Neuroanatomical and Functional Dimensionality of Reinforcement Sensitivity Theory Systems: An Exploratory Graph Analysis Approach

Neuroanatomical and Functional Dimensionality of Reinforcement Sensitivity Theory Systems: An Exploratory Graph Analysis Approach | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Neuroanatomical and Functional Dimensionality of Reinforcement Sensitivity Theory Systems: An Exploratory Graph Analysis Approach Oscar Perez-Diaz, Elena Lacomba-Arnau, Agustín Martínez-Molina, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9179108/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 7 You are reading this latest preprint version Abstract This study evaluates the neurobiological architecture of Reinforcement Sensitivity Theory (RST) using a network psychometric approach. We shift from region-based inference to system-level organization by studying structural and functional brain dimensionality. Using the Human Connectome Project dataset ( N = 1113 structural MRI; N = 176 naturalistic functional MRI), we applied Exploratory Graph Analysis (EGA) to 77 regions of interest from the HCPex atlas contingent to the RST. Bootstrap EGA assessed dimensional stability and item consistency. Structural results identified nine dimensions: the Behavioral Inhibition/Fight-Flight-Freeze (BIS/FFFS) and the Behavioral Approach (BAS) systems, a distinct Constraint Ventral Stream (CVS), and six partially dissociable Constraint Dorsal Stream (CDS) components. In contrast, functional data yielded a more integrated six-dimensional solution. While structural covariance reflects stable anatomical blueprints, functional networks reveal stimulus-driven co-activation, particularly between the BIS/FFFS and BAS. The subdivision of the CDS across both modalities supports a non-unitary control architecture visible only at high parcellation. Our findings demonstrate that RST systems emerge as coordinated, reproducible patterns of brain organization, providing a cumulative framework for linking psychological constructs to biological phenotypes in psychometric neuroscience. Exploratory Graph Analysis Reinforcement Sensitivity Theory Magnetic Resonance Imaging Biological Phenotypes Figures Figure 1 Figure 2 INTRODUCTION In behavioral neuroscience, psychological constructs are frequently used to describe and interpret patterns of brain structure and function, even when the relation between psychological constructs and neurobiological measure is not straightforward (Borsboom, 2006 ). This cross-level mapping problem is not merely interpretative, but epistemological. Psychological constructs are typically defined at the level of behavior, affect, or self-report, whereas MRI captures distributed statistical properties of neural organization. Clarifying this mapping is therefore essential for construct validity, interpretability, and cumulative integration across studies. Subsequently, Magnetic Resonance Imaging (MRI) measurements are interpreted as reflecting transient psychological processes or trait-like features of brain organization, despite the fact that these measures may capture complex and distributed properties of brain activation patterns as measures of psychological constructs expressions. Consequently, descriptions of psychological processes and constructs may not accurately reflect activation patterns commonly observed in the human brain (Bolt et al., 2018 ). This practice raises fundamental questions: to what extent do commonly used psychological constructs correspond to identifiable organizational principles in the human brain? What am I really measuring? And, is my interpretation adequate? When researchers conduct an analysis of variance or similar statistical test on MRI measures, and their effects are directly interpreted as transient psychological processes (i.e., working memory) or stable traits (i.e., personality), alternative multidimensional or latent class structures are often neglected (H. F. Golino & Epskamp, 2017 ; Van der Weele & Vansteelandt, 2022 ). This lack of attention to organizational structures of psychological theories or ontologies difficults the integration across tasks, samples, and analytic strategies, thereby constraining cumulative inference. Despite these limitations, the scientific community continues to value MRI indicators as in vivo neurobiological measures of whole-brain function and structure (Laurent et al., 2025 ; Logothetis, 2003 ; Logothetis et al., 2001 ; Rowley et al., 2015 ; Tesler et al., 2023 ), albeit explicitly modeling the dimensional structure of neurobiological systems may facilitate the transition from isolated statistical effects to reproducible principles of brain organization (Poldrack et al., 2017 ; Yarkoni, 2022 ). Equating MRI measures to theoretical attributes, such as working memory processes or personality traits, highlights the perceived proximity between behavioral–psychometric space and network states (Soreq et al., 2021 ). Prior research has suggested that patterns of brain activation may directly reflect latent cognitive abilities (Haslbeck & Waldorp, 2015 , 2020 ; Soreq et al., 2021 ) and can therefore serve as manifest indicators in psychometric models (Cooper et al., 2019 ). Structural MRI (sMRI) morphometry and functional MRI (fMRI) time-series are often used as a form of neurobiological chronometry, not only through subtraction-based methodologies linking brain to behavior across task conditions or group contrast, but also by analogy with reaction time in mental chronometry, thereby rendering Blood Oxygen Level Dependent (BOLD) contrasts conceptually equivalent to other psychological variables (Formisano & Goebel, 2003 ; Menon et al., 1998 ). In neuroimaging research, this inferential leap has been discussed in terms of reverse inference and construct validity: the assumption that activation in a given region necessarily implies engagement of a specific psychological process (Poldrack, 2006 ). Such interpretations are especially problematic for constructs defined at the systems level, where explanatory adequacy depends on coordinated patterns of activity rather than on single regional effects (Pessoa, 2014 ). In this context, the family of methods under the umbrella of network psychometrics (NP) provides an alternative to other traditional multivariate psychometric approaches, such as component extraction techniques (e.g., Principal Component Analysis) or confirmatory factor analysis (Cooper et al., 2019 ). NP explicitly accommodates heterogeneous variable types and supports personalized brain models, showing individual-level reliability comparable to those observed in behavioral measures (Borsboom, 2006 ). NP allows the representation of direct relationships among variables without assuming the existence of latent factors, thus enabling distributed patterns of alteration and interactions that do not necessarily conform to a single factor. At the same time, NP enables the analysis of network properties of latent factors derived from neuroimaging indicators. Within this framework, the brain is conceptualized as a distributed system characterized by non-hierarchical causality, emergent processes, and dynamic interactions. While network models do not by themselves demonstrate causal mechanisms, they offer a structured representation of system-level organization in which constructs are instantiated in patterns of conditional dependence among indicators. As such, NP furnishes a principled interface between theory-defined neurobehavioral systems and the multivariate architecture of neuroimaging data (Borsboom, 2006 ; Schmittmann et al., 2013 ). From this perspective, psychological constructs are conceptualized as complex systems of observable variables that dynamically and mutually reinforce one another. Constructs are thus defined by the causal (bi)directional relations among observed MRI variables rather than being caused by a single underlying latent entity, although latent variables extracted from theoretically grounded indicators can themselves serve as nodes within networks. Thus, rather than being caused by a single latent entity, constructs emerge from the pattern of (bi)directional relationships among observed MRI variables, although theoretically grounded latent variables may themselves be represented as nodes within such networks. The Reinforcement Sensitivity Theory (RST) offers a particularly stringent test case for NP modeling because its explanatory framework is inherently system-based: motivational and affective sensitivities are posited to arise from coordinated neural circuitry with partially overlapping yet functionally differentiated components, subserved by specific cortical and subcortical structures (Kennis et al., 2013 ; McNaughton & Gray, 2000 ). Our prior study explored the dimensionality of brain structural indicators (Lacomba-Arnau et al., 2025 ). That study underscored the critical importance of utilizing dimensionality determination techniques, specifically Parallel Analysis and Exploratory Graph Analysis (EGA), to empirically identify the true number of latent structures in MRI data rather than relying solely on theoretical assumptions. This previous work led to the discovery of a robust 4-factor model: the Behavioral Inhibition System / Fight-Freeze-Flight System (BIS/FFFS); the Behavioral Approach System (BAS); the Constraint Dorsal Stream (CDS); and the Constraint Ventral Stream (CVS). Demonstrating that these theory-defined systems emerge as reproducible dimensions within brain covariance structures would therefore constitute direct support for the neurobiological architecture proposed by the theory. Furthermore, the findings demonstrated the superiority of factorial approaches (such as EFA and ESEM) over Principal Component Analysis (PCA); while PCA conflates systematic and error variance and yielded the least favorable fit indices, factorial methods successfully separate shared variance from measurement error, offering a more precise and biologically valid representation of neurobiological systems. Nonetheless, two questions remained open from that study. First, the Neuromorphometric parcellation did not provide the level of detail necessary for a more refined subcortical parcellation of the BIS/FFFS and the BAS. A more detailed subcortical delineation is theoretically relevant because RST differentiates systems according to their functional roles (e.g., conflict monitoring and defensive responding versus incentive motivation and approach), which rely on partially overlapping yet functionally differentiated subcortical circuitry. Increasing anatomical resolution enables us to evaluate whether the hypothesized system boundaries remain coherent when the measurement framework captures this internal differentiation. Second, we wondered whether that dimensionality would be supported on functional as well as structural covariance. Therefore, we planned to run RST hypothesis driven dimensionality analysis to restrict the interpretation of multivariate covariance analyses to a set of indicators relevant for the theory at test, reducing the alternative explanations at the cost of loss of information. The evaluation of dimensionality or the number of underlying factors is a critical and fundamental step for the validation and understanding of latent structures in multivariate datasets (H. Golino et al., 2020 ), especially in complex domains such as neuroscience. Following the line of research that utilizes fMRI and sMRI indicators to address psychological constructs, as seen in the work by Lacomba-Arnau et al. ( 2025 ) which examines brain latent variables from structural covariance, we continue our interest in testing a methodology capable of handling the complexity and inherent interrelationships of cerebral metrics. In this context, NP has emerged as an advanced quantitative framework (H. Golino et al., 2021 ; H. F. Golino & Epskamp, 2017 ), where variables (e.g., brain parcels or regions of interest) are represented as nodes and their conditional associations as edges (links), creating a system of mutually influencing elements beyond merely reflecting a common latent cause (Christensen et al., 2020 ; H. Golino et al., 2021 ; Schmittmann et al., 2013 ). EGA is a novel technique used within the framework of network psychometrics, following the framework described by Golino and Epskamp ( 2017 ). EGA is designed to estimate the number and content of dimensions accurately and visually by identifying densely connected communities or clusters within the network (Christensen et al., 2023 , 2025 ; H. F. Golino & Epskamp, 2017 ; Kjellström & Golino, 2019 ). EGA operates by first estimating a Gaussian Graphical Model (H. F. Golino & Epskamp, 2017 )—often utilizing the Graphical LASSO (GLASSO) operator to obtain regularized partial correlations—followed by a community detection algorithm, such as Walktrap, which allows latent dimensions to emerge directly from the structure of interconnections among indicators (Christensen et al., 2023 , 2025 ; H. F. Golino & Epskamp, 2017 ; Kjellström & Golino, 2019 ). This approach provides an intuitive visual guide regarding which neuroimaging indicators cluster together, offering a detailed dimensional structure without attending to the rotation or the a priori assumptions of traditional factorial methods. Therefore, the use of EGA may permit an exploratory re-evaluation of the dimensions underlying cerebral covariance patterns for precisely specifying latent factors in agreement with the RST neuropsychological research. In sum, the aim of this study is to test the brain dimensionality of sMRI and fMRI measures in regions of interest (ROIs) involved in the different RST systems, the BAS, the BIS/FFFS and the CVS and CDS. We expect to find a similar dimensional organization of the brain from structural and functional indicators. Specifically, we expect that the ROIs theoretically suggested to gather under the BAS, BIS/FFFS, CVS and CDS will be differentiated in networks of nodes attending to each system. We evaluate whether these ROIs organize into coherent dimensions consistent with the proposed RST systems across structural and functional modalities. While a comparable dimensional structure across modalities would support the presence of shared system-level organization, differences between structural and functional covariance patterns would provide insight into how relatively stable anatomical constraints relate to context-dependent functional expression. Specifically, we expect that ROIs theoretically assigned to the BAS, BIS/FFFS, CVS, and CDS will group/cluster into differentiated networks of nodes corresponding to each system. METHODS Dataset This study uses sMRI and naturalistic fMRI data from the Human Connectome Project (HCP), a publicly available large-scale neuroimaging dataset containing healthy young adults. The sample includes up to 1200 participants (ages 22–35, 54% female) with no history of neurological or psychiatric disorders, recruited from families with twins and siblings to support analyses of heritability and individual variability in brain organization. All participants provided informed consent, and the study was approved by the institutional review board of Washington University in St. Louis. Structural MRI Unprocessed structural data are available for 1113 subjects with imaging data. High-resolution T1-weighted images were acquired using a Siemens 3T Connectome Skyra scanner with the following parameters: repetition time (TR) = 2400 ms, echo time (TE) = 2.14 ms, inversion time = 1000 ms, flip angle = 8°, voxel size = 0.7 mm isotropic, and field of view = 224 × 224 mm. Functional MRI – Naturalistic Paradigm The naturalistic fMRI data used in this study correspond to the HCP 7T Movie-Watching sub-dataset (Finn & Bandettini, 2021 ), which consists of functional acquisitions while participants viewed short audiovisual clips from commercial films and documentaries (1 to 4.3 min in length), which were concatenated and presented in four separate functional runs (total scan duration: 60 min). The movies contained diverse visual stimuli (people, animals, scenes, and objects), actions, sounds, music, speech, linguistic and social communications, and sometimes narratives. There were also 20 s rest periods between the movies. Across the 4 movie watching runs datasets there are 176 common subjects, which were used for later analyses. All data were acquired using a Siemens Magnetom 7 T MRI scanner equipped with a Nova 32-channel receive head coil, following the HCP 7 T imaging protocol. Gradient-echo echo-planar imaging (EPI) sequences were used with the following parameters: TR = 1000 ms, TE = 22.2 ms, flip angle = 45°, voxel size = 1.6 mm isotropic, multiband acceleration factor = 5, in-plane acceleration = 2, and partial Fourier = 7/8. A total of 85 slices were acquired per volume with a bandwidth of 1924 Hz/Px. Phase-encoding direction alternated between posterior–anterior and anterior–posterior across runs to minimize geometric distortions. Each movie-watching run lasted approximately 15 min, with a total viewing time of slightly over one hour across the full session. The stimulus presentation and timing information are fully documented in the HCP data dictionary. Preprocessing Structural MRI Although the HCP provides fully preprocessed structural data, in the present study we used the native T1-weighted images from the HCP Young Adult dataset and performed independent preprocessing using the CAT12 toolbox (version 12.6, r1450; Gaser et al., 2024 ) implemented in SPM12. The preprocessing followed the standard CAT12 “expert mode” pipeline with default parameters unless otherwise specified. The main steps included: (1) Segmentation of the T1-weighted image into gray matter, white matter, and cerebrospinal fluid. (2) Normalization to the MNI152 template using the DARTEL high-dimensional registration approach, preserving regional volume through modulation by the Jacobian determinants of the deformation fields. (3) Bias field correction and affine registration to standard space. (4) Estimation of regional gray matter volume in native space prior to normalization. (5) Quality control by visual inspection and sample homogeneity analysis based on covariance metrics. Functional MRI The fMRI data were distributed preprocessed by the HCP consortium. These preprocessing pipelines are described in detail in Glasser et al. ( 2013 ) and made available through the HCP minimal preprocessing pipelines (Glasser et al., 2013 , https://www.humanconnectome.org/ ). Briefly, the fMRIVolume pipeline performed gradient nonlinearity correction, motion correction, and EPI readout distortion correction. Functional images were co-registered to the subject’s high-resolution structural T1w scan using boundary-based registration and normalized to the standard MNI152 space. Intensity inhomogeneities (bias field) were corrected using the map estimated from the structural processing, and the time series were linearly detrended without aggressive high-pass filtering. All spatial transformations were concatenated and applied in a single spline interpolation step (one-step resampling) to minimize