of response inhibition by the anterior cingulate cortex in autism spectrum disorders. NeuroImage. 2010;52:336–47. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Carter Leno V, Chandler S, White P, Yorke I, Charman T, Jones CRG, et al. Associations between theory of mind and conduct problems in autistic and nonautistic youth. Autism Res. 2021;14:276–88. [DOI] [PubMed] [Google Scholar]

[CR60] 60.Cheng W, Rolls ET, Gu H, Zhang J, Feng J. Autism: reduced connectivity between cortical areas involved in face expression, theory of mind, and the sense of self. Brain. 2015;138:1382–93. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.Xu J, Wang C, Xu Z, Li T, Chen F, Chen K, et al. Specific functional connectivity patterns of middle temporal gyrus subregions in children and adults with autism spectrum disorder. Autism Res. 2020;13:410–22. [DOI] [PubMed] [Google Scholar]

[CR62] 62.Cooke DF, Graziano MSA. Sensorimotor integration in the precentral gyrus: polysensory neurons and defensive movements. J Neurophysiol. 2004;91:1648–60. [DOI] [PubMed] [Google Scholar]

[CR63] 63.Gogolla N. The insular cortex. Curr Biol. 2017;27:580–6. [DOI] [PubMed] [Google Scholar]

[CR64] 64.Tononi G. Consciousness and complexity. Science. 1998;282:1846–51. [DOI] [PubMed] [Google Scholar]

[CR65] 65.Edelman GM, Gally JA. Degeneracy and complexity in biological systems. Proc Natl Acad Sci. 2001;98:13763–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR66] 66.Albantakis L, Bernard C, Brenner N, Marder E, Narayanan R. The Brain’s best kept secret is its degenerate structure. J Neurosci. 2024;44:e1339242024. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR67] 67.Lake EMR, Finn ES, Noble SM, Vanderwal T, Shen X, Rosenberg MD, et al. The functional brain organization of an individual allows prediction of measures of social abilities transdiagnostically in autism and Attention-Deficit/Hyperactivity disorder. Biol Psychiatry. 2019;86:315–26. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR68] 68.Geschwind DH, Levitt P. Autism spectrum disorders: developmental disconnection syndromes. Curr Opin Neurobiol. 2007;17:103–11. [DOI] [PubMed] [Google Scholar]

[CR69] 69.Thomas Yeo BT, Krienen FM, Sepulcre J, Sabuncu MR, Lashkari D, Hollinshead M, et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J Neurophysiol. 2011;106:1125–65. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR70] 70.Bishop DVM, Norbury CF. Executive functions in children with communication impairments, in relation to autistic symptomatology: 2: response inhibition. Autism. 2005;9:29–43. [DOI] [PubMed] [Google Scholar]

[CR71] 71.Mosconi MW, Kay M, D’Cruz A-M, Seidenfeld A, Guter S, Stanford LD, et al. Impaired inhibitory control is associated with higher-order repetitive behaviors in autism spectrum disorders. Psychol Med. 2009;39:1559–66. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR72] 72.Geurts HM, Van Den Bergh SFWM, Ruzzano L. Prepotent response inhibition and interference control in autism spectrum disorders: two meta-analyses. Autism Res. 2014;7:407–20. [DOI] [PubMed] [Google Scholar]

[CR73] 73.Schmitt LM, White SP, Cook EH, Sweeney JA, Mosconi MW. Cognitive mechanisms of inhibitory control deficits in autism spectrum disorder. J Child Psychol Psychiatry. 2018;59:586–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR74] 74.Voorhies W, Dajani DR, Vij SG, Shankar S, Turan TO, Uddin LQ. Aberrant functional connectivity of inhibitory control networks in children with autism spectrum disorder. Autism Res. 2018;11:1468–78. [DOI] [PubMed] [Google Scholar]

[CR75] 75.De Lacy N, Doherty D, King BH, Rachakonda S, Calhoun VD. Disruption to control network function correlates with altered dynamic connectivity in the wider autism spectrum. NeuroImage Clin. 2017;15:513–24. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR76] 76.Wang L, Qin Y, Yang S, Jin D, Zhu Y, Li X, et al. Posterior default mode network is associated with the social performance in male children with autism spectrum disorder: a dynamic causal modeling analysis based on triple-network model. Hum Brain Mapp. 2024;45:e26750. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR77] 77.Durston S, Thomas KM, Yang Y, Uluğ AM, Zimmerman RD, Casey BJ. A neural basis for the development of inhibitory control. Dev Sci. 2002;5:9–16. [Google Scholar]

[CR78] 78.Schmitz N, Rubia K, Daly E, Smith A, Williams S, Murphy DGM. Neural correlates of executive function in autistic spectrum disorders. Biol Psychiatry. 2006;59:7–16. [DOI] [PubMed] [Google Scholar]

