Advertisement

Social Health Is Associated With Tract-Specific Brain White Matter Microstructure In Community-Dwelling Older Adults

Open AccessPublished:September 08, 2022DOI:https://doi.org/10.1016/j.bpsgos.2022.08.009

      Abstract

      BACKGROUND

      Poor social health has been linked to risk of neuropsychiatric disorders. Neuroimaging studies have shown associations between social health and global white matter microstructural integrity. We aim to identify which white matter tracts are involved in these associations.

      METHODS

      Social health markers (loneliness, perceived social support, partnership status) and white matter microstructural integrity of 15 white matter tracts (identified with probabilistic tractography after diffusion magnetic resonance imaging) were collected for 3352 participants (mean age 58.4, 54.9% female) from 2002-2008 in the Rotterdam Study. Cross-sectional associations were studied using multivariable linear regression.

      RESULTS

      Loneliness was associated with higher mean diffusivity (MD) in the superior thalamic radiation and the parahippocampal part of the cingulum (standardized mean difference for both tracts: 0.21, 95%CI: 0.09; 0.34). Better perceived social support was associated with lower MD in the forceps minor (standardized mean difference per point increase in social support: -0.06, 95%CI: -0.09; -0.03), inferior fronto-occipital fasciculus and uncinate fasciculus. In male participants, better perceived social support was associated with lower MD in the forceps minor, and not having a partner was associated with lower fractional anisotropy (FA) in the forceps minor. Loneliness was associated with higher MD in the superior thalamic radiation in female participants only.

      CONCLUSIONS

      Social health was associated with tract-specific white matter microstructure. Loneliness was associated with lower integrity of limbic and sensorimotor tracts, whereas better perceived social support was associated with higher integrity of association and commissural tracts, indicating that social health domains involve distinct neural pathways of the brain.

      Introduction

      Social health, which entails how the interaction between the direct social environment and the individual affects the individual perception of social life as a domain of general wellbeing (
      • Huber M.
      • Knottnerus J.A.
      • Green L.
      • Horst Hvd
      • Jadad A.R.
      • Kromhout D.
      • et al.
      How should we define health?.
      ,
      • Vernooij-Dassen M.
      • Jeon Y.H.
      Social health and dementia: the power of human capabilities.
      ,
      • Droes R.M.
      • Chattat R.
      • Diaz A.
      • Gove D.
      • Graff M.
      • Murphy K.
      • et al.
      Social health and dementia: a European consensus on the operationalization of the concept and directions for research and practice.
      ,
      • Vernooij-Dassen M.
      • Moniz-Cook E.
      • Verhey F.
      • Chattat R.
      • Woods B.
      • Meiland F.
      • et al.
      Bridging the divide between biomedical and psychosocial approaches in dementia research: the 2019 INTERDEM manifesto.
      ), is fundamental for human survival, and is heavily dependent on complex neurocognitive systems (
      • Insel T.R.
      • Fernald R.D.
      How the brain processes social information: searching for the social brain.
      ). Social dysfunction is accordingly one of the first and most common signs of major neuropsychiatric disorders (
      • Porcelli S.
      • Van Der Wee N.
      • van der Werff S.
      • Aghajani M.
      • Glennon J.C.
      • van Heukelum S.
      • et al.
      Social brain, social dysfunction and social withdrawal.
      ). Importantly, poor social health is also associated with higher mortality and morbidity risk, including dementia (
      • Holt-Lunstad J.
      The Major Health Implications of Social Connection.
      ,
      • Holt-Lunstad J.
      • Smith T.B.
      • Baker M.
      • Harris T.
      • Stephenson D.
      Loneliness and social isolation as risk factors for mortality: a meta-analytic review.
      ,
      • Holt-Lunstad J.
      • Smith T.B.
      • Layton J.B.
      Social relationships and mortality risk: a meta-analytic review.
      ,
      • Howick J.
      • Kelly P.
      • Kelly M.
      Establishing a causal link between social relationships and health using the Bradford Hill Guidelines.
      ,
      • Malcolm M.
      • Frost H.
      • Cowie J.
      Loneliness and social isolation causal association with health-related lifestyle risk in older adults: a systematic review and meta-analysis protocol.
      ,
      • Holt-Lunstad J.
      Why Social Relationships Are Important for Physical Health: A Systems Approach to Understanding and Modifying Risk and Protection.
      ,
      • Holt-Lunstad J.
      The Potential Public Health Relevance of Social Isolation and Loneliness: Prevalence, Epidemiology, and Risk Factors.
      ).
      To understand the complex relations between social health and the brain in the risk of neuropsychiatric disorders, numerous neuroimaging studies have been performed, focusing recently especially on white matter (
      • Ameis S.H.
      • Catani M.
      Altered white matter connectivity as a neural substrate for social impairment in Autism Spectrum Disorder.
      ,
      • Wang Y.
      • Metoki A.
      • Alm K.H.
      • Olson I.R.
      White matter pathways and social cognition.
      ,
      • Wang Y.
      • Olson I.R.
      The Original Social Network: White Matter and Social Cognition.
      ). A recent study showed that social health is associated with global white matter microstructure in a general population of older adults (
      • van der Velpen I.F.
      • Melis R.J.F.
      • Perry M.
      • Vernooij-Dassen M.J.F.
      • Ikram M.A.
      • Vernooij M.W.
      Social Health Is Associated With Structural Brain Changes in Older Adults: The Rotterdam Study.
      ). Yet, the relationship between social health and specific white matter tracts remains unclear.
      Research in clinical populations of patients with conditions characterized by symptoms in the social domain, showed that schizophrenia (
      • Pettersson-Yeo W.
      • Allen P.
      • Benetti S.
      • McGuire P.
      • Mechelli A.
      Dysconnectivity in schizophrenia: where are we now?.
      ,
      • Jalbrzikowski M.
      • Villalon-Reina J.E.
      • Karlsgodt K.H.
      • Senturk D.
      • Chow C.
      • Thompson P.M.
      • et al.
      Altered white matter microstructure is associated with social cognition and psychotic symptoms in 22q11.2 microdeletion syndrome.
      ,
      • Jung S.
      • Kim J.H.
      • Sung G.
      • Ko Y.G.
      • Bang M.
      • Park C.I.
      • et al.
      Uncinate fasciculus white matter connectivity related to impaired social perception and cross-sectional and longitudinal symptoms in patients with schizophrenia spectrum psychosis.
      ), autism spectrum disorder (
      • Ameis S.H.
      • Catani M.
      Altered white matter connectivity as a neural substrate for social impairment in Autism Spectrum Disorder.
      ,
      • Li Y.
      • Zhou Z.
      • Chang C.
      • Qian L.
      • Li C.
      • Xiao T.
      • et al.
      Anomalies in uncinate fasciculus development and social defects in preschoolers with autism spectrum disorder.
      ), social anxiety (
      • Roelofs E.F.
      • Bas-Hoogendam J.M.
      • van Ewijk H.
      • Ganjgahi H.
      • van der Werff S.J.A.
      • Barendse M.E.A.
      • et al.
      Investigating microstructure of white matter tracts as candidate endophenotypes of Social Anxiety Disorder - Findings from the Leiden Family Lab study on Social Anxiety Disorder (LFLSAD).
      ), and social cognition deficits (
      • Filley C.M.
      Social Cognition and White Matter: Connectivity and Cooperation.
      ) are associated with poor microstructure of specific white matter tracts, such as the frontal and temporal thalamic projections (
      • Ameis S.H.
      • Catani M.
      Altered white matter connectivity as a neural substrate for social impairment in Autism Spectrum Disorder.
      ), the corpus callosum (
      • Filley C.M.
      Social Cognition and White Matter: Connectivity and Cooperation.
      ,
      • Saito J.
      • Hori M.
      • Nemoto T.
      • Katagiri N.
      • Shimoji K.
      • Ito S.
      • et al.
      Longitudinal study examining abnormal white matter integrity using a tract-specific analysis in individuals with a high risk for psychosis.
      ), the uncinate fasciculus (
      • Ameis S.H.
      • Catani M.
      Altered white matter connectivity as a neural substrate for social impairment in Autism Spectrum Disorder.
      ,
      • Jalbrzikowski M.
      • Villalon-Reina J.E.
      • Karlsgodt K.H.
      • Senturk D.
      • Chow C.
      • Thompson P.M.
      • et al.
      Altered white matter microstructure is associated with social cognition and psychotic symptoms in 22q11.2 microdeletion syndrome.
      ,
      • Jung S.
      • Kim J.H.
      • Sung G.
      • Ko Y.G.
      • Bang M.
      • Park C.I.
      • et al.
      Uncinate fasciculus white matter connectivity related to impaired social perception and cross-sectional and longitudinal symptoms in patients with schizophrenia spectrum psychosis.
      ,
      • Li Y.
      • Zhou Z.
      • Chang C.
      • Qian L.
      • Li C.
      • Xiao T.
      • et al.
      Anomalies in uncinate fasciculus development and social defects in preschoolers with autism spectrum disorder.
      ,
      • Filley C.M.
      Social Cognition and White Matter: Connectivity and Cooperation.
      ), the superior and inferior longitudinal fasciculus (
      • Roelofs E.F.
      • Bas-Hoogendam J.M.
      • van Ewijk H.
      • Ganjgahi H.
      • van der Werff S.J.A.
      • Barendse M.E.A.
      • et al.
      Investigating microstructure of white matter tracts as candidate endophenotypes of Social Anxiety Disorder - Findings from the Leiden Family Lab study on Social Anxiety Disorder (LFLSAD).
      ,
      • Filley C.M.
      Social Cognition and White Matter: Connectivity and Cooperation.
      ), and the inferior fronto-occipital fasciculus (
      • Jalbrzikowski M.
      • Villalon-Reina J.E.
      • Karlsgodt K.H.
      • Senturk D.
      • Chow C.
      • Thompson P.M.
      • et al.
      Altered white matter microstructure is associated with social cognition and psychotic symptoms in 22q11.2 microdeletion syndrome.
      ). In addition, some studies have focused on particular aspects of social life of individuals, such as social network size (
      • Noonan M.P.
      • Mars R.B.
      • Sallet J.
      • Dunbar R.I.M.
      • Fellows L.K.
      The structural and functional brain networks that support human social networks.
      ) and diversity (
      • Molesworth T.
      • Sheu L.K.
      • Cohen S.
      • Gianaros P.J.
      • Verstynen T.D.
      Social network diversity and white matter microstructural integrity in humans.
      ), indicating that there may be differentiated relationships between different social domain components and specific neural circuits, and that these relationships may also show sex differences (
      • Spreng R.N.
      • Dimas E.
      • Mwilambwe-Tshilobo L.
      • Dagher A.
      • Koellinger P.
      • Nave G.
      • et al.
      The default network of the human brain is associated with perceived social isolation.
      ).
      Based on these findings, social health may be associated with the integrity of specific white matter tracts in the general population. Knowledge on this association is instrumental to better understand the neurobiological mechanisms that underlie neuropsychiatric disorders like dementia and schizophrenia (
      • Porcelli S.
      • Van Der Wee N.
      • van der Werff S.
      • Aghajani M.
      • Glennon J.C.
      • van Heukelum S.
      • et al.
      Social brain, social dysfunction and social withdrawal.
      ). Specifically, suboptimal social health may contribute to risk of neuropsychiatric disorders through changes in specific white matter tracts.
      Therefore, the aim of this study was to determine the associations between different aspects of social health and tract-specific white matter microstructure in community-dwelling older adults. We hypothesize that worse social health is associated with worse integrity of white matter microstructure in tracts that are important for cognitive functioning.

      Methods

      Design and population

      This study was embedded in the Rotterdam Study, a prospective, population-based cohort study based in Rotterdam, the Netherlands. It started in 1990 and is ongoing (
      • Ikram M.A.
      • Brusselle G.
      • Ghanbari M.
      • Goedegebure A.
      • Ikram M.K.
      • Kavousi M.
      • et al.
      Objectives, design and main findings until 2020 from the Rotterdam Study.
      ). Persons aged ≥40 years living in the Ommoord neighbourhood were invited to participate and were followed up every 3 to 4 years. The Rotterdam Study has been approved by the Medical Ethics Committee of the Erasmus MC (registration number MEC 02.1015) and by the Dutch Ministry of Health, Welfare and Sport (Population Screening Act WBO, license number 1071272-159521-PG). The Rotterdam Study Personal Registration Data collection is filed with the Erasmus MC Data Protection Officer under registration number EMC1712001. The Rotterdam Study has been entered into the Netherlands National Trial Register (NTR; www.trialregister.nl) and into the WHO International Clinical Trials Registry Platform (ICTRP; https://apps.who.int/trialsearch/) under shared catalogue number NL6645/ NTR6831. All participants provided written informed consent to participate in the study and to have their information obtained from treating physicians.
      Social health markers were collected from January 2002 to November 2008. Magnetic Resonance Imaging (MRI) data were collected from March 2006 to March 2010. Median time difference between assessment of social health markers and MRI scan was 90 days (IQR 57-195 days). Diffusion tensor imaging (DTI) processing was complete for 3,526 participants, of whom 3,488 participants had complete information on social health markers. After exclusion of participants with prevalent dementia (N=53) and cortical brain infarcts (N=83), 3,352 participants were included in the study population.

      Social health markers

      Social health markers collected during the home interview in the Rotterdam Study are loneliness, perceived social support, and partnership status. Loneliness and perceived social support are subjective measures of the perception of social life, while partnership status is an objective measure that can influence the individual perception of social life and thus social health. Loneliness, defined as “the subjective experience of an unpleasant lack of (quality of) social relationships”, was assessed with a single-item question in the Center for Epidemiological Studies Depression scale (CES-D) (
      • Radloff L.S.
      The CES-D Scale:A Self-Report Depression Scale for Research in the General Population.
      ). We dichotomized the responses into lonely (feelings of loneliness ≥1 day per week) and not lonely (feelings of loneliness <1 day per week). Perceived social support was assessed with a 5-item questionnaire modified from the Health and Lifestyle Survey. The questions were: “I know people, among my family and friends, 1) who do things that make me happy; 2) whom I can always count on; 3) who would make sure that I would get help if I would need it; 4) who give me the feeling that I am important in their lives; and 5) who accept me for who I am.” Participants had for each question the following options: agree, somewhat agree, or disagree. The sum scores range from 0 to 10, a higher value corresponds to better perceived social support. Scores were weighted to account for responses with one missing item. Scores with less than 4 responses were excluded. Cronbach’s alpha for social support questionnaire was 0.74 (95%CI 0.73-0.74). The answer options for the partnership status were “married/has a partner”, “widowed/divorced”, and “never married”, which were subsequently dichotomized into “has a current partner” and “does not have a current partner”.

      MRI acquisition and processing

      Brain MRI was performed for all participants on a single 1.5-T MRI scanner (GE Signa Excite) with an 8-channel head coil, without any major hardware or software upgrades during the time of the study. A detailed protocol of MRI in the Rotterdam Study including quality control has been extensively described previously (
      • Ikram M.A.
      • van der Lugt A.
      • Niessen W.J.
      • Koudstaal P.J.
      • Krestin G.P.
      • Hofman A.
      • et al.
      The Rotterdam Scan Study: design update 2016 and main findings.
      ). In brief, structural imaging included a T1-weighted sequence, a proton density (PD) weighted sequence, a T2-weighted fluid-attenuated inversion recovery (FLAIR) sequence, and a 3D T2*-weighted gradient-recalled echo (GRE) scan. Diffusion weighted imaging is described below. Due to a technical issue, a subset of 1312 participants was scanned with phase and frequency encoding directions switched for the DTI sequence between February 2007 and May 2008. This introduced a mild ghosting artefact in the phase encoding direction, with was adjusted for in statistical analysis (see statistical analyses). Brain volumetric measures (gray matter, white matter and cerebrospinal fluid) were quantified using a k-nearest neighbor algorithm (
      • Vrooman H.A.
      • Cocosco C.A.
      • van der Lijn F.
      • Stokking R.
      • Ikram M.A.
      • Vernooij M.W.
      • et al.
      Multi-spectral brain tissue segmentation using automatically trained k-Nearest-Neighbor classification.
      ). White matter hyperintensities (WMH) were quantified with an automated post-processing step based on the tissue segmentation and FLAIR image (
      • de Boer R.
      • Vrooman H.A.
      • van der Lijn F.
      • Vernooij M.W.
      • Ikram M.A.
      • van der Lugt A.
      • et al.
      White matter lesion extension to automatic brain tissue segmentation on MRI.
      ). All tissue segmentations were inspected for quality by trained raters and manually corrected if necessary. Presence of cortical brain infarcts was visually assessed by trained raters.

      Diffusion-MRI processing and tractography

      To obtain diffusion tensor imaging data we performed a single shot, diffusion-weighted spin-echo echo-planar imaging sequence. The maximum b value was 1000 s/mm2 in 25 noncollinear directions; 3 volumes were acquired without diffusion weighting (b value = 0 s/mm2). All diffusion data were pre-processed using a standardized pipeline (
      • Koppelmans V.
      • de Groot M.
      • de Ruiter M.B.
      • Boogerd W.
      • Seynaeve C.
      • Vernooij M.W.
      • et al.
      Global and focal white matter integrity in breast cancer survivors 20 years after adjuvant chemotherapy.
      ). In short, eddy current and head motion corrections were performed on the diffusion-weighted volumes. Diffusion tensors were estimated using a nonlinear Levenberg Marquardt estimator, available in ExploreDTI (
      • Leemans A.
      • Jeurissen B.
      • Sijbers J.
      • Jones D.K.
      ExploreDTI: a graphical toolbox for processing, analyzing, and visualizing diffusion MR data.
      ). Tensors were fit using the resampled data, which allowed computation of global mean FA and MD in the normal-appearing white matter, in combination with the tissue segmentation. Diffusion data were next used to segment white matter tracts using a probabilistic tractography approach, which has been described in detail previously (
      • de Groot M.
      • Ikram M.A.
      • Akoudad S.
      • Krestin G.P.
      • Hofman A.
      • van der Lugt A.
      • et al.
      Tract-specific white matter degeneration in aging: The Rotterdam Study.
      ). In brief, tractography was performed using PROBTRACKX (
      • Behrens T.E.
      • Berg H.J.
      • Jbabdi S.
      • Rushworth M.F.
      • Woolrich M.W.
      Probabilistic diffusion tractography with multiple fibre orientations: What can we gain?.
      ), which is a probabilistic Bayesian framework for white matter tractography and is available in FSL (version 4.1.4) (
      • Jenkinson M.
      • Beckmann C.F.
      • Behrens T.E.
      • Woolrich M.W.
      • Smith S.M.
      ). For 15 different white matter tracts (12 of which were segmented bilaterally) tract-specific white matter microstructural diffusion-MRI parameters (median FA and MD) were obtained with subsequent averaging of left and right measures. Average reproducibility (R2) of tract-specific measurements was 0.87 (
      • de Groot M.
      • Ikram M.A.
      • Akoudad S.
      • Krestin G.P.
      • Hofman A.
      • van der Lugt A.
      • et al.
      Tract-specific white matter degeneration in aging: The Rotterdam Study.
      ). Tracts were categorized, based on anatomy or presumed function, into association tracts (anterior thalamic radiation, inferior fronto-occipital fasciculus, inferior longitudinal fasciculus, posterior thalamic radiation, superior longitudinal fasciculus, uncinate fasciculus), commissural tracts (forceps major, forceps minor), limbic system tracts (cingulate gyrus part of the cingulum, parahippocampal part of cingulum, fornix), and sensorimotor tracts (corticospinal tract, middle cerebellar peduncle, medial lemniscus, superior thalamic radiation) (Figure 1). Tract segmentations were used to obtain tract-specific white matter volumes and tract-specific white matter hyperintensity volumes by combining the tissue and tract segmentation. The cerebellum could not be fully incorporated in the field of view of the diffusion-MRI scan, resulting in partial coverage of the medial lemniscus at the lower border of the scan. Alternative seed masks for tractography were selected until reasonable coverage was achieved to overcome this problem (
      • de Groot M.
      • Ikram M.A.
      • Akoudad S.
      • Krestin G.P.
      • Hofman A.
      • van der Lugt A.
      • et al.
      Tract-specific white matter degeneration in aging: The Rotterdam Study.
      ). This correction was treated as a potential confounder in all models that included the medial lemniscus. Further information on quality control for DTI and white matter tract segmentation can be found in the Supplemental Methods.
      A better microstructure is characterised by high fractional anisotropy and low mean diffusivity (
      • Thomason M.E.
      • Thompson P.M.
      Diffusion imaging, white matter, and psychopathology.
      ). FA values were computed based on the ratio between the longest and shortest axes of diffusion, giving values between zero and one: zero indicates isotropic diffusion (equal in all directions), indicating an absence of organized fiber tracts to constrain directionality. A value closer to one means that diffusion occurs more strongly along one axis, suggesting increased fiber organization and white matter integrity.

      Other measurements

      We selected covariates in our study based on their potential to be a cause of the exposure or the outcome, of both, or as a proxy of unmeasured confounding. Intracranial volume was defined as the sum of total brain volume and cerebrospinal fluid on MRI. Educational attainment was assessed during the baseline interview and categorized according to UNESCO classification. Cognitive function was assessed with the Mini Mental State Examination (MMSE) at the research center. Smoking status was assessed during interview and categorized as former, never or current smoker. Alcohol consumption was classified into none (no alcoholic beverages), moderate (≤1 beverage/day) and heavy (>1 beverage/day), based on grams of alcohol per day calculated from interview data on number and type of alcoholic beverages consumed. Body mass index (kg/m2) was calculated from body height and weight assessed at the research center.
      Hypertension was defined as a systolic blood pressure >140 mmHg, or a diastolic blood pressure >90 mmHg, or taking antihypertensive medication. Blood pressure was measured twice in sitting position using a random-zero sphygmomanometer at the research center. The average of the two measurements was used to define hypertension. Diabetes mellitus was determined based on fasting serum glucose level (>7.0 mmol/L) or, if unavailable, nonfasting serum glucose level (>11.1 mmol/L), or the use of antidiabetic medication. The ascertainment of coronary heart disease, heart failure and clinical stroke diagnoses was based on medical records and has been described in detail elsewhere (
      • Ikram M.A.
      • Brusselle G.
      • Ghanbari M.
      • Goedegebure A.
      • Ikram M.K.
      • Kavousi M.
      • et al.
      Objectives, design and main findings until 2020 from the Rotterdam Study.
      ).
      Depressive symptoms were assessed using the CES-D during the home interview. CES-D scores were weighted to account for missing responses if missing values were less than 25% (
      • Radloff L.S.
      The CES-D Scale:A Self-Report Depression Scale for Research in the General Population.
      ,
      • Beekman A.T.
      • Deeg D.J.
      • Van Limbeek J.
      • Braam A.W.
      • De Vries M.Z.
      • Van Tilburg W.
      Criterion validity of the Center for Epidemiologic Studies Depression scale (CES-D): results from a community-based sample of older subjects in The Netherlands.
      ). Presence of anxiety disorders was assessed using an adapted version of the Munich Composite International Diagnostic Interview to obtain diagnoses of generalized anxiety disorder, agoraphobia, social phobia, panic disorder and specific phobias in accordance with the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) (
      • Wittchen H.U.
      • Lachner G.
      • Wunderlich U.
      • Pfister H.
      Test-retest reliability of the computerized DSM-IV version of the Munich-Composite International Diagnostic Interview (M-CIDI).
      ).

      Statistical Analysis

      Missing covariate data was imputed with fivefold multiple imputation. FA and MD were standardized for each tract to facilitate comparison of associations. We used the mean FA and MD values of left and right hemispheres combined to limit the number of outcomes and the risk of spurious findings. We used multivariable linear regression models to study cross-sectional associations between social health markers and tract-specific FA and MD. We performed stepwise adjustment of the models to interpret the change of the effect estimates with each addition of a set of covariates. In model 1, we adjusted for age, sex, intracranial volume. We included phase encoding direction of the diffusion scan as a covariate and potential confounder in all models. In model 2, we added smoking status, alcohol consumption, body mass index, hypertension, education level, mini mental state evaluation, CES-D score, presence of anxiety disorders, coronary heart disease, diabetes diagnosis, heart failure diagnosis and clinical stroke. In model 3, we additionally adjusted the model for tract volume and tract-specific white matter hyperintensity volume. Next, we stratified all models on sex to assess effect modification by sex by interpreting the associations separately for male and female participants. Multiplicative interaction was studied by adding an interaction term for the product of each social health marker with sex to each model.
      We performed permutation testing to determine the number of independent tests, considering we included 3 determinants and 15 outcomes which are theoretically correlated. For each outcome variable, linear regressions were run with a random variable and repeated 10,000 times. The minimum p-value for each regression model (permutation) was extracted and these p-values were sorted to define the significance threshold, which was based on the 5% quantile (0.0037). We then divided 0.05 by this threshold to obtain the number of independent tests (n=13). We calculated the new significance threshold using Šidák correction (αn=1–(1-α)ˆ(1/number of independent tests), resulting in a multiple-testing adjusted p-value threshold of 0.0039 (
      • Sidak Z.
      Rectangular Confidence Regions for the Means of Multivariate Normal Distributions.
      ).
      To investigate if the social health markers are independent determinants also by mutually adjusting them, we performed a sensitivity analysis including all social health markers in the same linear regression model. Correlations between social health markers are presented in Figure S1.

      Results

      The characteristics of the study sample (N=3,352) are shown in Table 1. Mean age was 58.4 (SD 11.7) years, and 55% was female. Eighty-one percent of the participants had a partner, 88% reported to feel not lonely, and the median social support score was 10 (IQR: 10-10, variance 0.82).
      Table 1Characteristics of the Study Sample
      Overall (N=3352)
      Age (years), Mean (SD)58.4 (7.1)
      Sex, N female (%)1841 (54.9)
      Loneliness, N lonely (%)396 (11.8)
      Perceived social support, weighted score, Median (IQR)10 (10-10)
      Perceived social support, weighted score, Mean (SD)9.7 (0.9)
      Marital status, N (%)
      Married or has partner2711 (80.9)
      No current partner641 (19.1)
      Education, N (%)
      Primary education273 (8.1)
      Lower/intermediate general education or lower vocational education1211 (36.1)
      Intermediate vocational education or higher general education987 (29.4)
      Higher vocational education or university881 (26.3)
      MMSE score, Median (IQR)29.0 (27.0-29.0)
      Depressive symptoms score, Median CES-D score (IQR)3.0 (1.0 – 7.0)
      Clinically relevant depressive symptoms (CES-D≥16), N (%)281 (8.4)
      Anxiety disorder, N present (%)255 (7.6)
      Smoking status, N (%)
      Never1014 (30.3)
      Former1542 (46.0)
      Current796 (23.7)
      Alcohol use, N (%)
      None334 (10.0)
      Moderate (0-1 units per day)2029 (60.5)
      Heavy (>1 unit per day)989 (29.5)
      Body mass index (kg/mˆ2), Mean (SD)27.5 (4.1)
      Hypertension, N present (%)1765 (52.3)
      History of diabetes mellitus type 2, N present (%)329 (9.8)
      History of coronary heart disease, N present (%)120 (3.6)
      History of heart failure, N present (%)24 (0.7)
      History of clinical stroke, N present (%)48 (1.4)
      CES-D: Center for Epidemiological Studies Depression scale; MMSE: Mini Mental State Examination
      <0.05) in lighter colour, see legend. Non-significant associations are displayed in white.
      Associations between social health markers and tract-specific white matter integrity are presented in Figure 1 and Table 2. After adjusting for all covariates, loneliness was associated with a higher MD of the parahippocampal part of cingulum (standardized mean difference: 0.21, 95%CI: 0.09 to 0.34), and the superior thalamic radiation (standardized mean difference: 0.21, 95%CI: 0.09 to 0.34). Associations between loneliness and tract-specific FA were not statistically significant after multiple testing correction.
      Table 2Associations between social health markers and tract-specific white matter integrity
      Fractional Anisotropy
      Association tractsCommissural tractsLimbic tractsSensorimotor tracts
      Social health markersModelATRIFOILFPTRSLFUNCFMAFMICGCCGHFXCSTMCPMLSTR
      Social support (per point increase)10.010.040.040.050.040.040.040.05*0.010.010.010.02-0.020.020.03
      Social support (per point increase)20.010.030.040.040.040.040.030.050.01000.01-0.020.010.03
      Social support (per point increase)300.030.030.020.030.020.010.04000.010-0.020.020.02
      Partner status (no partner vs. partner)1-0.06-0.14*-0.13*-0.16*-0.08-0.06-0.13*-0.12*-0.06-0.040.05-0.030.02-0.03-0.05
      Partner status (no partner vs. partner)2-0.04-0.12-0.11-0.13*-0.06-0.05-0.1-0.1-0.05-0.030.06-0.010.01-0.02-0.02
      Partner status (no partner vs. partner)30-0.09-0.08-0.1-0.03-0.02-0.03-0.08-0.06-0.040.08-0.010.01-0.01-0.01
      Loneliness (lonely vs. not lonely)1-0.11-0.12-0.1-0.14-0.07-0.09-0.1-0.08-0.1-0.06-0.07-0.080.06-0.03-0.03
      Loneliness (lonely vs. not lonely)2-0.1-0.1-0.07-0.09-0.07-0.13-0.05-0.05-0.1-0.06-0.06-0.040.02-0.000.02
      Loneliness (lonely vs. not lonely)3-0.07-0.07-0.06-0.08-0.03-0.070-0.02-0.08-0.07-0.04-0.0100.010.04
      Mean Diffusivity
      Association tractsCommissural tractsLimbic tractsSensorimotor tracts
      Social health markersModelATRIFOILFPTRSLFUNCFMAFMICGCCGHFXCSTMCPMLSTR
      Social support (per point increase)1-0.04-0.06*-0.05-0.04-0.05-0.06*-0.03-0.07*-0.04-0.02-0.05*-0.040.01-0.05-0.02
      Social support (per point increase)2-0.03-0.06*-0.05-0.04-0.04-0.06*-0.03-0.06*-0.03-0.01-0.04-0.040.01-0.04-0.02
      Social support (per point increase)3-0.02-0.05*-0.04-0.02-0.04-0.05*-0.01-0.06*-0.03-0.01-0.02-0.030.01-0.04-0.01
      Partner status (no partner vs. partner)10.11*0.11*0.080.14*0.110.090.14*0.080.10.13*0.080.13*0.010.050.13*
      Partner status (no partner vs. partner)20.070.090.060.12*0.080.080.13*0.060.090.120.060.110.030.030.12
      Partner status (no partner vs. partner)30.020.070.040.090.060.060.070.040.090.11-0.010.090.030.030.1
      Loneliness (lonely vs. not lonely)10.080.130.120.110.110.070.110.10.090.19*0.110.17*-0.030.030.16*
      Loneliness (lonely vs. not lonely)20.030.090.080.060.090.080.090.040.090.22*0.050.19*0.03-0.010.22*
      Loneliness (lonely vs. not lonely)30.010.070.050.040.070.040.050.020.080.21*-0.010.160.07-0.010.21*
      Table 2: Standardized mean differences of model 1, model 2, and model 3, for fractional anisotropy (FA, top panel) and mean diffusivity (MD, bottom panel). In bold-black the statistical significant (p≤0.05) coefficients without multiple testing correction, asterisks indicate the statistical significant coefficients after multiple testing correction (Šidák correction). Tract abbreviations: ATR – anterior thalamic radiation, IFO – inferior fronto-occipital fasciculus, ILF – inferior longitudinal fasciculus, PTR – posterior thalamic radiation, SLF – superior longitudinal fasciculus, UNC – uncinate fasciculus, FMA – forceps major, FMI – forceps minor, CGC – cingulate gyrus part of cingulum, CGH – parahippocampal part of cingulum, FX – fornix, CST – corticospinal tract, MCP – middle cerebellar peduncle, ML – medial lemniscus, STR – superior thalamic radiation.
      Better perceived social support scores were associated with lower MD of the inferior fronto-occipital fasciculus (standardized mean difference of MD per point increase in social support: -0.05, 95%CI: -0.08 to -0.03), the uncinate fasciculus (standardized mean difference: -0.05, 95%CI: -0.08 to -0.02), and the forceps minor (standardized mean difference: -0.06, 95%CI: -0.09 to -0.03) (Table 2, model 3).
      Partnership status was no longer statistically significantly associated with white matter tract integrity after adjusting for all covariates and multiple testing correction. Partnership status was associated with FA and MD only in model 1, where not having a partner was associated with a lower FA for the inferior fronto-occipital fasciculus, inferior longitudinal fasciculus, posterior thalamic radiation, forceps major, and forceps minor after multiple testing correction. Not having a partner was associated with higher MD in the anterior thalamic radiation, inferior fronto-occipital fasciculus, posterior thalamic radiation, forceps major, parahippocampal part of cingulum, corticospinal tract and superior thalamic radiation only in model 1. Not having a partner remained associated with higher MD of the posterior thalamic radiation (standardized mean difference: 0.12, 95%CI: 0.04 to 0.20) and the forceps major (standardized mean difference: 0.13, 95%CI: 0.04 to 0.21) after adjusting for lifestyle and multimorbidity, but not after adjusting for tract volume and tract-specific WMH volume.
      After stratifying on sex model 3, only male participants showed an association of social support with MD of the forceps minor (standardized mean difference: -0.07, 95%CI: -0.11 to -0.03) (p for interaction = 0.80), and an association of partnership status with the fractional anisotropy of the forceps minor (standardized mean difference: -0.22, 95%CI: -0.36 to -0.07) (p for interaction = 0.02), as shown in Table 3. Female participants showed an association of loneliness with the mean diffusivity of the superior thalamic radiation (standardized mean difference: 0.31, 95%CI: 0.13 to 0.50) (p for interaction = 0.05).
      Table 3Associations between social health markers and tract-specific white matter integrity stratified on sex
      Fractional Anisotropy
      Association tractsCommissural tractsLimbic tractsSensorimotor tracts
      Social health markerSexATRIFOILFPTRSLFUNCFMAFMICGCCGHFXCSTMCPMLSTR
      Social support (per point increase)Male0.000.050.040.020.050.030.000.060.010.02-0.020.00-0.02-0.010.03
      Social support (per point increase)Female0.000.000.010.030.020.010.020.030.010.020.010.02-0.030.050.00
      Partner status (no partner vs. partner)Male0.07-0.12-0.07-0.07-0.04-0.05-0.02-0.22*-0.09-0.040.090.160.040.120.11
      Partner status (no partner vs. partner)Female-0.03-0.10-0.09-0.10-0.03-0.01-0.04-0.03-0.05-0.030.07-0.05-0.03-0.03-0.06
      Loneliness (lonely vs. not lonely)Male-0.08-0.010.03-0.040.010.000.090.020.00-0.06-0.010.010.11-0.110.07
      Loneliness (lonely vs. not lonely)Female-0.07-0.09-0.09-0.10-0.04-0.09-0.06-0.05-0.12-0.07-0.05-0.01-0.040.040.05
      Mean Diffusivity
      Association tractsCommissural tractsLimbic tractsSensorimotor tracts
      Social health markerSexATRIFOILFPTRSLFUNCFMAFMICGCCGHFXCSTMCPMLSTR
      Social Support (per point increase)Male-0.02-0.05-0.030.00-0.05-0.050.01-0.07*-0.04-0.02-0.02-0.03-0.01-0.02-0.02
      Social Support (per point increase)Female-0.03-0.05-0.05-0.03-0.03-0.05-0.02-0.05-0.030.00-0.02-0.020.03-0.070.00
      Partner status (no partner vs. partner)Male0.070.150.020.070.070.120.160.080.120.130.040.070.040.010.03
      Partner status (no partner vs. partner)Female0.020.050.030.090.060.030.030.020.100.11-0.010.110.040.040.14
      Loneliness (lonely vs. not lonely)Male-0.020.04-0.070.000.060.040.020.050.000.14-0.01-0.030.160.01-0.01
      Loneliness (lonely vs. not lonely)Female0.030.090.110.080.060.040.070.000.120.22-0.010.240.02-0.020.31*
      Table 3: Standardized mean differences of model 3 stratified by sex for fractional anisotropy (top panel) and for mean diffusivity (bottom panel). In bold-black the statistical significant (p≤0.05) coefficients without multiple testing correction, asterisks indicate the statistical significant coefficients after multiple testing correction (Šidák correction). Underlines indicate significant p-values for the interaction term of the social health marker with sex. Tract abbreviations: ATR – anterior thalamic radiation, IFO – inferior fronto-occipital fasciculus, ILF – inferior longitudinal fasciculus, PTR – posterior thalamic radiation, SLF – superior longitudinal fasciculus, UNC – uncinate fasciculus, FMA – forceps major, FMI – forceps minor, CGC – cingulate gyrus part of cingulum, CGH – parahippocampal part of cingulum, FX – fornix, CST – corticospinal tract, MCP – middle cerebellar peduncle, ML – medial lemniscus, STR – superior thalamic radiation.
      The sensitivity analyses did not change the interpretation of our findings (Table S1). In the mutually adjusted model, the association between social support and the uncinate fasciculus was no longer statistically significant after multiple testing correction.