spatial blurring and preserve data quality. ROI extraction Structural MRI CAT12 segmentation included the extraction of mean gray matter volumes from the HCPex atlas (Huang et al., 2022 ), a multimodal parcellation extending the HCP-MMP1 atlas by including 66 additional subcortical areas in volumetric form. The use of the HCPex atlas allowed for direct application to structural T1 data, ensuring full coverage of cortical and subcortical regions relevant to the RST. ROI-wise structural values were exported as CSV files for subsequent dimensional analysis. Additionally, ROI volumes were corrected for the total intracranial volume (TIV) of each subject following the residual method (Nordenskjöld et al., 2015 ), implemented with curve_fit (scipy.optimize) in Python 3.11, to obtain an additional dataset corresponding to corrected brain volumes. Functional MRI Functional time series were extracted from the preprocessed resting-state fMRI data using the fslmeants command from the FSL suite (FMRIB Software Library, Oxford, UK; Smith et al., 2004 ). The extraction was also based on the HCPex volumetric atlas, which includes 426 regions of interest (66 subcortical and 360 cortical areas), registered to the same MNI152 space as the preprocessed data. For each subject, the mean BOLD signal within each ROI was computed across all time points, resulting in one representative time series per region. ROI Selection and Data curation ROI indicators were selected from the HCPex atlas. See Table 1 for the 77 regions selected as indicators for the RST model according to their involvement and overlapping on the previous ROI definition using Neuromorphometrics in Lacomba-Arnau et al. ( 2025 ). Due to the high correlations between homotopic regions (Duboc et al., 2015 ; Mechelli et al., 2005 ), we parceled homotopic ROIs using a combined strategy that is both isolated and content-oriented, averaging each left-right homologue into a single parcel. This approach aligns the unit of analysis with the system-level focus of RST, which is formulated in terms of distributed neurobehavioral systems rather than hemispheric specialization. This idea is also supported by the results of Unique Variable Analysis (UVA), showing a considerable number of redundancies between homotopic regions in sMRI and fMRI, as well as EGA carried with the complete datasets, revealing an average of 99.16%, 98.98% and 91.77%, for the non-TIV corrected (non-cTIV), TIV corrected (cTIV) and functional datasets, respectively (See Online Resource 1). Table 1 Study ROIs. System Label HCPex Region Label HCPex Region BIS Hip Hippocampus PHip_2 ParaHippocampal_Area_2 PreS PreSubiculum PHip_3 ParaHippocampal_Area_3 Entorhinal Entorhinal_Cortex Amyg Amygdala Perirhinal_Ectorhinal Perirhinal_Ectorhinal_Cortex SN Septal_nucleus PHip_1 ParaHippocampal_Area_1 BAS Put Putamen GPi Globus_pallidus_internalis Cau Caudate SNpc Substantia_nigra_pars_compacta NAcc Nucleus_Accumbens SNpr Substantia_nigra_pars_reticulata GPe Globus_pallidus_externalis VTA Ventral_tegmenta_area CDS 23d Area_23d p32 Area_p32 31a Area_31a p32_p Area_p32_prime 31pd Area_31pd 44 Area_44 31p_v Area_31p_ventral 45 Area_45 7m Area_7m 47l Area_47l_(47_lateral) 23ab Area_dorsal_23_a + b Ant_47r Area_anterior_47r TVis Dorsal_Transitional_Visual_Area IFJa Area_IFJa PreC_Visual PreCuneus_Visual_Area IFJp Area_IFJp POS1 Parieto-Occipital_Sulcus_Area_1 IFSa Area_IFSa POS2 Parieto-Occipital_Sulcus_Area_2 IFSp Area_IFSp ProStriate ProStriate_Area Pos_47r Area_posterior_47r RSplenial RetroSplenial_Complex 46 Area_46 Ven_23ab Area_ventral_23_a + b 8Ad Area_8Ad 10r Area_10r 8Av Area_8Av 10v Area_10v Lat_8B Area_8B_Lateral 25 Area_25 8C Area_8C 33_p Area_33_prime 9-46d Area_9-46d 8BM Area_8BM Ant_9 Area_9_anterior 9_Mid Area_9_Middle Pos_9 Area_9_Posterior a24 Area_a24 Ant_9-46v Area_anterior_9-46v Ant_24_p Anterior_24_prime Inf_6–8 Inferior_6–8_Transitional_Area Ant_32_p Area_anterior_32_prime Pos_9-46v_L Area_posterior_9-46v Dor_32 Area_dorsal_32 Sup_6–8 Superior_6–8_Transitional_Area Pos_24 Area_posterior_24 SFL Superior_Frontal_Language_Area Pos_24_p Area_Posterior_24_prime CVS Pos_OFC posterior_OFC_Complex 47m Area_47m s32 Area_s32 47s Area_47s 10d Area_10d Ant_10p Area_anterior_10p Pol_10p Polar_10p OFC Orbital_Frontal_Complex 11l Area_11l Pos_10p Area_posterior_10p 13l Area_13l Furthermore, for the functional data we averaged the signal of each ROI for each subject, obtaining a single average value per ROI for each subject per movie, which were then averaged across the common subjects (176 subjects) across all 4 movies. Thus, we carried the analysis using an average signal matrix of dimensions (ROIs x Subjects). DIMENSIONALITY ANALYSIS The dimensionality assessment of the structural datasets was done using EGA, and Parallel Analysis (PA) as a comparative technique, for both cTIV and non-TIV corrected data. Parallel Analysis PA was undertaken to objectively estimate the optimal number of latent factors in the dataset and validate results from network-based modeling approaches. PA was extracted using principal axis factoring. PA was performed using the RAWPAR function from EGAnet ( R package, version 2.1.0 ), which implements permutation-based factor retention, generating 100 random datasets using the permuted randomization method, preserving the structure and distributional properties of the observed data. Principal Axis Factoring was employed to extract latent factors from both empirical and random datasets. Pearson correlations were used for both empirical and random structures. Factors were retained if observed eigenvalues exceeded the 95th percentile of corresponding randomly generated eigenvalues. Unique Variable Analysis Furthermore, to identify and handle redundant variables in whole-brain neuroimaging data, we applied the UVA procedure implemented in EGAnet. UVA uses the weighted Topological Overlap metric on a network estimated from the input variables to detect sets with strong local dependence, as described by Christensen, Garrido, and Golino (2020). In this case, the weighted Topological Overlap threshold was set as default (0.25). In this study, UVA was applied to the complete dataset (without hemisphere filtering), which predominantly revealed interhemispheric redundancies—that is, marked topological similarity between variables representing homologous regions across both hemispheres. For the structural data UVA revealed 34 redundancies out of which 13 were between homotopic regions, representing a 17% of the total ROIs selected; as for the functional data 18 redundancies were found, with 11 corresponding to homotopic regions (14% of ROIs) (See Online Resource 2). Exploratory Graph Analysis EGA is a novel technique used within the framework of network psychometrics. It was performed on the pre-processed datasets to estimate the dimensional structure of multivariate brain data using network psychometrics, following the framework described by Golino and Epskamp ( 2017 ), Golino et al. ( 2020 ). As established, a Gaussian graphical model was estimated via GLASSO with extended Bayesian information criterion (default gamma = 0.5) to select the optimal regularization parameter. Correlations were determined automatically ( cor_auto ). Community detection was carried out with the Louvain algorithm, applied to the estimated network to identify latent communities (dimensions) within the data using 1000 iterations. The Louvain algorithm was also used to assess unidimensionality as part of standard EGA procedure. Bootstrap Exploratory Graph Analysis To evaluate the stability and replicability of the dimensional solution, Bootstrap Exploratory Graph Analysis (bootEGA) was performed with 500 parametric bootstrap samples, following recommended guidelines (Christensen & Golino, 2021 ). After the bootstrap procedure, two stability metrics were examined: The proportion of times each empirical dimension (community) was exactly replicated across bootstrap samples (structural stability) and the proportion of times each item was assigned to the same dimension as in the empirical EGA solution (item stability). Items with item stability values below 0.75 were considered unstable and were removed from further analysis, except for those corresponding to central regions of the analyzed brain systems (based on theoretical relevance). Thus, we excluded any cortical indicator and subcortical indicator except central for the BIS/FFFS, such as the amygdala and hippocampus, or the BAS, such as the caudate, putamen, accumbens, SN or VTA. Then, the EGA and bootEGA were estimated again. This combined criterion preserved the integrity of core regions while reducing instability from peripheral or low-replicating items. RESULTS Structural data EGA on sMRI data before bootEGA We conducted two separate EGA dimensional analysis on gray matter volume indicators before and after correcting for TIV. The results were highly consistent across conditions, identifying 9 dimensions in both cTIV and non-cTIV datasets. The PA yielded 37 dimensions for cTIV and 23 non-cTIV. The discrepancies between the real and simulated eigenvalues from the ninth factor onward were in the PA on cTIV (0.98 − 0.003) and non-cTIV (0.43 − 0.009) indicators. Network loadings ranged from − 0.12 to 0.63 in the non-cTIV dataset, and from − 0.38 to 0.65 in cTIV dataset. Therefore, the lower bounds define the lower limit applied to obtain a network without any isolated node (Christensen & Golino, 2021 ; Golino et al., 2020 ). These network loadings are extracted after the factors have been extracted from the network’ structure. Total Entropy Fit Index (TEFI) values were 18.87 in non-cTIV and − 81.26 in cTIV. The regions showing different dimension locations were the pro-striate area (3rd dimension in cTIV and 1st in non-cTIV) and the inferior 6–8 transitional area (7th dimension in cTIV and 8th in non-cTIV). Furthermore, we observed differences in the order of the indicators in dimensions 7th and 8th mainly. However, the other indicators kept the same positions in cTIV and non-cTIV analyses. Attending these differences described previously, dimensions 1 and 9 gathered regions of the BIS/FFFS and BAS, respectively. Dimension 2 and 3 gathered regions of the posterior cingulate cortex (PCC), separately. Particularly, dimension 3 gathered regions between the PCC and the visual cortex, while dimension 2 gathered central regions of the PCC between retrosplenial and precuneus visual area. Dimension 4th gathered regions of the ventromedial prefrontal cortex (vmPFC) and anterior cingulate cortex (ACC). Dimension 5th gathered the rest of the ACC regions. Dimension 6 gathered all orbitofrontal cortex (OFC) regions and inferior frontal identified as ventrolateral prefrontal cortex (vlPFC). Dimension 7 included similar portions of the inferior frontal and dorsolateral prefrontal (dlPFC) in cTIV. However, dimension 7 in non-TIV included lateral inferior frontal regions except for the area posterior 9-46v included in the dlPFC subdivision of the HCPex. Dimension 8 included dlPFC in cTIV and non-TIV, but dimension 8 cTIV included a reduced number of dlPFC regions compared to non-TIV (See Online Resource 1 for a complete table of ROIs and their corresponding dimension). Network loadings, considered equivalent to factor loadings, are computed by standardizing node strength (sum of edge weights) within each dimension. They are standardized measures that quantify each variable´s contribution to the emergence of coherent dimensions within a network structure. These network loadings use regularized partial correlations and loadings below 0.15 are considered a small effect. The higher network loadings for each dimension (from 1st to 9th dimensions) in the cTIV (within parenthesis if not correspondence in non-cTIV) indicators were: Parahippocampal area 3, area dorsal 23 a + b, dorsal transitional visual area, area 10r (first 10v, second 10r), anterior 24 prime, area 47s, area IFSp, superior 6–8 transitional area and nucleus accumbens (NAcc) (globus pallidus internalis (GPi), second NAcc) (See Online Resource 3 for a complete table of network loadings). Attending to the same dimensions identified by EGA in non-TIV and cTIV structural data we identified nine dimensions: 1, the BIS/FFFS; 2, the CDS/PCC1; 3, the CDS/PCC2; 4, the ventromedial CDS; 5, the dorsomedial CDS; 6, the CVS; 7, the lateral inferior CDS; 8, the lateral superior CDS; and 9, the BAS. Based on the observed TEFI values, we continued working with the cTIV data, as they suggested a better structural fit. Corrected sMRI items and dimension stability The analysis of structural cTIV stability involved deleting the indicators with a stability < 0.75, as previously indicated (Table 2 ). In the first step, 6 indicators were excluded following this criterion. Also, dimensions 3, 6 and 7 had stability values below the threshold (dimensions stability range [0.72–0.98]), though only dimension 3 presented item stability average below threshold (0.72). Finally, all indicator stability was over threshold [range 0.80 -1], as well as dimensions stability (range [0.80 -1]), and average item stability (range [0.94-1]). Table 2 BootEGA sMRI dimension and item stability results. Dimension stability Dimension 1 2 3 4 5 6 7 8 9 Consistency 0.822 0.796 0.558 0.958 0.772 0.302 0.328 0.954 0.77 Item stability HCPex Region Stability ProStriate_Area 0.558 Posterior_OFC_Complex 0.742 Orbital_Frontal_Complex 0.742 Area_posterior_47r 0.504 Area_9-46d 0.384 Area_anterior_9-46v 0.504 The final EGA analysis revealed 9 dimensions on cTIV data after removing unstable items of a network with 71 nodes, 428 edges and edge density of 0.17 with mean edge weight of 0.076 ( SD = 0.120) (Fig. 1 ). Lambda = 0.0898 ( n = 100, ratio = 0.1). In this case, all dimensions have stability values above threshold and no unstable items (Table 3 ). However, TEFI value increased to -76.581. For the same set of data, the PA yielded 31 dimensions. The discrepancies between the real and simulated eigenvalues from the ninth factor onward were (0.985 − 0.038). Network loadings run from small (-0.395) to large (0.656) in cTIV. Table 3 cTIV dimension stability after bootEGA. Dimension 1 2 3 4 5 6 7 8 9 Consistency 0.996 0.802 1 0.956 0.848 0.936 0.916 1 0.898 Even after eliminating indicators due to instability, EGA maintains the number of dimensions as previously identified in the initial EGA analysis for the structural data: 1, the BIS/FFFS; 2, the CDS/PCC1; 3, the CDS/PCC2; 4, the ventromedial CDS; 5, the dorsomedial CDS; 6, the CVS; 7, the lateral inferior CDS; 8, the lateral superior CDS; and 9, the BAS. The higher network loadings for each dimension in the cTIV indicators (indicated if different in cTIV before bootEGA) were: Parahippocampal area 3, area dorsal 23 a + b, dorsal transitional visual area, area 10r, anterior 24 prime, area 47M (before bootstrap area 47s, now second), area posterior 9-46v (before bootstrap area IFSp, now second), superior 6–8 transitional area and NAcc (See Online Resource 3 for a complete table of network loadings). Functional data EGA on naturalistic fMRI data before bootEGA An EGA analysis was carried with the averaged functional dataset previously described in the methods section. This revealed a set of 8 dimensions, compared to the 10 dimensions calculated by the PA where the discrepancies of real and simulated values were (0.156 − 0.114). The EGA analysis resulted in a network with 77 nodes, 287 edges and edge density of 0.098 with mean edge weight of 0.08 ( SD = 0.094), also network loadings ranged from − 0.403 to 0.733. Lambda = 0.3176 ( n = 100, ratio = 0.1) and a TEFI value of -147.689. In this case, dimensions 1, 2 and 5 show items from the BIS/FFFS, while dimensions 5 and 8 contain BAS items. Also, dimension 2 gathered regions of the PCC. Then, dimension 3 combined ACC regions with a PCC region, as well as inferior frontal and dorsolateral prefrontal areas. As for the structural data, dimension 4 contained regions of the vmPFC and ACC, including some regions from the OFC and vlPFC. For the functional data, dimension 5 joined ACC and OFC regions with the amygdala and septal nucleus. Dimension 6 included some lateral OFC regions. Finally, lateral inferior frontal and dorsolateral prefrontal regions were included in dimension 7 (See Online Resource 1 for a complete table of ROIs and their corresponding dimension). The higher network loadings for each dimension in the naturalistic fMRI indicators were: Parahippocampal area 1, area 7m, area posterior 24, area10v, NAcc, area 47l, area 8Av and GPi (See Online Resource 3 for a complete table of network loadings). Corrected fMRI items and dimension stability Again, following the criteria of deleting indicators with stability lower than 0.75 except theoretically sound regions, 20 indicators were removed from the analysis. Dimensions from 2 to 7 presented stability values below threshold (dimensions stability range [0.206–0.85]), with dimensions 5 and 6 having average item stability lower than 0.75 (Table 4 ). Table 4 BootEGA fMRI dimension and item stability results. Dimension stability Dimension 1 2 3 4 5 6 7 8 Consistency 0.784 0.206 0.124 0.35 0.63 0.064 0.56 0.85 Item stability HCPex Region Stability HCPex Region Stability Entorhinal_Cortex 0.506 Area_45 0.066 Perirhinal_Ectorhinal_Cortex 0.506 Area_47l 0.066 Area_23d 0.532 Area_anterior_47r 0.064 ProStriate_Area 0.442 Area_IFSa 0.7 Area_25 0.71 Area_IFSp 0.732 Anterior_24_prime 0.678 Area_posterior_47r 0.632 Area_Posterior_24_prime 0.538 Area_8C 0.708 Posterior_OFC_Complex 0.71 Area_9-46d 0.57 Area_13l 0.476 Nucleus_Accumbens 0.71 Area_47m 0.582 Amygdala 0.63 Area_47s 0.45 Septal_nucleus 0.71 Area_44 0.728 The EGA analysis, carried after cleaning unstable items from the bootstrap results, using the parameters previously indicated in the "Methods" section reported 6 dimensions of a network with 57 nodes, 210 edges and edge density of 0.132 with mean edge weight of 0.096 ( SD = 0.110) (Fig. 2 ). Lambda = 0.2347 ( n = 100, ratio = 0.1). As with the structural data, the TEFI value increased to -102.055, and the network loading range was − 0.049 to 0.669. For this dataset, PA estimated 10 dimensions again, with discrepancies between the real and simulated eigenvalues from the ninth factor onward were (0.682 − 0.004). In this case, after eliminating indicators due to instability (some indicators have been retained due to their theoretical relevance), EGA yields unstable dimensions for the BIS in this data and identifies the following dimensions for the functional data: 1, the BIS/FFFS; 2, the CDS/PCC; 3, the ventromedial CDS; 4, the dorsomedial and lateral inferior CDS/CVS; 5, the lateral superior CDS; and 6, the BAS (Table 5 ). Table 5 fMRI dimension stability after bootEGA. Dimension 1 2 3 4 5 6 Consistency 0.584 0.85 0.932 0.926 0.998 0.84 The higher network loadings for each dimension in the naturalistic fMRI after bootstrap (noted if different before bootEGA) indicators were: Parahippocampal area 1 (NAcc appeared in a separate dimension before bootstrap and is second after bootEGA), area 7m, area10v, area posterior 24, Area 8Av and substantia nigra pars compacta (SNpc) (before it was GPi, which is the second one in this case). The higher network loadings items for each dimension in the naturalistic fMRI compared to those of cTIV after bootEGA (cTIV differences noted in parentheses) were: Parahippocampal area 1 (parahippocampal area 3), Area 7m (area dorsal 23a + b) for the dimensions BIS/FFFS and CDS/PCC (CDS/PCC1 and CDS/PCC2), respectively; area 10v (area 10r, area 10v was second for cTIV) for the ventromedial CDS; area posterior 24 (anterior 24 prime) in the dorsomedial and lateral inferior CDS/CVS (dorsomedial CDS and CVS); area 8Av (superior 6–8 transitional area) for the lateral superior CDS (the lateral inferior and lateral superior CDS); and SNpc, followed by GPi (NAcc had the higher loading followed by SNpc and GPi) for the BAS (See Online Resource 3 for a complete table of network loadings). DISCUSSION AND CONCLUSION This study provides new insights into the neurobiological foundations of the RST proposing a more detailed topology than our previous proposal (Lacomba-Arnau et al., 2025 ). While structural and functional MRI indicators suggest a similar number of dimensions, we interpret this cross-modality convergence as evidence for a shared system architecture that allows for modality-specific expression. Structural covariance may reflect long-term organizational constraints shaped by development, genetics, whereas naturalistic functional covariance may reflect the deployment of these constraints under rich, ecological stimulation. Specifically, the BIS-FFFS, the BAS and the CVS showed differentiated structural dimensionality keeping in agreement with our previous study. Notably, the CDS showed a dimensional division that gets simplified after bootstrap stability analysis on fMRI indicators. These results suggest that while brain dimensional organization identified by EGA keeps agreement with latent factors suggested by the RST (Kennis et al., 2013 ; Lacomba-Arnau et al., 2025 ), cortical organization may offer more complex parcellations in different neurobiological systems. Within the network psychometrics framework, these systems are treated as emergent organizational units reflected in patterns of conditional dependence among regions. While this does not establish causality, it provides a formal representation of coordinated system architecture consistent with the RST’s system-level principles. Our findings confirm that the structural covariance of brain regions identified with the BIS-FFFS and the BAS systems remains coupled (Lacomba-Arnau et al., 2025 ). However, while our previous study suggested these systems were latent causes describing brain organization, the current EGA results extend these parcels into finer sub-divisions. Specifically, the BIS-FFFS involves three partitions of the parahippocampi as well as the perirhinal system besides the entorhinal, adding layers to the differentiation found previously. Notably, the emergence of parahippocampal regions as highly central within the BIS/FFFS dimension suggests that, in structural covariance, the system may be organized around contextual-mnemonic circuitry rather than localized in a single threat-detection region. By contrast, the nucleus accumbens emerging as central within the BAS dimension is consistent with its well-established involvement in incentive motivation and approach-related circuitry. From a NP approach, these regions are viewed as a complex system of mutually reinforcing variables; a structural change in one region is statistically associated with others within the identified neurobiological system (Christensen et al., 2020 ; Christensen & Golino, 2021 ). The interpretation of structural and functional datasets provides complementary insights into RST dimensionality. Importantly, partial differences between structural and functional dimensionality are consistent with contemporary accounts of distributed brain organization, which emphasize that psychological functions arise from coordinated interactions among regions rather than isolated modules (Pessoa, 2014 ). From this perspective, structural covariance may reflect relatively stable anatomical constraints, likely influence of genetic, environmental, or neuroplasticity (e.g., learning) effects influenced by genetic, environmental, or neuroplasticity effects, whereas functional covariance captures the dynamic deployment of these systems under specific contextual demands. Importantly, naturalistic fMRI paradigms reliably elicit coordinated large-scale activity patterns that generalize across individuals, capturing functional network organization under ecologically meaningful conditions (Hasson et al., 2010 ; Sonkusare et al., 2019 ). The functional covariance structure observed here reflects the between-subject regularities in how these systems are engaged across complex environmental inputs, such as movie-watching. While structural dimensions represent the enduring organizational architecture, future studies are recommended to further explore the specific influence of genetic or environmental factors on these dimensions. Because RST concerns individual differences in motivational (McNaughton & Gray, 2000 ), the critical question is how such stability is instantiated neurally. Personality is typically defined as relatively stable patterns of behavior, affect, and cognition across situations, although expressed in context-dependent ways. Within this framework, these traits reflect stable differences in the sensitivity of brain systems mediating responses to reward and punishment (Corr & Matthews, 2009 ). Our findings suggest that structural and functional dimensions capture complementary aspects of these systems: structural covariance reflecting relatively stable system architecture, and functional covariance reflecting context-sensitive system expression. This suggests that while these dimensions capture stable individual differences in neurobiological responsiveness, they may also reflect stimulus-driven regularities. Distinguishing trait-like system sensitivity from stimulus-specific engagement remains an important goal for future work. In the context of NP, factors in a network model “emerge” from the causal connections between nodes (Christensen & Golino, 2021 ; Cramer et al., 2012 ). Thus, network loadings for a node are interpreted as its unique contribution to the emergence of dimension (Christensen et al., 2020 ). We observed that the central nodes, those with higher network loadings, remain comparable across structural methodologies (cTIV and non-cTIV). Interestingly, the NAcc was the central node related to the BAS, while the central node of the BIS was the parahippocampus as discussed before. Furthermore, the CVS showed the involvement of the ventrolateral prefrontal cortex as a central node (IFSp), whereas we found a high diversity in the nodes of the partialized CDS, further suggesting a non-unitary control architecture. Regarding the naturalistic fMRI dataset, the central nodes showed a high congruency before and after the bootstrap stability analysis, with the same central nodes for all dimensions except for the last one (and the ones absent after the bootstrap), reinforcing the robustness of the functional dimensions. However, a key distinction emerged between modalities: although the functional dimensions showed a separate BAS and BIS/FFF dimensions, central nodes like the NAcc (BAS) and amygdala (BIS/FFFS) cluster within a single dimension. This suggests that during complex, stimulus-driven activity, these systems may engage in a high degree of functional co-activation or shared variance that is not present in their underlying anatomical architecture. Such a finding underscores that the "core" regions of these systems may shift depending on whether one is measuring stable structural blueprints or the dynamic, integrated response to ecologically rich stimuli. The subdivision of the CDS into multiple dimensions may indicate that ‘constraint’ is not a unitary system at the neuroanatomical level, but rather a set of partially dissociable control components that become visible with higher brain-parcellation. This refinement may help reconcile RST-inspired systems with broader accounts of prefrontal control architecture. Specifically, we observed that the dimension identified in our previous work (Lacomba-Arnau et al., 2025 ) was divided into several dimensions here. The results suggest that there is an effect of the sample and the number of indicators in the dimensions extracted, as the current study utilized the HCP sample and the HCPex parcellation rather than the home-sample and the Neuromorphometric atlas used previously. This research is not without limitations. Although entropy-based fit indices suggested potential overfactoring, the persistence of the nine-dimensional solution across bootstrapped structural analyses supports its robustness. Future work should examine whether the present dimensional solution represents a stable characterization of RST systems or whether broader, more parsimonious configurations better capture their functional organization across different samples and tasks. Finally, regarding ROI selection, some regions were chosen based on correspondence with the Neuromorphometric atlas used in our previous research (Lacomba-Arnau et al., 2025 ) rather than their relevance in the systems (e.g., the posterior striate). Others showed proximity effects, such as the septal nucleus that always goes with the NAcc. It is important to signal those effects as an alternative hypothesis. In sum, our findings provide a systems-level test of a theory-defined architecture using multivariate brain covariance structure. Rather than treating MRI measures as direct proxies of isolated psychological processes, we demonstrate that theory-driven systems can be evaluated as reproducible organizational dimensions within distributed neural networks. This shifts the focus of personality neuroscience from region-based inference to the validation of system-level architectures, where stable dispositions, encompassing motivational, defensive, and regulatory processes, are expressed as coordinated patterns of structural and functional organization. By demonstrating that psychological constructs can be formally integrated with biological phenotypes, this study confirms that the neuroconceptual nervous system proposed by the RST is directly reflected in the brain’s intrinsic organizational dependencies. Ultimately, this approach contributes to a cumulative and testable framework for psychometric neuroscience, bridging the gap between behavioral theory and neurobiological phenotypes Declarations Competing interests The authors have no relevant financial or non-financial interests to disclose or potential conflicts of interest. Funding The publication is part of project PID2021-127340NB-C21, funded by Ministerio de Ciencia e Innovación, Spain (Grant No. MCIN/AEI/ 10.13039/501100011033/FEDER , European Union). Author Contribution OPD performed data curation and formal analysis, wrote the original draft, and led the review and editing of the final manuscript. ELA contributed to the interpretation of the results and the review of the manuscript. AMM participated in the conceptualization of the study and the development of the initial methodology, and the review of the manuscript. ABL conceptualized and supervised the study, participated in the interpretation of the data, and contributed to the writing of the original draft.All authors approved the final version of the manuscript. Acknowledgement The publication is part of project PID2021-127340NB-C21, funded by Ministerio de Ciencia e Innovación, Spain (Grant No. MCIN/AEI/10.13039/501100011033/FEDER, European Union). Data Availability The neuroimaging MRI data used in this study are publicly available through the Human Connectome Project (HCP) Young Adult dataset (https://www.humanconnectome.org/). The data were collected and shared by the Washington University-University of Minnesota (WU-Minn HCP) Consortium. Access to the data is subject to the HCP’s terms of use, which require registration and acceptance of Data Use Terms.The custom R scripts used for data curation, TIV correction, and Exploratory Graph Analysis (EGA) are available from the corresponding author upon reasonable request. References Bolt, T., Prince, E. B., Nomi, J. S., Messinger, D., Llabre, M. M., & Uddin, L. Q. (2018). Combining region- and network-level brain-behavior relationships in a structural equation model. NeuroImage , 165 , 158–169. https://doi.org/10.1016/j.neuroimage.2017.10.007 Borsboom, D. (2006). The Attack of the Psychometricians. Psychometrika , 71 (3), 425–440. https://doi.org/10.1007/S11336-006-1447-6 Christensen, A. P., Garrido, L. E., & Golino, H. (2023). Unique Variable Analysis: A Network Psychometrics Method to Detect Local Dependence. Multivariate Behavioral Research , 58 (6), 1165–1182. https://doi.org/10.1080/00273171.2023.2194606 Christensen, A. P., & Golino, H. (2021). Estimating the Stability of Psychological Dimensions via Bootstrap Exploratory Graph Analysis: A Monte Carlo Simulation and Tutorial. Psych 2021, Vol. 3, Pages 479-500 , 3 (3), 479–500. https://doi.org/10.3390/PSYCH3030032 Christensen, A. P., Golino, H., Abad, F. J., & Garrido, L. E. (2025). Revised network loadings. Behavior Research Methods 2025 57:4 , 57 (4), 114-. https://doi.org/10.3758/S13428-025-02640-3 Christensen, A. P., Golino, H., & Silvia, P. J. (2020). A Psychometric Network Perspective on the Validity and Validation of Personality Trait Questionnaires. European Journal of Personality , 34 (6), 1095–1108. https://doi.org/10.1002/PER.2265;WGROUP:STRING:PUBLICATION Cooper, S. R., Jackson, J. J., Barch, D. M., & Braver, T. S. (2019). Neuroimaging of individual differences: A latent variable modeling perspective. Neuroscience & Biobehavioral Reviews , 98 , 29–46. https://doi.org/10.1016/J.NEUBIOREV.2018.12.022 Corr, P. J., & Matthews, G. (2009). The Cambridge Handbook of Personality Psychology. In P. J. Corr & G. Matthews (Eds.), The Cambridge Handbook of Personality Psychology . Cambridge University Press. https://doi.org/10.1017/CBO9780511596544 Cramer, A. O. J., van der Sluis, S., Noordhof, A., Wichers, M., Geschwind, N., Aggen, S. H., Kendler, K. S., & Borsboom, D. (2012). Dimensions of normal personality as networks in search of equilibrium: You can’t like parties if you don’t like people. European Journal of Personality , 26 (4), 414–431. https://doi.org/10.1002/PER.1866;REQUESTEDJOURNAL:JOURNAL:10990984;PAGEGROUP:STRING:PUBLICATION Duboc, V., Dufourcq, P., Blader, P., & Roussigné, M. (2015). Asymmetry of the Brain: Development and Implications. Annual Review of Genetics , 49 , 647–672. https://doi.org/10.1146/ANNUREV-GENET-112414-055322 Finn, E. S., & Bandettini, P. A. (2021). Movie-watching outperforms rest for functional connectivity-based prediction of behavior. NeuroImage , 235 , 117963. https://doi.org/10.1016/J.NEUROIMAGE.2021.117963 Formisano, E., & Goebel, R. (2003). Tracking cognitive processes with functional MRI mental chronometry. Current Opinion in Neurobiology , 13 (2), 174–181. https://doi.org/10.1016/S0959-4388(03)00044-8 Gaser, C., Dahnke, R., Thompson, P. M., Kurth, F., & Luders, E. (2024). CAT: a computational anatomy toolbox for the analysis of structural MRI data. GigaScience , 13 . https://doi.org/10.1093/GIGASCIENCE/GIAE049 Glasser, M. F., Sotiropoulos, S. N., Wilson, J. A., Coalson, T. S., Fischl, B., Andersson, J. L., Xu, J., Jbabdi, S., Webster, M., Polimeni, J. R., Van Essen, D. C., & Jenkinson, M. (2013). The Minimal Preprocessing Pipelines for the Human Connectome Project. NeuroImage , 80 , 105. https://doi.org/10.1016/J.NEUROIMAGE.2013.04.127 Golino, H. F., & Epskamp, S. (2017). Exploratory graph analysis: A new approach for estimating the number of dimensions in psychological research. PLOS ONE , 12 (6), e0174035. https://doi.org/10.1371/JOURNAL.PONE.0174035 Golino, H., Moulder, R., Shi, D., Christensen, A. P., Garrido, L. E., Nieto, M. D., Nesselroade, J., Sadana, R., Thiyagarajan, J. A., & Boker, S. M. (2021). Entropy Fit Indices: New Fit Measures for Assessing the Structure and Dimensionality of Multiple Latent Variables. Multivariate Behavioral Research , 56 (6), 1–29. https://doi.org/10.1080/00273171.2020.1779642 Golino, H., Shi, D., Christensen, A. P., Garrido, L. E., Nieto, M. D., Sadana, R., Thiyagarajan, J. A., & Martínez-Molina, A. (2020). Investigating the performance of Exploratory Graph Analysis and traditional techniques to identify the number of latent factors: A simulation and tutorial. Psychological Methods , 25 (3), 292–320. https://doi.org/10.1037/MET0000255 Haslbeck, J. M. B., & Waldorp, L. (2015). Structure estimation for mixed graphical models in high-dimensional data. ArXiv: Applications . Haslbeck, J. M. B., & Waldorp, L. J. (2020). mgm: Estimating Time-Varying Mixed Graphical Models in High-Dimensional Data. Journal of Statistical Software , 93 , 1–46. https://doi.org/10.18637/jss.v093.i08 Hasson, U., Malach, R., & Heeger, D. J. (2010). Reliability of cortical activity during natural stimulation. Trends in Cognitive Sciences , 14 (1), 40–48. https://doi.org/10.1016/j.tics.2009.10.011 Huang, C. C., Rolls, E. T., Feng, J., & Lin, C. P. (2022). An extended Human Connectome Project multimodal parcellation atlas of the human cortex and subcortical areas. Brain Structure and Function , 227 (3), 763–778. https://doi.org/10.1007/S00429-021-02421-6, Kennis, M., Rademaker, A. R., & Geuze, E. (2013). Neural correlates of personality: An integrative review. Neuroscience and Biobehavioral Reviews , 37 (1), 73–95. https://doi.org/10.1016/j.neubiorev.2012.10.012 Kjellström, S., & Golino, H. (2019). Mining concepts of health responsibility using text mining and exploratory graph analysis. Scandinavian Journal of Occupational Therapy , 26 (6), 395–410. https://doi.org/10.1080/11038128.2018.1455896 Lacomba-Arnau, E., Martínez-Molina, A., Garrido, L. E., & Barrós-Loscertales, A. (2025). Neural Topologies of Reinforcement Sensitivity Theory: A Latent Variable Approach to Magnetic Resonance Imaging Data. Biological Psychiatry Global Open Science , 5 (5). https://doi.org/10.1016/j.bpsgos.2025.100526 Laurent, M.-A., Jacques, C., Yan, X., Jurczynski, P., Colnat-Coulbois, S., Maillard, L., Le Cam, S., Ranta, R., Cottereau, B. R., Koessler, L., Jonas, J., & Rossion, B. (2025). A tight relationship between BOLD fMRI activation/deactivation and increase/decrease in single neuron responses in human association cortex. ELife , 14 . https://doi.org/10.7554/ELIFE.104779 Logothetis, N. K. (2003). The Underpinnings of the BOLD Functional Magnetic Resonance Imaging Signal. Journal of Neuroscience , 23 (10), 3963–3971. https://doi.org/10.1523/JNEUROSCI.23-10-03963.2003 Logothetis, N. K., Pauls, J., Augath, M., Trinath, T., & Oeltermann, A. (2001). Neurophysiological investigation of the basis of the fMRI signal. Nature 2001 412:6843 , 412 (6843), 150–157. https://doi.org/10.1038/35084005 McNaughton, N., & Gray, J. A. (2000). Anxiolytic action on the behavioural inhibition system implies multiple types of arousal contribute to anxiety. Journal of Affective Disorders , 61 (3), 161–176. https://doi.org/10.1016/S0165-0327(00)00344-X Mechelli, A., Friston, K. J., Frackowiak, R. S., & Price, C. J. (2005). Structural covariance in the human cortex. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience , 25 (36), 8303–8310. https://doi.org/10.1523/JNEUROSCI.0357-05.2005 Menon, R. S., Luknowsky, D. C., & Gati, J. S. (1998). Mental chronometry using latency-resolved functional MRI. Proceedings of the National Academy of Sciences of the United States of America , 95 (18), 10902–10907. https://doi.org/10.1073/pnas.95.18.10902 Nordenskjöld, R., Malmberg, F., Larsson, E. M., Simmons, A., Ahlström, H., Johansson, L., & Kullberg, J. (2015). Intracranial volume normalization methods: Considerations when investigating gender differences in regional brain volume. Psychiatry Research: Neuroimaging , 231 (3), 227–235. https://doi.org/10.1016/J.PSCYCHRESNS.2014.11.011 Pessoa, L. (2014). Understanding brain networks and brain organization. Physics of Life Reviews , 11 (3), 400. https://doi.org/10.1016/j.plrev.2014.03.005 Poldrack, R. A. (2006). Can cognitive processes be inferred from neuroimaging data? Trends in Cognitive Sciences , 10 (2), 59–63. https://doi.org/10.1016/j.tics.2005.12.004 Poldrack, R. A., Baker, C. I., Durnez, J., Gorgolewski, K. J., Matthews, P. M., Munafò, M. R., Nichols, T. E., Poline, J. B., Vul, E., & Yarkoni, T. (2017). Scanning the horizon: towards transparent and reproducible neuroimaging research. Nature Reviews. Neuroscience , 18 (2), 115. https://doi.org/10.1038/nrn.2016.167 Rowley, C. D., Bazin, P. L., Tardif, C. L., Sehmbi, M., Hashim, E., Zaharieva, N., Minuzzi, L., Frey, B. N., & Bock, N. A. (2015). Assessing intracortical myelin in the living human brain using myelinated cortical thickness. Frontiers in Neuroscience , 9 (OCT). https://doi.org/10.3389/fnins.2015.00396 Schmittmann, V. D., Cramer, A. O. J., Waldorp, L. J., Epskamp, S., Kievit, R. A., & Borsboom, D. (2013). Deconstructing the construct: A network perspective on psychological phenomena. New Ideas in Psychology , 31 (1), 43–53. https://doi.org/10.1016/J.NEWIDEAPSYCH.2011.02.007 Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E. J., Johansen-Berg, H., Bannister, P. R., De Luca, M., Drobnjak, I., Flitney, D. E., Niazy, R. K., Saunders, J., Vickers, J., Zhang, Y., De Stefano, N., Brady, J. M., & Matthews, P. M. (2004). Advances in functional and structural MR image analysis and implementation as FSL. NeuroImage , 23 Suppl 1 (SUPPL. 1). https://doi.org/10.1016/J.NEUROIMAGE.2004.07.051 Sonkusare, S., Breakspear, M., & Guo, C. (2019). Naturalistic Stimuli in Neuroscience: Critically Acclaimed. Trends in Cognitive Sciences , 23 (8), 699–714. https://doi.org/10.1016/j.tics.2019.05.004 Soreq, E., Violante, I. R., Daws, R. E., & Hampshire, A. (2021). Neuroimaging evidence for a network sampling theory of individual differences in human intelligence test performance. Nature Communications 2021 12:1 , 12 (1), 2072-. https://doi.org/10.1038/s41467-021-22199-9 Tesler, F., Linne, M. L., & Destexhe, A. (2023). Modeling the relationship between neuronal activity and the BOLD signal: contributions from astrocyte calcium dynamics. Scientific Reports 2023 13:1 , 13 (1), 6451-. https://doi.org/10.1038/s41598-023-32618-0 Van der Weele, T. J., & Vansteelandt, S. (2022). A Statistical Test to Reject the Structural interpretation of a Latent Factor Model. Journal of the Royal Statistical Society Series B: Statistical Methodology , 84 (5), 2032–2054. https://doi.org/10.1111/RSSB.12555 Yarkoni, T. (2022). The generalizability crisis. Behavioral and Brain Sciences , 45 , e1. https://doi.org/10.1017/S0140525X20001685 Additional Declarations No competing interests reported. Supplementary Files SI1.xlsx SI2.xlsx SI3.xlsx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 04 May, 2026 Reviews received at journal 15 Apr, 2026 Reviewers agreed at journal 04 Apr, 2026 Reviewers invited by journal 03 Apr, 2026 Editor assigned by journal 23 Mar, 2026 Submission checks completed at journal 22 Mar, 2026 First submitted to journal 20 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9179108","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":619819864,"identity":"25704e9e-044e-41ca-81fc-96f33d9cd5aa","order_by":0,"name":"Oscar Perez-Diaz","email":"","orcid":"","institution":"Universitat Jaume I","correspondingAuthor":false,"prefix":"","firstName":"Oscar","middleName":"","lastName":"Perez-Diaz","suffix":""},{"id":619819865,"identity":"f06202d5-6986-4c17-a8f5-fda44042e36a","order_by":1,"name":"Elena Lacomba-Arnau","email":"","orcid":"","institution":"Luxembourg Institute of Health","correspondingAuthor":false,"prefix":"","firstName":"Elena","middleName":"","lastName":"Lacomba-Arnau","suffix":""},{"id":619819866,"identity":"0f886030-fcf4-4081-8d20-05d5321c1633","order_by":2,"name":"Agustín Martínez-Molina","email":"","orcid":"","institution":"Universidad Autónoma de Madrid","correspondingAuthor":false,"prefix":"","firstName":"Agustín","middleName":"","lastName":"Martínez-Molina","suffix":""},{"id":619819867,"identity":"5257d703-0100-468d-94c5-7668a725a2e0","order_by":3,"name":"Alfonso Barrós-Loscertales","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAApklEQVRIiWNgGAWjYBACAwhlw8DYTKKWNNK1HCbeXQzm7Gefffjw53xiczsD48MfxGix7Ek3njmz7XZiYzMDszEPUQ47kMbMzNsA1sImTZTDDM4/Y2bm+XMOpIX9J1EOM7gBtIWH7QDYFgaiHGY54xkz48y2ZOPGZsZmaaK0mPOnMTN8+GMnu7H/8MGPRDkMDgwbGBtI0sDAIE+i+lEwCkbBKBhBAAC1TS6wFf0bjAAAAABJRU5ErkJggg==","orcid":"","institution":"Universitat Jaume I","correspondingAuthor":true,"prefix":"","firstName":"Alfonso","middleName":"","lastName":"Barrós-Loscertales","suffix":""}],"badges":[],"createdAt":"2026-03-20 12:54:32","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9179108/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9179108/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106582754,"identity":"70cde807-1b2d-4067-87ec-eebcaf554b80","added_by":"auto","created_at":"2026-04-10 06:57:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":441018,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003ecTIV network after bootEGA\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9179108/v1/612350307778ee6c009f6dd1.png"},{"id":106582809,"identity":"eff56223-cbbd-4e55-aee7-bb212eb0a34a","added_by":"auto","created_at":"2026-04-10 06:57:35","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":324282,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eNaturalistic fMRI network after bootEGA\u003c/em\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9179108/v1/ecb6f2064be64bdf10335bb3.png"},{"id":106582822,"identity":"691d6928-ba28-4bf3-a3e7-9388404fcb8d","added_by":"auto","created_at":"2026-04-10 06:57:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1644285,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9179108/v1/62d7c2cf-888a-4040-904f-49dbda3031ea.pdf"},{"id":106582700,"identity":"2ee78d83-2eab-482c-ad9e-01032bad7fe6","added_by":"auto","created_at":"2026-04-10 06:57:24","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":31076,"visible":true,"origin":"","legend":"","description":"","filename":"SI1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9179108/v1/38fa40e9246fab3bc81936db.xlsx"},{"id":106582808,"identity":"0236b9db-4a52-47c0-b88b-d01eed9c4778","added_by":"auto","created_at":"2026-04-10 06:57:34","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":33202,"visible":true,"origin":"","legend":"","description":"","filename":"SI2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9179108/v1/1ac7f7b202e72c4f13673768.xlsx"},{"id":106582701,"identity":"369d59fc-1fa7-4fe9-a254-242a8c63a064","added_by":"auto","created_at":"2026-04-10 06:57:25","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":43396,"visible":true,"origin":"","legend":"","description":"","filename":"SI3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9179108/v1/42d05c5cd9be37df11be8455.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Neuroanatomical and Functional Dimensionality of Reinforcement Sensitivity Theory Systems: An Exploratory Graph Analysis Approach","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eIn behavioral neuroscience, psychological constructs are frequently used to describe and interpret patterns of brain structure and function, even when the relation between psychological constructs and neurobiological measure is not straightforward (Borsboom, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). This cross-level mapping problem is not merely interpretative, but epistemological. Psychological constructs are typically defined at the level of behavior, affect, or self-report, whereas MRI captures distributed statistical properties of neural organization. Clarifying this mapping is therefore essential for construct validity, interpretability, and cumulative integration across studies. Subsequently, Magnetic Resonance Imaging (MRI) measurements are interpreted as reflecting transient psychological processes or trait-like features of brain organization, despite the fact that these measures may capture complex and distributed properties of brain activation patterns as measures of psychological constructs expressions. Consequently, descriptions of psychological processes and constructs may not accurately reflect activation patterns commonly observed in the human brain (Bolt et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis practice raises fundamental questions: to what extent do commonly used psychological constructs correspond to identifiable organizational principles in the human brain? What am I really measuring? And, is my interpretation adequate? When researchers conduct an analysis of variance or similar statistical test on MRI measures, and their effects are directly interpreted as transient psychological processes (i.e., working memory) or stable traits (i.e., personality), alternative multidimensional or latent class structures are often neglected (H. F. Golino \u0026amp; Epskamp, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Van der Weele \u0026amp; Vansteelandt, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This lack of attention to organizational structures of psychological theories or ontologies difficults the integration across tasks, samples, and analytic strategies, thereby constraining cumulative inference. Despite these limitations, the scientific community continues to value MRI indicators as in vivo neurobiological measures of whole-brain function and structure (Laurent et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Logothetis, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Logothetis et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Rowley et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Tesler et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), albeit explicitly modeling the dimensional structure of neurobiological systems may facilitate the transition from isolated statistical effects to reproducible principles of brain organization (Poldrack et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Yarkoni, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eEquating MRI measures to theoretical attributes, such as working memory processes or personality traits, highlights the perceived proximity between behavioral\u0026ndash;psychometric space and network states (Soreq et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Prior research has suggested that patterns of brain activation may directly reflect latent cognitive abilities (Haslbeck \u0026amp; Waldorp, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2015\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Soreq et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and can therefore serve as manifest indicators in psychometric models (Cooper et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Structural MRI (sMRI) morphometry and functional MRI (fMRI) time-series are often used as a form of neurobiological chronometry, not only through subtraction-based methodologies linking brain to behavior across task conditions or group contrast, but also by analogy with reaction time in mental chronometry, thereby rendering Blood Oxygen Level Dependent (BOLD) contrasts conceptually equivalent to other psychological variables (Formisano \u0026amp; Goebel, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Menon et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). In neuroimaging research, this inferential leap has been discussed in terms of reverse inference and construct validity: the assumption that activation in a given region necessarily implies engagement of a specific psychological process (Poldrack, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Such interpretations are especially problematic for constructs defined at the systems level, where explanatory adequacy depends on coordinated patterns of activity rather than on single regional effects (Pessoa, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). In this context, the family of methods under the umbrella of network psychometrics (NP) provides an alternative to other traditional multivariate psychometric approaches, such as component extraction techniques (e.g., Principal Component Analysis) or confirmatory factor analysis (Cooper et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). NP explicitly accommodates heterogeneous variable types and supports personalized brain models, showing individual-level reliability comparable to those observed in behavioral measures (Borsboom, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNP allows the representation of direct relationships among variables without assuming the existence of latent factors, thus enabling distributed patterns of alteration and interactions that do not necessarily conform to a single factor. At the same time, NP enables the analysis of network properties of latent factors derived from neuroimaging indicators. Within this framework, the brain is conceptualized as a distributed system characterized by non-hierarchical causality, emergent processes, and dynamic interactions. While network models do not by themselves demonstrate causal mechanisms, they offer a structured representation of system-level organization in which constructs are instantiated in patterns of conditional dependence among indicators. As such, NP furnishes a principled interface between theory-defined neurobehavioral systems and the multivariate architecture of neuroimaging data (Borsboom, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Schmittmann et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). From this perspective, psychological constructs are conceptualized as complex systems of observable variables that dynamically and mutually reinforce one another. Constructs are thus defined by the causal (bi)directional relations among observed MRI variables rather than being caused by a single underlying latent entity, although latent variables extracted from theoretically grounded indicators can themselves serve as nodes within networks. Thus, rather than being caused by a single latent entity, constructs emerge from the pattern of (bi)directional relationships among observed MRI variables, although theoretically grounded latent variables may themselves be represented as nodes within such networks.\u003c/p\u003e \u003cp\u003eThe Reinforcement Sensitivity Theory (RST) offers a particularly stringent test case for NP modeling because its explanatory framework is inherently system-based: motivational and affective sensitivities are posited to arise from coordinated neural circuitry with partially overlapping yet functionally differentiated components, subserved by specific cortical and subcortical structures (Kennis et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; McNaughton \u0026amp; Gray, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). Our prior study explored the dimensionality of brain structural indicators (Lacomba-Arnau et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). That study underscored the critical importance of utilizing dimensionality determination techniques, specifically Parallel Analysis and Exploratory Graph Analysis (EGA), to empirically identify the true number of latent structures in MRI data rather than relying solely on theoretical assumptions. This previous work led to the discovery of a robust 4-factor model: the Behavioral Inhibition System / Fight-Freeze-Flight System (BIS/FFFS); the Behavioral Approach System (BAS); the Constraint Dorsal Stream (CDS); and the Constraint Ventral Stream (CVS). Demonstrating that these theory-defined systems emerge as reproducible dimensions within brain covariance structures would therefore constitute direct support for the neurobiological architecture proposed by the theory. Furthermore, the findings demonstrated the superiority of factorial approaches (such as EFA and ESEM) over Principal Component Analysis (PCA); while PCA conflates systematic and error variance and yielded the least favorable fit indices, factorial methods successfully separate shared variance from measurement error, offering a more precise and biologically valid representation of neurobiological systems.\u003c/p\u003e \u003cp\u003eNonetheless, two questions remained open from that study. First, the Neuromorphometric parcellation did not provide the level of detail necessary for a more refined subcortical parcellation of the BIS/FFFS and the BAS. A more detailed subcortical delineation is theoretically relevant because RST differentiates systems according to their functional roles (e.g., conflict monitoring and defensive responding versus incentive motivation and approach), which rely on partially overlapping yet functionally differentiated subcortical circuitry. Increasing anatomical resolution enables us to evaluate whether the hypothesized system boundaries remain coherent when the measurement framework captures this internal differentiation. Second, we wondered whether that dimensionality would be supported on functional as well as structural covariance. Therefore, we planned to run RST hypothesis driven dimensionality analysis to restrict the interpretation of multivariate covariance analyses to a set of indicators relevant for the theory at test, reducing the alternative explanations at the cost of loss of information.