[CR79] 79.Levy BJ, Wagner AD. Cognitive control and right ventrolateral prefrontal cortex: reflexive reorienting, motor inhibition, and action updating. Ann N Y Acad Sci. 2011;1224:40–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR80] 80.Norman LJ, Sudre G, Bouyssi-Kobar M, Jiao M, Gligorovic S, Jean J, et al. Cross-sectional mega-analysis of resting-state alterations associated with autism and attention-deficit/hyperactivity disorder in children and adolescents. Nat Ment Health. 2025;3:709–23. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR81] 81.Adams NC, Jarrold C. Inhibition in autism: children with autism have difficulty inhibiting irrelevant distractors but not prepotent responses. J Autism Dev Disord. 2012;42:1052–63. [DOI] [PubMed] [Google Scholar]

[CR82] 82.Voldsbekk I, Kjelkenes R, Frogner ER, Westlye LT, Alnæs D. Testing the sensitivity of diagnosis-derived patterns in functional brain networks to symptom burden in a Norwegian youth sample. Hum Brain Mapp. 2024;45:e26631. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR83] 83.Wang X, Auyeung B, Pan N, Lin L-Z, Chen Q, Chen J-J, et al. Empathy, theory of mind, and prosocial behaviors in autistic children. Front Psychiatry. 2022;13:844578. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR84] 84.Guo X, Xu C, Chen J, Wu Z, Hou S, Wei Z. Disrupted cognitive and affective empathy network interactions in autistic children viewing social animation. Soc Cogn Affect Neurosci. 2024;19:nsae028. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR85] 85.Balsters JH, Apps MAJ, Bolis D, Lehner R, Gallagher L, Wenderoth N. Disrupted prediction errors index social deficits in autism spectrum disorder. Brain. 2017;140:235–46. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR86] 86.Gu X, Eilam-Stock T, Zhou T, Anagnostou E, Kolevzon A, Soorya L, et al. Autonomic and brain responses associated with empathy deficits in autism spectrum disorder. Hum Brain Mapp. 2015;36:3323–38. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR87] 87.Loomes R, Hull L, Mandy WPL. What is the male-to-female ratio in autism spectrum disorder? A systematic review and meta-analysis. J Am Acad Child Adolesc Psychiatry. 2017;56:466–74. [DOI] [PubMed] [Google Scholar]

[CR88] 88.Lee JK, Amaral DG, Solomon M, Rogers SJ, Ozonoff S, Nordahl CW. Sex differences in the amygdala resting-state connectome of children with autism spectrum disorder. Biol Psychiatry Cogn Neurosci Neuroimaging. 2020;5:320–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR89] 89.Walsh MJM, Wallace GL, Gallegos SM, Braden BB. Brain-based sex differences in autism spectrum disorder across the lifespan: a systematic review of structural MRI, fMRI, and DTI findings. NeuroImage Clin. 2021;31:102719. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR90] 90.Zielinski BA, Andrews DS, Lee JK, Solomon M, Rogers SJ, Heath B, et al. Sex-dependent structure of socioemotional salience, executive control, and default mode networks in preschool-aged children with autism. NeuroImage. 2022;257:119252. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR91] 91.Han S, Jia H, Vilaseca G, Vilaró N, Duan F, Sun Z, et al. Revealing structural brain-cognition relationships in children: a comparison of morphometric similarity and INverse divergence networks. Neuroinformatics. 2026;24:4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR92] 92.Gao J, Hu Y. MIND versus MSN: a systematic evaluation of test–retest reliability and age sensitivity for T1-weighted structural similarity networks. Netw Neurosci. 2026. [DOI] [PMC free article] [PubMed]

PERMALINK

Aberrant multivariate mapping between behavioral profiles and cortical morphological brain networks in children with autism spectrum disorder

Yuqing Liao

Ning Pan

Xiaofan Qiu

Jin Jing

Lizi Lin

Xin Wang

Jinhui Wang

Abstract

Introduction

Methods and materials

Participants

MRI data acquisition and preprocessing

Morphological brain network construction

Sparse canonical correlation analysis

Results

Demographic and behavioral characteristics

Table 1.

MC-behavior associations in the TD group

Fig. 1. sCCA modes in the TD group that are derived based on the 1000 top-ranked MC features.

Fig. 2. Contributions of behavioral features to the sCCA modes in the TD group derived from group-specific 1000 top ranked MC features and cross-group projection.

Cross-group projection

MC-behavior associations in the ASD group

Fig. 3. sCCA modes in the ASD group that are derived based on the 1000 top-ranked MC features.

MC-behavior associations in a unified framework of MC features

Fig. 4. MC features common to the TD and ASD groups that significantly contribute to the autistic trait mode derived based on the assembly of the group-specific 1000 top-ranked MC features.

Fig. 5. MC features unique to each group that significantly contribute to the autistic trait mode derived based on the assembly of the group-specific 1000 top-ranked MC features.

Between-group differences in MC features

Validation results

Discussion

MC-behavior association modes in TD children

Altered MC-behavior associations in children with ASD

Divergent neural substrates of the autistic trait mode between children with ASD and TD children

Absence of the prosociality mode in children with ASD

Limitations

Conclusion

Supplementary information

Acknowledgements

Author contributions

Funding

Data availability

Code availability

Ethics declarations

Competing interests

Footnotes

Contributor Information

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases