Group-mean comparisons often identify atypical functional connectivity in autism, but it remains unclear whether these findings consistently manifest at the individual level. Here we use normative modeling to quantify the interindividual heterogeneity of atypical functional connectivity across multiple brain scales using multicenter resting-state functional magnetic resonance imaging data from 1,824 participants (796 autistic individuals and 1,028 neurotypical controls) in a cross-sectional study across 32 sites. We find that no single functional connectivity estimate showed extreme deviation from normative expectations in more than 4% of people in either group. However, these deviations converged on common regions and networks in autistic people, who showed up to double the level of overlap compared with controls. Specifically, autistic participants demonstrated convergent hypoconnectivity in sensorimotor and attention regions and convergent hyperconnectivity between frontoparietal and default mode networks. Functional connectivity deviation patterns significantly predicted social and cognitive abilities. These findings demonstrate that autism exhibits scale-dependent heterogeneity, characterized by normative variability at the connection level but significant convergence at regional and network scales. These convergent regions and networks may be used to identify targets for individualized therapeutic development.

Dreaming represents a complex and universal aspect of human sleep, yet it remains an intriguing phenomenon, with the neural mechanisms underlying dream experiences and their frequency not fully understood. This study employs a multimodal neuroimaging approach, integrating quantitative multi-parameter mapping, diffusion tensor imaging, and resting-state functional MRI, to investigate the neural correlates of dream recall frequency (DRF) in a large cohort of 258 healthy individuals. By employing Linked Independent Component Analysis (LICA), we were able to discern distinctive patterns of brain structure and function that correlated with variations in DRF. Our findings elucidate a complex relationship between dream recall and brain microstructure integrity, particularly in white matter regions of the orbitofrontal cortex, parahippocampal gyrus, superior parietal lobule, and occipital cortex. Higher DRF was related to increased white matter microstructure integrity in these regions and decreased gray matter volume in occipital and temporal areas. In terms of functional measures, higher DRF was associated with reduced connectivity across a range of resting-state networks, including the default mode, visual, and dorsal attention networks. This was particularly evident in the right precuneus and posterior cingulate cortex. These results suggest that enhanced dream recall may be related to the organization of higher-order visual and cognitive processing areas, supporting a top-down model of dreaming. This study contributes to a more comprehensive understanding of the neural substrates underlying individual differences in dream recall, offering a foundation for future investigations into the neurobiology and causal relationships of dreaming. ### Competing Interest Statement The authors have declared no competing interest.

BACKGROUND:Interindividual brain differences likely precede the emergence of mood and anxiety disorders; however, the specific brain alterations remain unclear. While many studies focus on a single imaging modality in isolation, recent advances in multimodal image analysis allow for a more comprehensive understanding of the complex neurobiology that underlies mental health. METHODS:In a large population-based cohort of children from the ABCD (Adolescent Brain Cognitive Development) Study (N > 10,000), we applied data-driven linked independent component analysis to identify linked variations in cortical structure and white matter microstructure that together predict longitudinal behavioral and mental health symptoms. Brain differences were examined in a subsample of twins depending on the presence of at-risk behaviors. RESULTS:Two multimodal brain signatures at ages 9 to 10 years predicted longitudinal mental health symptoms from 9 to 12 years, with small effect sizes. Cortical variations in association, limbic, and default mode regions linked with peripheral white matter microstructure together predicted higher depression and anxiety symptoms across 2 independent split-halves. The brain signature differed between depression and anxiety symptom trajectories and related to emotion regulation network functional connectivity. Linked variations of subcortical structures and projection tract microstructure variably predicted behavioral inhibition, sensation seeking, and psychosis symptom severity over time in male participants. These brain patterns were significantly different between pairs of twins discordant for self-injurious behavior. CONCLUSIONS:Our results demonstrate reliable, multimodal brain patterns in childhood, before mood and anxiety disorders tend to emerge, that lay the foundation for long-term mental health outcomes and offer targets for early identification of children at risk.

To support self-generating cognition and behaviour, neural communication must be highly flexible and dynamic, but also structured. While whole-brain fMRI measures have revealed robust yet changing patterns of statistical dependencies between regions, it is unclear whether these statistical patterns -referred to as functional connectivity (FC)- can reflect dynamic large-scale communication in a way that is relevant to human cognition; or just reflect, for example, homeostatic processes. For FC to reflect cognition, and therefore actual communication, we propose that at least three conditions must be met: it must span sufficient temporal complexity to support cognition's demands while being highly organized so that the system behaves reliably; it must be able to adjust to behavioural circumstances; and it must exhibit fluctuations at timescales compatible with cognition' timescales. We trained multiple models of time-varying FC on fMRI data from the Human Connectome Project across three behavioural conditions: at rest, during a working memory task, and a motor task; and characterised them using Principal Component Analysis. We show that FC follows low- yet multi-dimensional trajectories that can be reliably measured, and that these trajectories can satisfy the aforementioned requirements. Although these are necessary but not sufficient conditions, it remains possible that time-varying FC may potentially index key aspects of neural communication underlying cognitive function.

Atypical face processing is commonly reported in autism. Its neural correlates have been explored extensively across single neuroimaging modalities within key regions of the face processing network, such as the fusiform gyrus (FFG). Nonetheless, it is poorly understood how variation in brain anatomy and function jointly impacts face processing and social functioning. Here we leveraged a large multimodal sample to study the cross-modal signature of face processing within the FFG across four imaging modalities (structural magnetic resonance imaging (MRI), resting-state functional magnetic resonance imaging, task-functional magnetic resonance imaging and electroencephalography) in 204 autistic and nonautistic individuals aged 7–30 years (case–control design). We combined two methodological innovations—normative modeling and linked independent component analysis—to integrate individual-level deviations across modalities and assessed how multimodal components differentiated groups and informed social functioning in autism. Groups differed significantly in a multimodal component driven by bilateral resting-state functional MRI, bilateral structure, right task-functional MRI and left electroencephalography loadings in face-selective and retinotopic FFG. Multimodal components outperformed unimodal ones in differentiating groups. In autistic individuals, multimodal components were associated with cognitive and clinical features linked to social, but not nonsocial, functioning. These findings underscore the importance of elucidating multimodal neural associations of social functioning in autism, offering potential for the identification of mechanistic and prognostic biomarkers. The authors leveraged a large multimodal sample and combined normative modeling and linked independent component analysis to study a cross-modal signature of face processing within the fusiform gyrus in autism.

Traditional group-level fMRI analysis approaches, such as Independent Component Analysis (ICA), often rely on unsupervised dimensionality reduction to map subjects into a common feature space. While effective for capturing common variance across all subjects, the preservation of discriminative features between groups of participants is not guaranteed. To address this limitation, we introduce Independent Filter Analysis (IFA), a supervised extension of group ICA that explicitly models group-discriminative information as part of the dimensionality reduction steps. Prior to unmixing, IFA constructs a subspace that simultaneously retains both shared and group-specific information, enhancing sensitivity to group effects while preserving biological interpretability. We validated IFA using simulated data and paired condition comparisons from three Human Connectome Project (HCP) tasks. In the simulation, IFA achieved 95% classification accuracy, outperforming group ICA, which failed to detect subtle group differences. On the HCP data, IFA increased network matrix classification accuracy by up to 15% and produced spatial maps that more precisely reflected task-relevant differences.

In functional magnetic resonance imaging, multivariate proxies of functional brain networks are commonly extracted using spatial independent component analysis. The theoretical premises of spatial overlap among functional processes and the time-varying nature of functional connectivity prompt the question of how to accurately model spatially overlapping and time-varying functional sources. Well-known functional networks have previously been shown to divide into spatially overlapping and functionally distinct subprocesses termedTemporal Functional Modes (TFM)using temporal independent component analysis on the time courses obtained via spatial independent component analysis. In this model, spatial and temporal modes of organisation interact through a single mixing matrix with fixed coefficients. Here, we introduce a time-resolved version termedTime-Resolved Instantaneous Functional Loci Estimation (TRIFLE)to estimate time-varying changes in source allocation. We analytically demonstrate that the originally fixed TFM mixing matrix can be expressed as the temporal average of a time-resolved mixing matrix, which in turn can be obtained in closed form and provides instantaneous estimates of brain network reconfigurations involved in distinct temporal functional modes. We apply TRIFLE to a high-temporal resolution functional magnetic resonance imaging dataset. We demonstrate that spatial source allocation aligns with expectations based on the experimental task design and that successful and unsuccessful trials have different allocation profiles. The proposed method sheds light on the temporal evolution of brain network reconfigurations while explicitly accounting for potential neuroanatomical overlap.

Brain structure and function have largely been studied separately in relation to the neurobiology and genetics of language. Here we used linked independent component analysis to integrate language network functional connectivity with brain volumetric and white matter structure in 32,677 UK Biobank participants, followed by analysis of behavioural, neurobiological and genetic correlates of the derived multimodal structure-function imaging components. Stronger functional connectivity between brain language areas was associated with increased volume of parts of the cerebellum and motor cortex, together with smaller ventricles and sensory parietal and occipital areas. The brain structure-function language components mediated an association between vocabulary level and polygenic scores for reading ability. We report 18 genomic loci associated with brain structure-function language components. Single-nucleotide polymorphism (SNP)-based heritability estimates for these components were 23-30%, and there was significant enrichment of heritability in primate-conserved genomic loci and fetal brain human-gained enhancer elements. This study revealed that structural correlates of functional language network connectivity extend well beyond previously defined language areas of the brain, and highlights the value of multimodal brain phenotyping for human neurogenetic discovery.

Wealth inequality is one of the most profound challenges confronting society today. However, an important issue in addressing inequality lies in formalizing the diversity of individual perspectives regarding what constitutes a fair distribution of resources. We tackle this topic by simulating wealth inequality through the allocation of bonus endowments in both Dictator Game (DG) and Ultimatum Game (UG) settings and capturing distributive decisions. By integrating a computational model, we quantify individual differences in the interplay between financial self-interest and competing pro-social motivations that emerge in the context of pre-existing wealth inequity. Our behavioral results show that, on average, pre-existing wealth influences distributive preferences across both allocations and proposals. Yet, inequality elicits non-uniform fairness concerns. Using a hierarchical clustering approach, we objectively categorise participants' behavior elucidating four distinct decision strategies: 'Pro-Self', 'Table Egalitarianism', 'Total Egalitarianism', and 'Moral Opportunism'. A balanced distribution of strategies is observed during allocations (DG), whereas Table Egalitarianism prevails in strategic proposals (UG), highlighting the influence of strategic considerations on decision strategy. Furthermore, we demonstrate an association between strategies across decision contexts. Our findings thus contribute a principled framework to formalize distributive preferences, revealing that, with respect to both altruistic allocations and strategic proposals, competing ideals of fairness underlie distributive preferences under wealth inequality.

Wealth inequality often leads to heated debates about resource distribution. An important challenge in addressing polarisation and economic disparities therefore lies in formalising the diversity of individual perspectives regarding what constitutes a fair distribution of resources. We tackle this issue by simulating wealth inequality through the allocation of bonus endowments to participants in both Dictator Game (DG) and Ultimatum Game (UG) settings and capturing distributive decisions. By integrating a computational model, we further quantify individual differences in the interplay between financial self-interest and competing pro-social motivations that emerge in the context of pre-existing wealth inequity. Our behavioural results show that, on average, wealth influences distributive preferences across allocations and proposals, yet existing inequality elicits non-uniform fairness concerns among individuals. Using a hierarchical clustering approach, we objectively categorise participants' behaviour, offering a systematic framework that elucidates four distinct decision strategies: ‘Pro-Self’, ‘Table Egalitarianism’, ‘Total Egalitarianism’, and ‘Moral Opportunism’. A relatively balanced distribution of strategies is observed during allocations, whereas Table Egalitarianism prevails in strategic proposals, highlighting the significant influence of strategic considerations on chosen decision strategy. Furthermore, we demonstrate an association between strategies between decision contexts. Our findings thus contribute a principled framework to formalise the individualised nature of distributive preferences, revealing that, with respect to both altruistic allocations and strategic proposals, competing ideals of fairness underlie the issue of resource distribution under wealth inequality.

Finding an interpretable and compact representation of complex neuroimaging data is extremely useful for understanding brain behavioral mapping and hence for explaining the biological underpinnings of mental disorders. However, hand-crafted representations, as well as linear transformations, may inadequately capture the considerable variability across individuals. Here, we implemented a data-driven approach using a three-dimensional autoencoder on two large-scale datasets. This approach provides a latent representation of high-dimensional task-fMRI data which can account for demographic characteristics whilst also being readily interpretable both in the latent space learned by the autoencoder and in the original voxel space. This was achieved by addressing a joint optimization problem that simultaneously reconstructs the data and predicts clinical or demographic variables. We then applied normative modeling to the latent variables to define summary statistics (‘latent indices’) and establish a multivariate mapping to non-imaging measures. Our model, trained with multi-task fMRI data from the Human Connectome Project (HCP) and UK biobank task-fMRI data, demonstrated high performance in age and sex predictions and successfully captured complex behavioral characteristics while preserving individual variability through a latent representation. Our model also performed competitively with respect to various baseline models including several variants of principal components analysis, independent components analysis and classical regions of interest, both in terms of reconstruction accuracy and strength of association with behavioral variables.

Background Autism spectrum disorder (henceforth autism) is a complex neurodevelopmental condition associated with differences in gray matter (GM) volume covariations, as reported in our previous study of the Longitudinal European Autism Project (LEAP) data. To make progress on the identification of potential neural markers and to validate the robustness of our previous findings, we aimed to replicate our results using data from the Enhancing Neuroimaging Genetics Through Meta-Analysis (ENIGMA) autism working group. Methods We studied 781 autistic and 927 non-autistic individuals (6–30 years, IQ ≥ 50), across 37 sites. Voxel-based morphometry was used to quantify GM volume as before. Subsequently, we used spatial maps of the two autism-related independent components (ICs) previously identified in the LEAP sample as templates for regression analyses to separately estimate the ENIGMA-participant loadings to each of these two ICs. Between-group differences in participants’ loadings on each component were examined, and we additionally investigated the relation between participant loadings and autistic behaviors within the autism group. Results The two components of interest, previously identified in the LEAP dataset, showed significant between-group differences upon regressions into the ENIGMA cohort. The associated brain patterns were consistent with those found in the initial identification study. The first IC was primarily associated with increased volumes of bilateral insula, inferior frontal gyrus, orbitofrontal cortex, and caudate in the autism group relative to the control group ( β = 0.129, p = 0.013). The second IC was related to increased volumes of the bilateral amygdala, hippocampus, and parahippocampal gyrus in the autism group relative to non-autistic individuals ( β = 0.116, p = 0.024). However, when accounting for the site-by-group interaction effect, no significant main effect of the group can be identified ( p > 0.590). We did not find significant univariate association between the brain measures and behavior in autism ( p > 0.085). Limitations The distributions of age, IQ, and sex between LEAP and ENIGMA are statistically different from each other. Owing to limited access to the behavioral data of the autism group, we were unable to further our understanding of the neural basis of behavioral dimensions of the sample. Conclusions The current study is unable to fully replicate the autism-related brain patterns from LEAP in the ENIGMA cohort. The diverse group effects across ENIGMA sites demonstrate the challenges of generalizing the average findings of the GM covariation patterns to a large-scale cohort integrated retrospectively from multiple studies. Further analyses need to be conducted to gain additional insights into the generalizability of these two GM covariation patterns.

ABSTRACTAtypical functional connectivity (FC) in autism is a common finding, but the results of individual studies are often inconsistent and sometimes contradictory. Classical reliance on case-control comparisons of group means that ignore the inter-individual heterogeneity in autism may be a key drive of this inconsistency. Here, we used normative modelling to examine FC heterogeneity at the level of pair-wise inter-regional connections, specific brain regions, and broader functional networks in 1,824 participants (796 autistic) aged 5-58 years recruited across 32 different sites. Connection-level heterogeneity was high in both groups, with no single connection deviating in more than 4% of participants. However, deviant connections tended to converge on common regions and networks in autistic individuals more than in controls. Autistic individuals showed significantly greater overlap for positive deviations (i.e., atypically increased FC) in transmodal systems and negative deviations (atypically decreased FC) in sensory-motor areas. FC deviation patterns across coarser levels correlated with social functioning symptoms and intellectual ability. This work suggests that clinical variability in autism may be associated with extreme heterogeneity in the specific functional connections, whereas commonalities may be driven by convergence of atypical FC increases in transmodal systems and atypical decreases in sensorimotor networks, pointing to an imbalance in the functional organization of the brain’s sensorimotor-association axis.

Understanding brain-behaviour correlates is challenging. Multimodal approaches increase the sensitivity to brain-behaviour relationships. Here we integrated grey matter density maps, structural connectivity metrics and functional connectopic (gradient) maps from 676 participants of the HCP, with the hypothesis that these more advanced measures of both structural and functional organisation would improve sensitivity to brain-behaviour associations. Our results show 1 component significantly associated with various behavioural measures including multiple measures of alcohol use. This component was composed of structural modalities. There was little shared variance between structural measures and functional gradients across the components.

In line with the Research Domain Criteria (RDoC) , we set out to investigate the brain basis of psychopathology within a transdiagnostic, dimensional framework. We performed an integrative structural-functional linked independent component analysis to study the relationship between brain measures and a broad set of biobehavioral measures in a sample (n = 295) with both mentally healthy participants and patients with diverse non-psychotic psychiatric disorders (i.e. mood, anxiety, addiction, and neurodevelopmental disorders). To get a more complete understanding of the underlying brain mechanisms, we used gray and white matter measures for brain structure and both resting-state and stress scans for brain function. The results emphasize the importance of the executive control network (ECN) during the functional scans for the understanding of transdiagnostic symptom dimensions. The connectivity between the ECN and the frontoparietal network in the aftermath of stress was correlated with symptom dimensions across both the cognitive and negative valence domains, and also with various other health-related biological and behavioral measures. Finally, we identified a multimodal component that was specifically associated with the diagnosis of autism spectrum disorder (ASD). The involvement of the default mode network, precentral gyrus, and thalamus across the different modalities of this component may reflect the broad functional domains that may be affected in ASD, like theory of mind, motor problems, and sensitivity to sensory stimuli, respectively. Taken together, the findings from our extensive, exploratory analyses emphasize the importance of a dimensional and more integrative approach for getting a better understanding of the brain basis of psychopathology.

Sensory atypicalities are particularly common in autism spectrum disorders (ASD). Nevertheless, our knowledge about the divergent functioning of the underlying somatosensory region and its association with ASD phenotype features is limited. We applied a data-driven approach to map the fine-grained variations in functional connectivity of the primary somatosensory cortex (S1) to the rest of the brain in 240 autistic and 164 neurotypical individuals from the EU-AIMS LEAP dataset, aged between 7 and 30. We estimated the S1 connection topography ('connectopy') at rest and during the emotional face-matching (Hariri) task, an established measure of emotion reactivity, and accessed its association with a set of clinical and behavioral variables. We first demonstrated that the S1 connectopy is organized along a dorsoventral axis, mapping onto the S1 somatotopic organization. We then found that its spatial characteristics were linked to the individuals' adaptive functioning skills, as measured by the Vineland Adaptive Behavior Scales, across the whole sample. Higher functional differentiation characterized the S1 connectopies of individuals with higher daily life adaptive skills. Notably, we detected significant differences between rest and the Hariri task in the S1 connectopies, as well as their projection maps onto the rest of the brain suggesting a task-modulating effect on S1 due to emotion processing. All in all, variation of adaptive skills appears to be reflected in the brain's mesoscale neural circuitry, as shown by the S1 connectivity profile, which is also differentially modulated during rest and emotional processing.

Recent tractography and microdissection studies have shown that the left arcuate fasciculus (AF)—a fiber tract thought to be crucial for speech production—consists of a minimum of 2 subtracts directly connecting the temporal and frontal cortex. These subtracts link the posterior superior temporal gyrus (STG) and middle temporal gyrus (MTG) to the inferior frontal gyrus. Although they have been hypothesized to mediate different functions in speech production, direct evidence for this hypothesis is lacking. To functionally segregate the 2 AF segments, we combined functional magnetic resonance imaging with diffusion-weighted imaging and probabilistic tractography using 2 prototypical speech production tasks, namely spoken pseudoword repetition (tapping sublexical phonological mapping) and verb generation (tapping lexical-semantic mapping). We observed that the repetition of spoken pseudowords is mediated by the subtract of STG, while generating an appropriate verb to a spoken noun is mediated by the subtract of MTG. Our findings provide strong evidence for a functional dissociation between the AF subtracts, namely a sublexical phonological mapping by the STG subtract and a lexical-semantic mapping by the MTG subtract. Our results contribute to the unraveling of a century-old controversy concerning the functional role in speech production of a major fiber tract involved in language.

Neuroimaging analyses of brain structure and function in autism have typically been conducted in isolation, missing the sensitivity gains of linking data across modalities. Here we focus on the integration of structural and functional organisational properties of brain regions. We aim to identify novel brain-organisation phenotypes of autism. We utilised multimodal MRI (T1-, diffusion-weighted and resting state functional), behavioural and clinical data from the EU AIMS Longitudinal European Autism Project (LEAP) from autistic ( n = 206) and non-autistic ( n = 196) participants. Of these, 97 had data from 2 timepoints resulting in a total scan number of 466. Grey matter density maps, probabilistic tractography connectivity matrices and connectopic maps were extracted from respective MRI modalities and were then integrated with Linked Independent Component Analysis. Linear mixed-effects models were used to evaluate the relationship between components and group while accounting for covariates and non-independence of participants with longitudinal data. Additional models were run to investigate associations with dimensional measures of behaviour. We identified one component that differed significantly between groups (coefficient = 0.33, p adj = 0.02). This was driven (99%) by variance of the right fusiform gyrus connectopic map 2. While there were multiple nominal (uncorrected p < 0.05) associations with behavioural measures, none were significant following multiple comparison correction. Our analysis considered the relative contributions of both structural and functional brain phenotypes simultaneously, finding that functional phenotypes drive associations with autism. These findings expanded on previous unimodal studies by revealing the topographic organisation of functional connectivity patterns specific to autism and warrant further investigation.