\u003c/p\u003e \u003cp\u003eThe evaluation of dimensionality or the number of underlying factors is a critical and fundamental step for the validation and understanding of latent structures in multivariate datasets (H. Golino et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), especially in complex domains such as neuroscience. Following the line of research that utilizes fMRI and sMRI indicators to address psychological constructs, as seen in the work by Lacomba-Arnau et al. (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) which examines brain latent variables from structural covariance, we continue our interest in testing a methodology capable of handling the complexity and inherent interrelationships of cerebral metrics. In this context, NP has emerged as an advanced quantitative framework (H. Golino et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; H. F. Golino \u0026amp; Epskamp, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), where variables (e.g., brain parcels or regions of interest) are represented as nodes and their conditional associations as edges (links), creating a system of mutually influencing elements beyond merely reflecting a common latent cause (Christensen et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; H. Golino et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Schmittmann et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eEGA is a novel technique used within the framework of network psychometrics, following the framework described by Golino and Epskamp (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). EGA is designed to estimate the number and content of dimensions accurately and visually by identifying densely connected \u003cem\u003ecommunities\u003c/em\u003e or \u003cem\u003eclusters\u003c/em\u003e within the network (Christensen et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; H. F. Golino \u0026amp; Epskamp, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Kjellstr\u0026ouml;m \u0026amp; Golino, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). EGA operates by first estimating a Gaussian Graphical Model (H. F. Golino \u0026amp; Epskamp, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u0026mdash;often utilizing the Graphical LASSO (GLASSO) operator to obtain regularized partial correlations\u0026mdash;followed by a community detection algorithm, such as Walktrap, which allows latent dimensions to emerge directly from the structure of interconnections among indicators (Christensen et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; H. F. Golino \u0026amp; Epskamp, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Kjellstr\u0026ouml;m \u0026amp; Golino, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). This approach provides an intuitive visual guide regarding which neuroimaging indicators cluster together, offering a detailed dimensional structure without attending to the rotation or the \u003cem\u003ea priori\u003c/em\u003e assumptions of traditional factorial methods. Therefore, the use of EGA may permit an exploratory re-evaluation of the dimensions underlying cerebral covariance patterns for precisely specifying latent factors in agreement with the RST neuropsychological research.\u003c/p\u003e \u003cp\u003eIn sum, the aim of this study is to test the brain dimensionality of sMRI and fMRI measures in regions of interest (ROIs) involved in the different RST systems, the BAS, the BIS/FFFS and the CVS and CDS. We expect to find a similar dimensional organization of the brain from structural and functional indicators. Specifically, we expect that the ROIs theoretically suggested to gather under the BAS, BIS/FFFS, CVS and CDS will be differentiated in networks of nodes attending to each system. We evaluate whether these ROIs organize into coherent dimensions consistent with the proposed RST systems across structural and functional modalities. While a comparable dimensional structure across modalities would support the presence of shared system-level organization, differences between structural and functional covariance patterns would provide insight into how relatively stable anatomical constraints relate to context-dependent functional expression. Specifically, we expect that ROIs theoretically assigned to the BAS, BIS/FFFS, CVS, and CDS will group/cluster into differentiated networks of nodes corresponding to each system.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eDataset\u003c/h2\u003e \u003cp\u003eThis study uses sMRI and naturalistic fMRI data from the Human Connectome Project (HCP), a publicly available large-scale neuroimaging dataset containing healthy young adults. The sample includes up to 1200 participants (ages 22\u0026ndash;35, 54% female) with no history of neurological or psychiatric disorders, recruited from families with twins and siblings to support analyses of heritability and individual variability in brain organization. All participants provided informed consent, and the study was approved by the institutional review board of Washington University in St. Louis.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStructural MRI\u003c/h3\u003e\n\u003cp\u003eUnprocessed structural data are available for 1113 subjects with imaging data. High-resolution T1-weighted images were acquired using a Siemens 3T Connectome Skyra scanner with the following parameters: repetition time (TR)\u0026thinsp;=\u0026thinsp;2400 ms, echo time (TE)\u0026thinsp;=\u0026thinsp;2.14 ms, inversion time\u0026thinsp;=\u0026thinsp;1000 ms, flip angle\u0026thinsp;=\u0026thinsp;8\u0026deg;, voxel size\u0026thinsp;=\u0026thinsp;0.7 mm isotropic, and field of view\u0026thinsp;=\u0026thinsp;224 \u0026times; 224 mm.\u003c/p\u003e\n\u003ch3\u003eFunctional MRI – Naturalistic Paradigm\u003c/h3\u003e\n\u003cp\u003eThe naturalistic fMRI data used in this study correspond to the \u003cem\u003eHCP 7T Movie-Watching\u003c/em\u003e sub-dataset (Finn \u0026amp; Bandettini, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), which consists of functional acquisitions while participants viewed short audiovisual clips from commercial films and documentaries (1 to 4.3 min in length), which were concatenated and presented in four separate functional runs (total scan duration: 60 min). The movies contained diverse visual stimuli (people, animals, scenes, and objects), actions, sounds, music, speech, linguistic and social communications, and sometimes narratives. There were also 20 s rest periods between the movies. Across the 4 movie watching runs datasets there are 176 common subjects, which were used for later analyses.\u003c/p\u003e \u003cp\u003eAll data were acquired using a Siemens Magnetom 7 T MRI scanner equipped with a Nova 32-channel receive head coil, following the HCP 7 T imaging protocol. Gradient-echo echo-planar imaging (EPI) sequences were used with the following parameters: TR\u0026thinsp;=\u0026thinsp;1000 ms, TE\u0026thinsp;=\u0026thinsp;22.2 ms, flip angle\u0026thinsp;=\u0026thinsp;45\u0026deg;, voxel size\u0026thinsp;=\u0026thinsp;1.6 mm isotropic, multiband acceleration factor\u0026thinsp;=\u0026thinsp;5, in-plane acceleration\u0026thinsp;=\u0026thinsp;2, and partial Fourier\u0026thinsp;=\u0026thinsp;7/8. A total of 85 slices were acquired per volume with a bandwidth of 1924 Hz/Px. Phase-encoding direction alternated between posterior\u0026ndash;anterior and anterior\u0026ndash;posterior across runs to minimize geometric distortions. Each movie-watching run lasted approximately 15 min, with a total viewing time of slightly over one hour across the full session. The stimulus presentation and timing information are fully documented in the HCP data dictionary.\u003c/p\u003e\n\u003ch3\u003ePreprocessing\u003c/h3\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStructural MRI\u003c/h2\u003e \u003cp\u003eAlthough the HCP provides fully preprocessed structural data, in the present study we used the \u003cem\u003enative T1-weighted images\u003c/em\u003e from the HCP Young Adult dataset and performed independent preprocessing using the CAT12 toolbox (version 12.6, r1450; Gaser et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) implemented in SPM12. The preprocessing followed the standard CAT12 \u0026ldquo;expert mode\u0026rdquo; pipeline with default parameters unless otherwise specified. The main steps included: (1) Segmentation of the T1-weighted image into gray matter, white matter, and cerebrospinal fluid. (2) Normalization to the MNI152 template using the DARTEL high-dimensional registration approach, preserving regional volume through modulation by the Jacobian determinants of the deformation fields. (3) Bias field correction and affine registration to standard space. (4) Estimation of regional gray matter volume in native space prior to normalization. (5) Quality control by visual inspection and sample homogeneity analysis based on covariance metrics.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eFunctional MRI\u003c/h2\u003e \u003cp\u003eThe fMRI data were distributed preprocessed by the HCP consortium. These preprocessing pipelines are described in detail in Glasser et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) and made available through the HCP minimal preprocessing pipelines (Glasser et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.humanconnectome.org/\u003c/span\u003e\u003cspan address=\"https://www.humanconnectome.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e).\u003c/span\u003e Briefly, the fMRIVolume pipeline performed gradient nonlinearity correction, motion correction, and EPI readout distortion correction. Functional images were co-registered to the subject\u0026rsquo;s high-resolution structural T1w scan using boundary-based registration and normalized to the standard MNI152 space. Intensity inhomogeneities (bias field) were corrected using the map estimated from the structural processing, and the time series were linearly detrended without aggressive high-pass filtering. All spatial transformations were concatenated and applied in a single spline interpolation step (one-step resampling) to minimize spatial blurring and preserve data quality.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eROI extraction\u003c/h3\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eStructural MRI\u003c/h2\u003e \u003cp\u003eCAT12 segmentation included the extraction of mean gray matter volumes from the HCPex atlas (Huang et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), a multimodal parcellation extending the HCP-MMP1 atlas by including 66 additional subcortical areas in volumetric form. The use of the HCPex atlas allowed for direct application to structural T1 data, ensuring full coverage of cortical and subcortical regions relevant to the RST. ROI-wise structural values were exported as CSV files for subsequent dimensional analysis.\u003c/p\u003e \u003cp\u003eAdditionally, ROI volumes were corrected for the total intracranial volume (TIV) of each subject following the residual method (Nordenskj\u0026ouml;ld et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), implemented with \u003cem\u003ecurve_fit\u003c/em\u003e (scipy.optimize) in Python 3.11, to obtain an additional dataset corresponding to corrected brain volumes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eFunctional MRI\u003c/h2\u003e \u003cp\u003eFunctional time series were extracted from the preprocessed resting-state fMRI data using the \u003cem\u003efslmeants\u003c/em\u003e command from the FSL suite (FMRIB Software Library, Oxford, UK; Smith et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). The extraction was also based on the HCPex volumetric atlas, which includes 426 regions of interest (66 subcortical and 360 cortical areas), registered to the same MNI152 space as the preprocessed data. For each subject, the mean BOLD signal within each ROI was computed across all time points, resulting in one representative time series per region.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eROI Selection and Data curation\u003c/h2\u003e \u003cp\u003eROI indicators were selected from the HCPex atlas. See Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e for the 77 regions selected as indicators for the RST model according to their involvement and overlapping on the previous ROI definition using Neuromorphometrics in Lacomba-Arnau et al. (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Due to the high correlations between homotopic regions (Duboc et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Mechelli et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), we parceled homotopic ROIs using a combined strategy that is both isolated and content-oriented, averaging each left-right homologue into a single parcel. This approach aligns the unit of analysis with the system-level focus of RST, which is formulated in terms of distributed neurobehavioral systems rather than hemispheric specialization. This idea is also supported by the results of Unique Variable Analysis (UVA), showing a considerable number of redundancies between homotopic regions in sMRI and fMRI, as well as EGA carried with the complete datasets, revealing an average of 99.16%, 98.98% and 91.77%, for the non-TIV corrected (non-cTIV), TIV corrected (cTIV) and functional datasets, respectively (See Online Resource 1).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStudy ROIs.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSystem\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLabel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHCPex Region\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLabel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHCPex Region\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBIS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHip\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHippocampus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePHip_2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eParaHippocampal_Area_2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePreS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePreSubiculum\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePHip_3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eParaHippocampal_Area_3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEntorhinal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEntorhinal_Cortex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAmyg\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAmygdala\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePerirhinal_Ectorhinal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePerirhinal_Ectorhinal_Cortex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSeptal_nucleus\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePHip_1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eParaHippocampal_Area_1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBAS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePut\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePutamen\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGPi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGlobus_pallidus_internalis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCau\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCaudate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSNpc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSubstantia_nigra_pars_compacta\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNAcc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNucleus_Accumbens\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSNpr\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSubstantia_nigra_pars_reticulata\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGPe\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGlobus_pallidus_externalis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eVTA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eVentral_tegmenta_area\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCDS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_23d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_p32\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_31a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep32_p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_p32_prime\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31pd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_31pd\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31p_v\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_31p_ventral\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_7m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e47l\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_47l_(47_lateral)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23ab\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_dorsal_23_a\u0026thinsp;+\u0026thinsp;b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAnt_47r\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_anterior_47r\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTVis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDorsal_Transitional_Visual_Area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIFJa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_IFJa\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePreC_Visual\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePreCuneus_Visual_Area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIFJp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_IFJp\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePOS1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eParieto-Occipital_Sulcus_Area_1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIFSa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_IFSa\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePOS2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eParieto-Occipital_Sulcus_Area_2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIFSp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_IFSp\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProStriate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eProStriate_Area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePos_47r\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_posterior_47r\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRSplenial\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRetroSplenial_Complex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_46\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVen_23ab\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_ventral_23_a\u0026thinsp;+\u0026thinsp;b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8Ad\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_8Ad\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10r\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_10r\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8Av\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_8Av\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10v\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_10v\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLat_8B\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_8B_Lateral\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_8C\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33_p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_33_prime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9-46d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_9-46d\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8BM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_8BM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAnt_9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_9_anterior\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9_Mid\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_9_Middle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePos_9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_9_Posterior\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ea24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_a24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAnt_9-46v\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_anterior_9-46v\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAnt_24_p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAnterior_24_prime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eInf_6\u0026ndash;8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInferior_6\u0026ndash;8_Transitional_Area\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAnt_32_p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_anterior_32_prime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePos_9-46v_L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_posterior_9-46v\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDor_32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_dorsal_32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSup_6\u0026ndash;8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSuperior_6\u0026ndash;8_Transitional_Area\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePos_24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_posterior_24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSFL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSuperior_Frontal_Language_Area\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePos_24_p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_Posterior_24_prime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCVS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePos_OFC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eposterior_OFC_Complex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e47m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_47m\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003es32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_s32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e47s\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_47s\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_10d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAnt_10p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_anterior_10p\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePol_10p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePolar_10p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOFC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eOrbital_Frontal_Complex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11l\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_11l\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePos_10p\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eArea_posterior_10p\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13l\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eArea_13l\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFurthermore, for the functional data we averaged the signal of each ROI for each subject, obtaining a single average value per ROI for each subject per movie, which were then averaged across the common subjects (176 subjects) across all 4 movies. Thus, we carried the analysis using an average signal matrix of dimensions (ROIs x Subjects).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eDIMENSIONALITY ANALYSIS\u003c/h2\u003e \u003cp\u003eThe dimensionality assessment of the structural datasets was done using EGA, and Parallel Analysis (PA) as a comparative technique, for both cTIV and non-TIV corrected data.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eParallel Analysis\u003c/h2\u003e \u003cp\u003ePA was undertaken to objectively estimate the optimal number of latent factors in the dataset and validate results from network-based modeling approaches. PA was extracted using principal axis factoring. PA was performed using the RAWPAR function from \u003cem\u003eEGAnet\u003c/em\u003e (\u003cem\u003eR package, version 2.1.0\u003c/em\u003e), which implements permutation-based factor retention, generating 100 random datasets using the permuted randomization method, preserving the structure and distributional properties of the observed data. Principal Axis Factoring was employed to extract latent factors from both empirical and random datasets. Pearson correlations were used for both empirical and random structures. Factors were retained if observed eigenvalues exceeded the 95th percentile of corresponding randomly generated eigenvalues.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eUnique Variable Analysis\u003c/h2\u003e \u003cp\u003eFurthermore, to identify and handle redundant variables in whole-brain neuroimaging data, we applied the UVA procedure implemented in EGAnet. UVA uses the weighted Topological Overlap metric on a network estimated from the input variables to detect sets with strong local dependence, as described by Christensen, Garrido, and Golino (2020). In this case, the weighted Topological Overlap threshold was set as default (0.25).\u003c/p\u003e \u003cp\u003eIn this study, UVA was applied to the complete dataset (without hemisphere filtering), which predominantly revealed interhemispheric redundancies\u0026mdash;that is, marked topological similarity between variables representing homologous regions across both hemispheres. For the structural data UVA revealed 34 redundancies out of which 13 were between homotopic regions, representing a 17% of the total ROIs selected; as for the functional data 18 redundancies were found, with 11 corresponding to homotopic regions (14% of ROIs) (See Online Resource 2).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eExploratory Graph Analysis\u003c/h2\u003e \u003cp\u003eEGA is a novel technique used within the framework of network psychometrics. It was performed on the pre-processed datasets to estimate the dimensional structure of multivariate brain data using network psychometrics, following the framework described by Golino and Epskamp (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), Golino et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAs established, a Gaussian graphical model was estimated via GLASSO with extended Bayesian information criterion (default gamma\u0026thinsp;=\u0026thinsp;0.5) to select the optimal regularization parameter. Correlations were determined automatically (\u003cem\u003ecor_auto\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eCommunity detection was carried out with the Louvain algorithm, applied to the estimated network to identify latent communities (dimensions) within the data using 1000 iterations. The Louvain algorithm was also used to assess unidimensionality as part of standard EGA procedure.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eBootstrap Exploratory Graph Analysis\u003c/h2\u003e \u003cp\u003eTo evaluate the stability and replicability of the dimensional solution, Bootstrap Exploratory Graph Analysis (bootEGA) was performed with 500 parametric bootstrap samples, following recommended guidelines (Christensen \u0026amp; Golino, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). After the bootstrap procedure, two stability metrics were examined: The proportion of times each empirical dimension (community) was exactly replicated across bootstrap samples (structural stability) and the proportion of times each item was assigned to the same dimension as in the empirical EGA solution (item stability).\u003c/p\u003e \u003cp\u003eItems with item stability values below 0.75 were considered unstable and were removed from further analysis, except for those corresponding to central regions of the analyzed brain systems (based on theoretical relevance). Thus, we excluded any cortical indicator and subcortical indicator except central for the BIS/FFFS, such as the amygdala and hippocampus, or the BAS, such as the caudate, putamen, accumbens, SN or VTA. Then, the EGA and bootEGA were estimated again. This combined criterion preserved the integrity of core regions while reducing instability from peripheral or low-replicating items.\u003c/p\u003e \u003c/div\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eStructural data\u003c/h2\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003ch2\u003eEGA on sMRI data before bootEGA\u003c/h2\u003e \u003cp\u003eWe conducted two separate EGA dimensional analysis on gray matter volume indicators before and after correcting for TIV. The results were highly consistent across conditions, identifying 9 dimensions in both cTIV and non-cTIV datasets. The PA yielded 37 dimensions for cTIV and 23 non-cTIV. The discrepancies between the real and simulated eigenvalues from the ninth factor onward were in the PA on cTIV (0.98 − 0.003) and non-cTIV (0.43 − 0.009) indicators. Network loadings ranged from − 0.12 to 0.63 in the non-cTIV dataset, and from − 0.38 to 0.65 in cTIV dataset. Therefore, the lower bounds define the lower limit applied to obtain a network without any isolated node (Christensen \u0026amp; Golino, \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e; Golino et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e). These network loadings are extracted after the factors have been extracted from the network’ structure. Total Entropy Fit Index (TEFI) values were 18.87 in non-cTIV and − 81.26 in cTIV.\u003c/p\u003e \u003cp\u003eThe regions showing different dimension locations were the pro-striate area (3rd dimension in cTIV and 1st in non-cTIV) and the inferior 6–8 transitional area (7th dimension in cTIV and 8th in non-cTIV). Furthermore, we observed differences in the order of the indicators in dimensions 7th and 8th mainly. However, the other indicators kept the same positions in cTIV and non-cTIV analyses.\u003c/p\u003e \u003cp\u003eAttending these differences described previously, dimensions 1 and 9 gathered regions of the BIS/FFFS and BAS, respectively. Dimension 2 and 3 gathered regions of the posterior cingulate cortex (PCC), separately. Particularly, dimension 3 gathered regions between the PCC and the visual cortex, while dimension 2 gathered central regions of the PCC between retrosplenial and precuneus visual area. Dimension 4th gathered regions of the ventromedial prefrontal cortex (vmPFC) and anterior cingulate cortex (ACC). Dimension 5th gathered the rest of the ACC regions. Dimension 6 gathered all orbitofrontal cortex (OFC) regions and inferior frontal identified as ventrolateral prefrontal cortex (vlPFC). Dimension 7 included similar portions of the inferior frontal and dorsolateral prefrontal (dlPFC) in cTIV. However, dimension 7 in non-TIV included lateral inferior frontal regions except for the area posterior 9-46v included in the dlPFC subdivision of the HCPex. Dimension 8 included dlPFC in cTIV and non-TIV, but dimension 8 cTIV included a reduced number of dlPFC regions compared to non-TIV (See Online Resource 1 for a complete table of ROIs and their corresponding dimension).\u003c/p\u003e \u003cp\u003eNetwork loadings, considered equivalent to factor loadings, are computed by standardizing node strength (sum of edge weights) within each dimension. They are standardized measures that quantify each variable´s contribution to the emergence of coherent dimensions within a network structure. These network loadings use regularized partial correlations and loadings below 0.15 are considered a small effect. The higher network loadings for each dimension (from 1st to 9th dimensions) in the cTIV (within parenthesis if not correspondence in non-cTIV) indicators were: Parahippocampal area 3, area dorsal 23 a + b, dorsal transitional visual area, area 10r (first 10v, second 10r), anterior 24 prime, area 47s, area IFSp, superior 6–8 transitional area and nucleus accumbens (NAcc) (globus pallidus internalis (GPi), second NAcc) (See Online Resource 3 for a complete table of network loadings).\u003c/p\u003e \u003cp\u003eAttending to the same dimensions identified by EGA in non-TIV and cTIV structural data we identified nine dimensions: 1, the BIS/FFFS; 2, the CDS/PCC1; 3, the CDS/PCC2; 4, the ventromedial CDS; 5, the dorsomedial CDS; 6, the CVS; 7, the lateral inferior CDS; 8, the lateral superior CDS; and 9, the BAS.\u003c/p\u003e \u003cp\u003eBased on the observed TEFI values, we continued working with the cTIV data, as they suggested a better structural fit.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eCorrected sMRI items and dimension stability\u003c/h2\u003e \u003cp\u003eThe analysis of structural cTIV stability involved deleting the indicators with a stability \u0026lt; 0.75, as previously indicated (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). In the first step, 6 indicators were excluded following this criterion. Also, dimensions 3, 6 and 7 had stability values below the threshold (dimensions stability range [0.72–0.98]), though only dimension 3 presented item stability average below threshold (0.72). Finally, all indicator stability was over threshold [range 0.80 -1], as well as dimensions stability (range [0.80 -1]), and average item stability (range [0.94-1]).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab2\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBootEGA sMRI dimension and item stability results.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"12\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"12\"\u003e \u003cp\u003e\u003cem\u003eDimension stability\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eDimension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eConsistency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.822\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.796\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.558\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.958\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.772\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.302\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.328\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.954\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"12\"\u003e \u003cp\u003e\u003cem\u003eItem stability\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eHCPex Region\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eStability\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eProStriate_Area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.558\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003ePosterior_OFC_Complex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.742\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eOrbital_Frontal_Complex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.742\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_posterior_47r\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.504\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_9-46d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.384\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_anterior_9-46v\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.504\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eThe final EGA analysis revealed 9 dimensions on cTIV data after removing unstable items of a network with 71 nodes, 428 edges and edge density of 0.17 with mean edge weight of 0.076 (\u003cem\u003eSD\u003c/em\u003e = 0.120) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Lambda = 0.0898 (\u003cem\u003en\u003c/em\u003e = 100, ratio = 0.1). In this case, all dimensions have stability values above threshold and no unstable items (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). However, TEFI value increased to -76.581. For the same set of data, the PA yielded 31 dimensions. The discrepancies between the real and simulated eigenvalues from the ninth factor onward were (0.985 − 0.038). Network loadings run from small (-0.395) to large (0.656) in cTIV.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab3\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ecTIV dimension stability after bootEGA.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eDimension\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eConsistency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.996\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.802\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.956\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.848\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.936\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.916\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.898\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eEven after eliminating indicators due to instability, EGA maintains the number of dimensions as previously identified in the initial EGA analysis for the structural data: 1, the BIS/FFFS; 2, the CDS/PCC1; 3, the CDS/PCC2; 4, the ventromedial CDS; 5, the dorsomedial CDS; 6, the CVS; 7, the lateral inferior CDS; 8, the lateral superior CDS; and 9, the BAS.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe higher network loadings for each dimension in the cTIV indicators (indicated if different in cTIV before bootEGA) were: Parahippocampal area 3, area dorsal 23 a + b, dorsal transitional visual area, area 10r, anterior 24 prime, area 47M (before bootstrap area 47s, now second), area posterior 9-46v (before bootstrap area IFSp, now second), superior 6–8 transitional area and NAcc (See Online Resource 3 for a complete table of network loadings).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eFunctional data\u003c/h2\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eEGA on naturalistic fMRI data before bootEGA\u003c/h2\u003e \u003cp\u003eAn EGA analysis was carried with the averaged functional dataset previously described in the methods section. This revealed a set of 8 dimensions, compared to the 10 dimensions calculated by the PA where the discrepancies of real and simulated values were (0.156 − 0.114). The EGA analysis resulted in a network with 77 nodes, 287 edges and edge density of 0.098 with mean edge weight of 0.08 (\u003cem\u003eSD\u003c/em\u003e = 0.094), also network loadings ranged from − 0.403 to 0.733. Lambda = 0.3176 (\u003cem\u003en\u003c/em\u003e = 100, ratio = 0.1) and a TEFI value of -147.689. In this case, dimensions 1, 2 and 5 show items from the BIS/FFFS, while dimensions 5 and 8 contain BAS items. Also, dimension 2 gathered regions of the PCC. Then, dimension 3 combined ACC regions with a PCC region, as well as inferior frontal and dorsolateral prefrontal areas. As for the structural data, dimension 4 contained regions of the vmPFC and ACC, including some regions from the OFC and vlPFC. For the functional data, dimension 5 joined ACC and OFC regions with the amygdala and septal nucleus. Dimension 6 included some lateral OFC regions. Finally, lateral inferior frontal and dorsolateral prefrontal regions were included in dimension 7 (See Online Resource 1 for a complete table of ROIs and their corresponding dimension).\u003c/p\u003e \u003cp\u003eThe higher network loadings for each dimension in the naturalistic fMRI indicators were: Parahippocampal area 1, area 7m, area posterior 24, area10v, NAcc, area 47l, area 8Av and GPi (See Online Resource 3 for a complete table of network loadings).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eCorrected fMRI items and dimension stability\u003c/h2\u003e \u003cp\u003eAgain, following the criteria of deleting indicators with stability lower than 0.75 except theoretically sound regions, 20 indicators were removed from the analysis. Dimensions from 2 to 7 presented stability values below threshold (dimensions stability range [0.206–0.85]), with dimensions 5 and 6 having average item stability lower than 0.75 (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab4\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBootEGA fMRI dimension and item stability results.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"12\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"12\"\u003e \u003cp\u003e\u003cem\u003eDimension stability\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eDimension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eConsistency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.206\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"12\"\u003e \u003cp\u003e\u003cem\u003eItem stability\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eHCPex Region\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eStability\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eHCPex Region\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003eStability\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eEntorhinal_Cortex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.506\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.066\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003ePerirhinal_Ectorhinal_Cortex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.506\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_47l\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.066\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_23d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.532\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_anterior_47r\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.064\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eProStriate_Area\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.442\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_IFSa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_IFSp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.732\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eAnterior_24_prime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.678\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_posterior_47r\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.632\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_Posterior_24_prime\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.538\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_8C\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.708\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003ePosterior_OFC_Complex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eArea_9-46d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.57\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_13l\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.476\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eNucleus_Accumbens\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_47m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.582\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eAmygdala\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_47s\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e \u003cp\u003eSeptal_nucleus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003eArea_44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\"\u003e \u003cp\u003e0.728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003eThe EGA analysis, carried after cleaning unstable items from the bootstrap results, using the parameters previously indicated in the \"Methods\" section reported 6 dimensions of a network with 57 nodes, 210 edges and edge density of 0.132 with mean edge weight of 0.096 (\u003cem\u003eSD\u003c/em\u003e = 0.110) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Lambda = 0.2347 (\u003cem\u003en\u003c/em\u003e = 100, ratio = 0.1). As with the structural data, the TEFI value increased to -102.055, and the network loading range was − 0.049 to 0.669. For this dataset, PA estimated 10 dimensions again, with discrepancies between the real and simulated eigenvalues from the ninth factor onward were (0.682 − 0.004).\u003c/p\u003e \u003cp\u003eIn this case, after eliminating indicators due to instability (some indicators have been retained due to their theoretical relevance), EGA yields unstable dimensions for the BIS in this data and identifies the following dimensions for the functional data: 1, the BIS/FFFS; 2, the CDS/PCC; 3, the ventromedial CDS; 4, the dorsomedial and lateral inferior CDS/CVS; 5, the lateral superior CDS; and 6, the BAS (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab5\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003efMRI dimension stability after bootEGA.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003eDimension\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eConsistency\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.584\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.932\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.926\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.998\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/table\u003e\u003c/div\u003e \u003cp\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe higher network loadings for each dimension in the naturalistic fMRI after bootstrap (noted if different before bootEGA) indicators were: Parahippocampal area 1 (NAcc appeared in a separate dimension before bootstrap and is second after bootEGA), area 7m, area10v, area posterior 24, Area 8Av and substantia nigra pars compacta (SNpc) (before it was GPi, which is the second one in this case).\u003c/p\u003e \u003cp\u003eThe higher network loadings items for each dimension in the naturalistic fMRI compared to those of cTIV after bootEGA (cTIV differences noted in parentheses) were: Parahippocampal area 1 (parahippocampal area 3), Area 7m (area dorsal 23a + b) for the dimensions BIS/FFFS and CDS/PCC (CDS/PCC1 and CDS/PCC2), respectively; area 10v (area 10r, area 10v was second for cTIV) for the ventromedial CDS; area posterior 24 (anterior 24 prime) in the dorsomedial and lateral inferior CDS/CVS (dorsomedial CDS and CVS); area 8Av (superior 6–8 transitional area) for the lateral superior CDS (the lateral inferior and lateral superior CDS); and SNpc, followed by GPi (NAcc had the higher loading followed by SNpc and GPi) for the BAS (See Online Resource 3 for a complete table of network loadings).\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"DISCUSSION AND CONCLUSION","content":"\u003cp\u003eThis study provides new insights into the neurobiological foundations of the RST proposing a more detailed topology than our previous proposal (Lacomba-Arnau et al., \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). While structural and functional MRI indicators suggest a similar number of dimensions, we interpret this cross-modality convergence as evidence for a shared system architecture that allows for modality-specific expression. Structural covariance may reflect long-term organizational constraints shaped by development, genetics, whereas naturalistic functional covariance may reflect the deployment of these constraints under rich, ecological stimulation. Specifically, the BIS-FFFS, the BAS and the CVS showed differentiated structural dimensionality keeping in agreement with our previous study. Notably, the CDS showed a dimensional division that gets simplified after bootstrap stability analysis on fMRI indicators. These results suggest that while brain dimensional organization identified by EGA keeps agreement with latent factors suggested by the RST (Kennis et al., \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e; Lacomba-Arnau et al., \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e), cortical organization may offer more complex parcellations in different neurobiological systems. Within the network psychometrics framework, these systems are treated as emergent organizational units reflected in patterns of conditional dependence among regions. While this does not establish causality, it provides a formal representation of coordinated system architecture consistent with the RST’s system-level principles.\u003c/p\u003e\u003cp\u003eOur findings confirm that the structural covariance of brain regions identified with the BIS-FFFS and the BAS systems remains coupled (Lacomba-Arnau et al., \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). However, while our previous study suggested these systems were latent causes describing brain organization, the current EGA results extend these parcels into finer sub-divisions. Specifically, the BIS-FFFS involves three partitions of the parahippocampi as well as the perirhinal system besides the entorhinal, adding layers to the differentiation found previously. Notably, the emergence of parahippocampal regions as highly central within the BIS/FFFS dimension suggests that, in structural covariance, the system may be organized around contextual-mnemonic circuitry rather than localized in a single threat-detection region. By contrast, the nucleus accumbens emerging as central within the BAS dimension is consistent with its well-established involvement in incentive motivation and approach-related circuitry. From a NP approach, these regions are viewed as a complex system of mutually reinforcing variables; a structural change in one region is statistically associated with others within the identified neurobiological system (Christensen et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e; Christensen \u0026amp; Golino, \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe interpretation of structural and functional datasets provides complementary insights into RST dimensionality. Importantly, partial differences between structural and functional dimensionality are consistent with contemporary accounts of distributed brain organization, which emphasize that psychological functions arise from coordinated interactions among regions rather than isolated modules (Pessoa, \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e). From this perspective, structural covariance may reflect relatively stable anatomical constraints, likely influence of genetic, environmental, or neuroplasticity (e.g., learning) effects influenced by genetic, environmental, or neuroplasticity effects, whereas functional covariance captures the dynamic deployment of these systems under specific contextual demands. Importantly, naturalistic fMRI paradigms reliably elicit coordinated large-scale activity patterns that generalize across individuals, capturing functional network organization under ecologically meaningful conditions (Hasson et al., \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e; Sonkusare et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). The functional covariance structure observed here reflects the between-subject regularities in how these systems are engaged across complex environmental inputs, such as movie-watching. While structural dimensions represent the enduring organizational architecture, future studies are recommended to further explore the specific influence of genetic or environmental factors on these dimensions.\u003c/p\u003e\u003cp\u003eBecause RST concerns individual differences in motivational (McNaughton \u0026amp; Gray, \u003cspan class=\"CitationRef\"\u003e2000\u003c/span\u003e), the critical question is how such stability is instantiated neurally. Personality is typically defined as relatively stable patterns of behavior, affect, and cognition across situations, although expressed in context-dependent ways. Within this framework, these traits reflect stable differences in the sensitivity of brain systems mediating responses to reward and punishment (Corr \u0026amp; Matthews, \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e). Our findings suggest that structural and functional dimensions capture complementary aspects of these systems: structural covariance reflecting relatively stable system architecture, and functional covariance reflecting context-sensitive system expression. This suggests that while these dimensions capture stable individual differences in neurobiological responsiveness, they may also reflect stimulus-driven regularities. Distinguishing trait-like system sensitivity from stimulus-specific engagement remains an important goal for future work.\u003c/p\u003e\u003cp\u003eIn the context of NP, factors in a network model “emerge” from the causal connections between nodes (Christensen \u0026amp; Golino, \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e; Cramer et al., \u003cspan class=\"CitationRef\"\u003e2012\u003c/span\u003e). Thus, network loadings for a node are interpreted as its unique contribution to the emergence of dimension (Christensen et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e). We observed that the central nodes, those with higher network loadings, remain comparable across structural methodologies (cTIV and non-cTIV). Interestingly, the NAcc was the central node related to the BAS, while the central node of the BIS was the parahippocampus as discussed before. Furthermore, the CVS showed the involvement of the ventrolateral prefrontal cortex as a central node (IFSp), whereas we found a high diversity in the nodes of the partialized CDS, further suggesting a non-unitary control architecture.\u003c/p\u003e\u003cp\u003eRegarding the naturalistic fMRI dataset, the central nodes showed a high congruency before and after the bootstrap stability analysis, with the same central nodes for all dimensions except for the last one (and the ones absent after the bootstrap), reinforcing the robustness of the functional dimensions. However, a key distinction emerged between modalities: although the functional dimensions showed a separate BAS and BIS/FFF dimensions, central nodes like the NAcc (BAS) and amygdala (BIS/FFFS) cluster within a single dimension. This suggests that during complex, stimulus-driven activity, these systems may engage in a high degree of functional co-activation or shared variance that is not present in their underlying anatomical architecture. Such a finding underscores that the \"core\" regions of these systems may shift depending on whether one is measuring stable structural blueprints or the dynamic, integrated response to ecologically rich stimuli.\u003c/p\u003e\u003cp\u003eThe subdivision of the CDS into multiple dimensions may indicate that ‘constraint’ is not a unitary system at the neuroanatomical level, but rather a set of partially dissociable control components that become visible with higher brain-parcellation. This refinement may help reconcile RST-inspired systems with broader accounts of prefrontal control architecture. Specifically, we observed that the dimension identified in our previous work (Lacomba-Arnau et al., \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e) was divided into several dimensions here. The results suggest that there is an effect of the sample and the number of indicators in the dimensions extracted, as the current study utilized the HCP sample and the HCPex parcellation rather than the home-sample and the Neuromorphometric atlas used previously.\u003c/p\u003e\u003cp\u003eThis research is not without limitations. Although entropy-based fit indices suggested potential overfactoring, the persistence of the nine-dimensional solution across bootstrapped structural analyses supports its robustness. Future work should examine whether the present dimensional solution represents a stable characterization of RST systems or whether broader, more parsimonious configurations better capture their functional organization across different samples and tasks. Finally, regarding ROI selection, some regions were chosen based on correspondence with the Neuromorphometric atlas used in our previous research (Lacomba-Arnau et al., \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e) rather than their relevance in the systems (e.g., the posterior striate). Others showed proximity effects, such as the septal nucleus that always goes with the NAcc. It is important to signal those effects as an alternative hypothesis.\u003c/p\u003e\u003cp\u003eIn sum, our findings provide a systems-level test of a theory-defined architecture using multivariate brain covariance structure. Rather than treating MRI measures as direct proxies of isolated psychological processes, we demonstrate that theory-driven systems can be evaluated as reproducible organizational dimensions within distributed neural networks. This shifts the focus of personality neuroscience from region-based inference to the validation of system-level architectures, where stable dispositions, encompassing motivational, defensive, and regulatory processes, are expressed as coordinated patterns of structural and functional organization. By demonstrating that psychological constructs can be formally integrated with biological phenotypes, this study confirms that the neuroconceptual nervous system proposed by the RST is directly reflected in the brain’s intrinsic organizational dependencies. Ultimately, this approach contributes to a cumulative and testable framework for psychometric neuroscience, bridging the gap between behavioral theory and neurobiological phenotypes\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors have no relevant financial or non-financial interests to disclose or potential conflicts of interest.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThe publication is part of project PID2021-127340NB-C21, funded by Ministerio de Ciencia e Innovaci\u0026oacute;n, Spain (Grant No. MCIN/AEI/\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.13039/501100011033/FEDER\u003c/span\u003e\u003cspan address=\"10.13039/501100011033/FEDER\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, European Union).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eOPD performed data curation and formal analysis, wrote the original draft, and led the review and editing of the final manuscript. ELA contributed to the interpretation of the results and the review of the manuscript. AMM participated in the conceptualization of the study and the development of the initial methodology, and the review of the manuscript. ABL conceptualized and supervised the study, participated in the interpretation of the data, and contributed to the writing of the original draft.All authors approved the final version of the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe publication is part of project PID2021-127340NB-C21, funded by Ministerio de Ciencia e Innovaci\u0026oacute;n, Spain (Grant No. MCIN/AEI/10.13039/501100011033/FEDER, European Union).\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe neuroimaging MRI data used in this study are publicly available through the Human Connectome Project (HCP) Young Adult dataset (https://www.humanconnectome.org/). The data were collected and shared by the Washington University-University of Minnesota (WU-Minn HCP) Consortium. Access to the data is subject to the HCP\u0026rsquo;s terms of use, which require registration and acceptance of Data Use Terms.The custom R scripts used for data curation, TIV correction, and Exploratory Graph Analysis (EGA) are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBolt, T., Prince, E. B., Nomi, J. S., Messinger, D., Llabre, M. M., \u0026amp; Uddin, L. Q. (2018). Combining region- and network-level brain-behavior relationships in a structural equation model. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e165\u003c/em\u003e, 158\u0026ndash;169. https://doi.org/10.1016/j.neuroimage.2017.10.007\u003c/li\u003e\n\u003cli\u003eBorsboom, D. (2006). The Attack of the Psychometricians. \u003cem\u003ePsychometrika\u003c/em\u003e, \u003cem\u003e71\u003c/em\u003e(3), 425\u0026ndash;440. https://doi.org/10.1007/S11336-006-1447-6\u003c/li\u003e\n\u003cli\u003eChristensen, A. P., Garrido, L. E., \u0026amp; Golino, H. (2023). Unique Variable Analysis: A Network Psychometrics Method to Detect Local Dependence. \u003cem\u003eMultivariate Behavioral Research\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(6), 1165\u0026ndash;1182. https://doi.org/10.1080/00273171.2023.2194606\u003c/li\u003e\n\u003cli\u003eChristensen, A. P., \u0026amp; Golino, H. (2021). Estimating the Stability of Psychological Dimensions via Bootstrap Exploratory Graph Analysis: A Monte Carlo Simulation and Tutorial. \u003cem\u003ePsych 2021, Vol. 3, Pages 479-500\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(3), 479\u0026ndash;500. https://doi.org/10.3390/PSYCH3030032\u003c/li\u003e\n\u003cli\u003eChristensen, A. P., Golino, H., Abad, F. J., \u0026amp; Garrido, L. E. (2025). Revised network loadings. \u003cem\u003eBehavior Research Methods 2025 57:4\u003c/em\u003e, \u003cem\u003e57\u003c/em\u003e(4), 114-. https://doi.org/10.3758/S13428-025-02640-3\u003c/li\u003e\n\u003cli\u003eChristensen, A. P., Golino, H., \u0026amp; Silvia, P. J. (2020). A Psychometric Network Perspective on the Validity and Validation of Personality Trait Questionnaires. \u003cem\u003eEuropean Journal of Personality\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(6), 1095\u0026ndash;1108. https://doi.org/10.1002/PER.2265;WGROUP:STRING:PUBLICATION\u003c/li\u003e\n\u003cli\u003eCooper, S. R., Jackson, J. J., Barch, D. M., \u0026amp; Braver, T. S. (2019). Neuroimaging of individual differences: A latent variable modeling perspective. \u003cem\u003eNeuroscience \u0026amp; Biobehavioral Reviews\u003c/em\u003e, \u003cem\u003e98\u003c/em\u003e, 29\u0026ndash;46. https://doi.org/10.1016/J.NEUBIOREV.2018.12.022\u003c/li\u003e\n\u003cli\u003eCorr, P. J., \u0026amp; Matthews, G. (2009). The Cambridge Handbook of Personality Psychology. In P. J. Corr \u0026amp; G. Matthews (Eds.), \u003cem\u003eThe Cambridge Handbook of Personality Psychology\u003c/em\u003e. Cambridge University Press. https://doi.org/10.1017/CBO9780511596544\u003c/li\u003e\n\u003cli\u003eCramer, A. O. J., van der Sluis, S., Noordhof, A., Wichers, M., Geschwind, N., Aggen, S. H., Kendler, K. S., \u0026amp; Borsboom, D. (2012). Dimensions of normal personality as networks in search of equilibrium: You can\u0026rsquo;t like parties if you don\u0026rsquo;t like people. \u003cem\u003eEuropean Journal of Personality\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(4), 414\u0026ndash;431. https://doi.org/10.1002/PER.1866;REQUESTEDJOURNAL:JOURNAL:10990984;PAGEGROUP:STRING:PUBLICATION\u003c/li\u003e\n\u003cli\u003eDuboc, V., Dufourcq, P., Blader, P., \u0026amp; Roussign\u0026eacute;, M. (2015). Asymmetry of the Brain: Development and Implications. \u003cem\u003eAnnual Review of Genetics\u003c/em\u003e, \u003cem\u003e49\u003c/em\u003e, 647\u0026ndash;672. https://doi.org/10.1146/ANNUREV-GENET-112414-055322\u003c/li\u003e\n\u003cli\u003eFinn, E. S., \u0026amp; Bandettini, P. A. (2021). Movie-watching outperforms rest for functional connectivity-based prediction of behavior. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e235\u003c/em\u003e, 117963. https://doi.org/10.1016/J.NEUROIMAGE.2021.117963\u003c/li\u003e\n\u003cli\u003eFormisano, E., \u0026amp; Goebel, R. (2003). Tracking cognitive processes with functional MRI mental chronometry. \u003cem\u003eCurrent Opinion in Neurobiology\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(2), 174\u0026ndash;181. https://doi.org/10.1016/S0959-4388(03)00044-8\u003c/li\u003e\n\u003cli\u003eGaser, C., Dahnke, R., Thompson, P. M., Kurth, F., \u0026amp; Luders, E. (2024). CAT: a computational anatomy toolbox for the analysis of structural MRI data. \u003cem\u003eGigaScience\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e. https://doi.org/10.1093/GIGASCIENCE/GIAE049\u003c/li\u003e\n\u003cli\u003eGlasser, M. F., Sotiropoulos, S. N., Wilson, J. A., Coalson, T. S., Fischl, B., Andersson, J. L., Xu, J., Jbabdi, S., Webster, M., Polimeni, J. R., Van Essen, D. C., \u0026amp; Jenkinson, M. (2013). The Minimal Preprocessing Pipelines for the Human Connectome Project. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e80\u003c/em\u003e, 105. https://doi.org/10.1016/J.NEUROIMAGE.2013.04.127\u003c/li\u003e\n\u003cli\u003eGolino, H. F., \u0026amp; Epskamp, S. (2017). Exploratory graph analysis: A new approach for estimating the number of dimensions in psychological research. \u003cem\u003ePLOS ONE\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(6), e0174035. https://doi.org/10.1371/JOURNAL.PONE.0174035\u003c/li\u003e\n\u003cli\u003eGolino, H., Moulder, R., Shi, D., Christensen, A. P., Garrido, L. E., Nieto, M. D., Nesselroade, J., Sadana, R., Thiyagarajan, J. A., \u0026amp; Boker, S. M. (2021). Entropy Fit Indices: New Fit Measures for Assessing the Structure and Dimensionality of Multiple Latent Variables. \u003cem\u003eMultivariate Behavioral Research\u003c/em\u003e, \u003cem\u003e56\u003c/em\u003e(6), 1\u0026ndash;29. https://doi.org/10.1080/00273171.2020.1779642\u003c/li\u003e\n\u003cli\u003eGolino, H., Shi, D., Christensen, A. P., Garrido, L. E., Nieto, M. D., Sadana, R., Thiyagarajan, J. A., \u0026amp; Mart\u0026iacute;nez-Molina, A. (2020). Investigating the performance of Exploratory Graph Analysis and traditional techniques to identify the number of latent factors: A simulation and tutorial. \u003cem\u003ePsychological Methods\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(3), 292\u0026ndash;320. https://doi.org/10.1037/MET0000255\u003c/li\u003e\n\u003cli\u003eHaslbeck, J. M. B., \u0026amp; Waldorp, L. (2015). Structure estimation for mixed graphical models in high-dimensional data. \u003cem\u003eArXiv: Applications\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eHaslbeck, J. M. B., \u0026amp; Waldorp, L. J. (2020). mgm: Estimating Time-Varying Mixed Graphical Models in High-Dimensional Data. \u003cem\u003eJournal of Statistical Software\u003c/em\u003e, \u003cem\u003e93\u003c/em\u003e, 1\u0026ndash;46. https://doi.org/10.18637/jss.v093.i08\u003c/li\u003e\n\u003cli\u003eHasson, U., Malach, R., \u0026amp; Heeger, D. J. (2010). Reliability of cortical activity during natural stimulation. \u003cem\u003eTrends in Cognitive Sciences\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 40\u0026ndash;48. https://doi.org/10.1016/j.tics.2009.10.011\u003c/li\u003e\n\u003cli\u003eHuang, C. C., Rolls, E. T., Feng, J., \u0026amp; Lin, C. P. (2022). An extended Human Connectome Project multimodal parcellation atlas of the human cortex and subcortical areas. \u003cem\u003eBrain Structure and Function\u003c/em\u003e, \u003cem\u003e227\u003c/em\u003e(3), 763\u0026ndash;778. https://doi.org/10.1007/S00429-021-02421-6,\u003c/li\u003e\n\u003cli\u003eKennis, M., Rademaker, A. R., \u0026amp; Geuze, E. (2013). Neural correlates of personality: An integrative review. \u003cem\u003eNeuroscience and Biobehavioral Reviews\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(1), 73\u0026ndash;95. https://doi.org/10.1016/j.neubiorev.2012.10.012\u003c/li\u003e\n\u003cli\u003eKjellstr\u0026ouml;m, S., \u0026amp; Golino, H. (2019). Mining concepts of health responsibility using text mining and exploratory graph analysis. \u003cem\u003eScandinavian Journal of Occupational Therapy\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(6), 395\u0026ndash;410. https://doi.org/10.1080/11038128.2018.1455896\u003c/li\u003e\n\u003cli\u003eLacomba-Arnau, E., Mart\u0026iacute;nez-Molina, A., Garrido, L. E., \u0026amp; Barr\u0026oacute;s-Loscertales, A. (2025). Neural Topologies of Reinforcement Sensitivity Theory: A Latent Variable Approach to Magnetic Resonance Imaging Data. \u003cem\u003eBiological Psychiatry Global Open Science\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(5). https://doi.org/10.1016/j.bpsgos.2025.100526\u003c/li\u003e\n\u003cli\u003eLaurent, M.-A., Jacques, C., Yan, X., Jurczynski, P., Colnat-Coulbois, S., Maillard, L., Le Cam, S., Ranta, R., Cottereau, B. R., Koessler, L., Jonas, J., \u0026amp; Rossion, B. (2025). A tight relationship between BOLD fMRI activation/deactivation and increase/decrease in single neuron responses in human association cortex. \u003cem\u003eELife\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e. https://doi.org/10.7554/ELIFE.104779\u003c/li\u003e\n\u003cli\u003eLogothetis, N. K. (2003). The Underpinnings of the BOLD Functional Magnetic Resonance Imaging Signal. \u003cem\u003eJournal of Neuroscience\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(10), 3963\u0026ndash;3971. https://doi.org/10.1523/JNEUROSCI.23-10-03963.2003\u003c/li\u003e\n\u003cli\u003eLogothetis, N. K., Pauls, J., Augath, M., Trinath, T., \u0026amp; Oeltermann, A. (2001). Neurophysiological investigation of the basis of the fMRI signal. \u003cem\u003eNature 2001 412:6843\u003c/em\u003e, \u003cem\u003e412\u003c/em\u003e(6843), 150\u0026ndash;157. https://doi.org/10.1038/35084005\u003c/li\u003e\n\u003cli\u003eMcNaughton, N., \u0026amp; Gray, J. A. (2000). Anxiolytic action on the behavioural inhibition system implies multiple types of arousal contribute to anxiety. \u003cem\u003eJournal of Affective Disorders\u003c/em\u003e, \u003cem\u003e61\u003c/em\u003e(3), 161\u0026ndash;176. https://doi.org/10.1016/S0165-0327(00)00344-X\u003c/li\u003e\n\u003cli\u003eMechelli, A., Friston, K. J., Frackowiak, R. S., \u0026amp; Price, C. J. (2005). Structural covariance in the human cortex. \u003cem\u003eThe Journal of Neuroscience : The Official Journal of the Society for Neuroscience\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(36), 8303\u0026ndash;8310. https://doi.org/10.1523/JNEUROSCI.0357-05.2005\u003c/li\u003e\n\u003cli\u003eMenon, R. S., Luknowsky, D. C., \u0026amp; Gati, J. S. (1998). Mental chronometry using latency-resolved functional MRI. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e95\u003c/em\u003e(18), 10902\u0026ndash;10907. https://doi.org/10.1073/pnas.95.18.10902\u003c/li\u003e\n\u003cli\u003eNordenskj\u0026ouml;ld, R., Malmberg, F., Larsson, E. M., Simmons, A., Ahlstr\u0026ouml;m, H., Johansson, L., \u0026amp; Kullberg, J. (2015). Intracranial volume normalization methods: Considerations when investigating gender differences in regional brain volume. \u003cem\u003ePsychiatry Research: Neuroimaging\u003c/em\u003e, \u003cem\u003e231\u003c/em\u003e(3), 227\u0026ndash;235. https://doi.org/10.1016/J.PSCYCHRESNS.2014.11.011\u003c/li\u003e\n\u003cli\u003ePessoa, L. (2014). Understanding brain networks and brain organization. \u003cem\u003ePhysics of Life Reviews\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(3), 400. https://doi.org/10.1016/j.plrev.2014.03.005\u003c/li\u003e\n\u003cli\u003ePoldrack, R. A. (2006). Can cognitive processes be inferred from neuroimaging data? \u003cem\u003eTrends in Cognitive Sciences\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(2), 59\u0026ndash;63. https://doi.org/10.1016/j.tics.2005.12.004\u003c/li\u003e\n\u003cli\u003ePoldrack, R. A., Baker, C. I., Durnez, J., Gorgolewski, K. J., Matthews, P. M., Munaf\u0026ograve;, M. R., Nichols, T. E., Poline, J. B., Vul, E., \u0026amp; Yarkoni, T. (2017). Scanning the horizon: towards transparent and reproducible neuroimaging research. \u003cem\u003eNature Reviews. Neuroscience\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(2), 115. https://doi.org/10.1038/nrn.2016.167\u003c/li\u003e\n\u003cli\u003eRowley, C. D., Bazin, P. L., Tardif, C. L., Sehmbi, M., Hashim, E., Zaharieva, N., Minuzzi, L., Frey, B. N., \u0026amp; Bock, N. A. (2015). Assessing intracortical myelin in the living human brain using myelinated cortical thickness. \u003cem\u003eFrontiers in Neuroscience\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(OCT). https://doi.org/10.3389/fnins.2015.00396\u003c/li\u003e\n\u003cli\u003eSchmittmann, V. D., Cramer, A. O. J., Waldorp, L. J., Epskamp, S., Kievit, R. A., \u0026amp; Borsboom, D. (2013). Deconstructing the construct: A network perspective on psychological phenomena. \u003cem\u003eNew Ideas in Psychology\u003c/em\u003e, \u003cem\u003e31\u003c/em\u003e(1), 43\u0026ndash;53. https://doi.org/10.1016/J.NEWIDEAPSYCH.2011.02.007\u003c/li\u003e\n\u003cli\u003eSmith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens, T. E. J., Johansen-Berg, H., Bannister, P. R., De Luca, M., Drobnjak, I., Flitney, D. E., Niazy, R. K., Saunders, J., Vickers, J., Zhang, Y., De Stefano, N., Brady, J. M., \u0026amp; Matthews, P. M. (2004). Advances in functional and structural MR image analysis and implementation as FSL. \u003cem\u003eNeuroImage\u003c/em\u003e, \u003cem\u003e23 Suppl 1\u003c/em\u003e(SUPPL. 1). https://doi.org/10.1016/J.NEUROIMAGE.2004.07.051\u003c/li\u003e\n\u003cli\u003eSonkusare, S., Breakspear, M., \u0026amp; Guo, C. (2019). Naturalistic Stimuli in Neuroscience: Critically Acclaimed. \u003cem\u003eTrends in Cognitive Sciences\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(8), 699\u0026ndash;714. https://doi.org/10.1016/j.tics.2019.05.004\u003c/li\u003e\n\u003cli\u003eSoreq, E., Violante, I. R., Daws, R. E., \u0026amp; Hampshire, A. (2021). Neuroimaging evidence for a network sampling theory of individual differences in human intelligence test performance. \u003cem\u003eNature Communications 2021 12:1\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(1), 2072-. https://doi.org/10.1038/s41467-021-22199-9\u003c/li\u003e\n\u003cli\u003eTesler, F., Linne, M. L., \u0026amp; Destexhe, A. (2023). Modeling the relationship between neuronal activity and the BOLD signal: contributions from astrocyte calcium dynamics. \u003cem\u003eScientific Reports 2023 13:1\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(1), 6451-. https://doi.org/10.1038/s41598-023-32618-0\u003c/li\u003e\n\u003cli\u003eVan der Weele, T. J., \u0026amp; Vansteelandt, S. (2022). A Statistical Test to Reject the Structural interpretation of a Latent Factor Model. \u003cem\u003eJournal of the Royal Statistical Society Series B: Statistical Methodology\u003c/em\u003e, \u003cem\u003e84\u003c/em\u003e(5), 2032\u0026ndash;2054. https://doi.org/10.1111/RSSB.12555\u003c/li\u003e\n\u003cli\u003eYarkoni, T. (2022). The generalizability crisis. \u003cem\u003eBehavioral and Brain Sciences\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e, e1. https://doi.org/10.1017/S0140525X20001685\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"brain-structure-and-function","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bsaf","sideBox":"Learn more about [Brain Structure and Function](https://www.springer.com/journal/429)","snPcode":"429","submissionUrl":"https://submission.nature.com/new-submission/429/3","title":"Brain Structure and Function","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Exploratory Graph Analysis, Reinforcement Sensitivity Theory, Magnetic Resonance Imaging, Biological Phenotypes","lastPublishedDoi":"10.21203/rs.3.rs-9179108/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9179108/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study evaluates the neurobiological architecture of Reinforcement Sensitivity Theory (RST) using a network psychometric approach. We shift from region-based inference to system-level organization by studying structural and functional brain dimensionality. Using the Human Connectome Project dataset (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1113 structural MRI; \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;176 naturalistic functional MRI), we applied Exploratory Graph Analysis (EGA) to 77 regions of interest from the HCPex atlas contingent to the RST. Bootstrap EGA assessed dimensional stability and item consistency. Structural results identified nine dimensions: the Behavioral Inhibition/Fight-Flight-Freeze (BIS/FFFS) and the Behavioral Approach (BAS) systems, a distinct Constraint Ventral Stream (CVS), and six partially dissociable Constraint Dorsal Stream (CDS) components. In contrast, functional data yielded a more integrated six-dimensional solution. While structural covariance reflects stable anatomical blueprints, functional networks reveal stimulus-driven co-activation, particularly between the BIS/FFFS and BAS. The subdivision of the CDS across both modalities supports a non-unitary control architecture visible only at high parcellation. Our findings demonstrate that RST systems emerge as coordinated, reproducible patterns of brain organization, providing a cumulative framework for linking psychological constructs to biological phenotypes in psychometric neuroscience.\u003c/p\u003e","manuscriptTitle":"Neuroanatomical and Functional Dimensionality of Reinforcement Sensitivity Theory Systems: An Exploratory Graph Analysis Approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-10 06:56:26","doi":"10.21203/rs.3.rs-9179108/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-04T09:54:41+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-15T13:00:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"294491476318779704915694372067727450297","date":"2026-04-04T07:20:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-03T16:18:30+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-23T13:14:39+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-23T03:50:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"Brain Structure and Function","date":"2026-03-20T12:42:06+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"brain-structure-and-function","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bsaf","sideBox":"Learn more about [Brain Structure and Function](https://www.springer.com/journal/429)","snPcode":"429","submissionUrl":"https://submission.nature.com/new-submission/429/3","title":"Brain Structure and Function","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"6dd39840-8a87-43ce-b4a5-9222670bba84","owner":[],"postedDate":"April 10th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-04T09:54:41+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-04T10:09:12+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-10 06:56:26","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9179108","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9179108","identity":"rs-9179108","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}