      Discussion

      In this study, we aimed to determine the associations between different aspects of social health and tract-specific brain white matter microstructure in community-dwelling older adults. We found that better perceived social support was associated with lower mean diffusivity (reflecting higher structural integrity) of the inferior fronto-occipital fasciculus, the uncinate fasciculus, and the forceps minor, whereas loneliness was associated with higher mean diffusivity (reflecting lower integrity) of the parahippocampal part of cingulum and of the superior thalamic radiation. Additionally, we found potential interaction effects of sex on the associations between partner status and loneliness with specifically the integrity of the forceps minor and superior thalamic radiation. We will briefly discuss each of these findings.
      Better perceived social support was associated with lower mean diffusivity of the association and commissural tracts, namely the inferior fronto-occipital fasciculus, the uncinate fasciculus, and the forceps minor. Two recent extensive literature review papers on structural white matter have described that the inferior fronto-occipital fasciculus and uncinate fasciculus may play key roles in socio-emotional processing and social cognition (
      • Ameis S.H.
      • Catani M.
      Altered white matter connectivity as a neural substrate for social impairment in Autism Spectrum Disorder.
      ,
      • Wang Y.
      • Metoki A.
      • Alm K.H.
      • Olson I.R.
      White matter pathways and social cognition.
      ). The inferior fronto-occipital fasciculus runs from the ventral occipital cortex through the temporal cortex and terminates in the orbitofrontal, medial prefrontal and inferior frontal cortex. As such, it is involved in face-processing and -perception, mentalizing (Theory of Mind) and embodied cognition (mirroring) (
      • Wang Y.
      • Metoki A.
      • Alm K.H.
      • Olson I.R.
      White matter pathways and social cognition.
      ). The uncinate fasciculus is a limbic tract that connects medial temporal areas to the medial and lateral orbitofrontal cortex, and has been linked to emotion recognition and empathy abilities (
      • Wang Y.
      • Metoki A.
      • Alm K.H.
      • Olson I.R.
      White matter pathways and social cognition.
      ). The forceps minor, representing the anterior commissure of the corpus callosum, connects bilateral anterior frontal regions through the genu. Malformation or disruptions of the anterior corpus callosum have been associated with deficits in cognitive functioning and social communication (
      • Buyanova I.S.
      • Arsalidou M.
      Cerebral White Matter Myelination and Relations to Age, Gender, and Cognition: A Selective Review.
      ), and microstructural integrity of the forceps minor has more recently been linked to social network size and diversity (
      • Noonan M.P.
      • Mars R.B.
      • Sallet J.
      • Dunbar R.I.M.
      • Fellows L.K.
      The structural and functional brain networks that support human social networks.
      ,
      • Molesworth T.
      • Sheu L.K.
      • Cohen S.
      • Gianaros P.J.
      • Verstynen T.D.
      Social network diversity and white matter microstructural integrity in humans.
      ). These findings indicate that the perception of social support is associated with white matter structures that are important for the adequate processing of social cues of others and responding adequately to others’ emotions. Potentially, perceiving good social support stimulates healthy brain structures that are needed for the ability to appropriately respond to social cues.
      Loneliness in our study was associated with higher mean diffusivity of limbic and sensorimotor tracts, specifically the parahippocampal part of cingulum and the superior thalamic radiation. Several previous studies have reported on associations between loneliness and white matter structure, including a recent population-based study which found that loneliness was mainly linked to impaired microstructure of the fornix, but not other structures (
      • Spreng R.N.
      • Dimas E.
      • Mwilambwe-Tshilobo L.
      • Dagher A.
      • Koellinger P.
      • Nave G.
      • et al.
      The default network of the human brain is associated with perceived social isolation.
      ). Another study in young adults found that loneliness was associated with reduced white matter density in brain areas that are related to social cognition, empathy, and self- cognition and –efficacy (
      • Nakagawa S.
      • Takeuchi H.
      • Taki Y.
      • Nouchi R.
      • Sekiguchi A.
      • Kotozaki Y.
      • et al.
      White matter structures associated with loneliness in young adults.
      ). Loneliness has not specifically been linked yet to the cingulum and thalamic radiations in previous research. The parahippocampal part of cingulum is part of the limbic system and connects the medial prefrontal cortex and anterior cingulate cortex through the precuneus to medial temporal regions proximal to the hippocampus (
      • Wang Y.
      • Metoki A.
      • Alm K.H.
      • Olson I.R.
      White matter pathways and social cognition.
      ). It is involved in functions related to memory and socio-emotional processing, and has been suggested as part of the brain network related to empathy (
      • Wang Y.
      • Metoki A.
      • Alm K.H.
      • Olson I.R.
      White matter pathways and social cognition.
      ). On the other hand, the superior thalamic radiation has (to the best of our knowledge) not been mentioned before in relation to social processes. The superior thalamic radiation connects the thalamus to the parietal lobe, penetrating the posterior limb of the internal capsule, and has mainly been thought to be involved in processing sensory input. Recently however, findings from the UK Biobank indicated that depressive symptomology and major depressive disorder are associated with altered microstructure of thalamic radiations, including the superior thalamic radiation (
      • Shen X.
      • Reus L.M.
      • Cox S.R.
      • Adams M.J.
      • Liewald D.C.
      • Bastin M.E.
      • et al.
      Subcortical volume and white matter integrity abnormalities in major depressive disorder: findings from UK Biobank imaging data.
      ). Thus, loneliness may be associated with altered processing of sensory information in the brain, with implications for brain structures required for memory, socio-emotional and sensory processing.
      Partnership status was not significantly associated with tract-specific white matter integrity after adjusting for tract-white matter volume and tract-white matter hyperintensity volume, indicating that the association between partnership status and white matter integrity may be better explained by white matter macrostructure than by partnership status and covariates alone. Prior to multiple testing correction, partnership status showed an association with the microstructural integrity of several white matter tracts, including association, limbic, and sensorimotor tracts, where not having a current partner was consistently associated with worse microstructure. Future studies with higher statistical power should investigate these associations further.
      Lastly, our findings suggest sex-specific differences in the associations between social health and white matter microstructure. We found that not having a partner was associated with lower fractional anisotropy of the forceps minor in males only, and that the association between loneliness and higher mean diffusivity of the superior thalamic radiation was only present in female participants. This is in line with sex differences found in previous studies, where loneliness in men is associated with greater structural integrity of the fornix (
      • Spreng R.N.
      • Dimas E.
      • Mwilambwe-Tshilobo L.
      • Dagher A.
      • Koellinger P.
      • Nave G.
      • et al.
      The default network of the human brain is associated with perceived social isolation.
      ), and with smaller global white matter volume (
      • van der Velpen I.F.
      • Melis R.J.F.
      • Perry M.
      • Vernooij-Dassen M.J.F.
      • Ikram M.A.
      • Vernooij M.W.
      Social Health Is Associated With Structural Brain Changes in Older Adults: The Rotterdam Study.
      ). Importantly, these results could though be affected by the fact that men tend to report themselves lonely less often than women due to stigma on loneliness for men (
      • Borys S.
      • Perlman D.
      Gender Differences in Loneliness.
      ). More research is required to elucidate the complex associations between sex, gender, social health and the brain.
      The large sample size and the population-based setting are important strengths of this study. A further strength is that we controlled for a large number of potential confounders, including markers of physical and mental health, lifestyle behaviours, and tract volume.
      Limitations include that we considered only 15, major, white matter tracts, excluding minor tracts from our analysis. For tracts that are located on both right and left hemispheres, we only analysed averaged values for the entire tracts and did not consider the portion of the tracts in the two hemispheres separately. We regarded FA and MD as reflecting microstructural integrity, which is a simplification of the underlying biological processes, but is generally accepted as such a representation. Furthermore, the perceived social support questionnaire has not been formally validated and the direct question on loneliness in the CES-D may have led to underreporting of loneliness in our sample. Finally, due to the cross-sectional analysis, the causal relationship between social health variables and white matter microstructure cannot be inferred by the present study and should be elucidated by future investigations, which should also consider the possibility the negative (early) live experiences such as childhood maltreatment could have influenced both white matter and social health and thus represent a possible additional mechanism underlying the present findings [42,43].

      Conclusions

      We investigated the associations between social health markers and microstructure of specific white matter tracts, finding that a better social health is related with a higher microstructural integrity of specific tracts: the superior thalamic radiation, the parahippocampal part of cingulum, the uncinate fasciculus, the inferior fronto-occipital fasciculus, and the forceps minor. Different social health markers such as loneliness and perceived social support associate with the microstructural integrity of different white matter tracts, suggesting that various social health domains influence different neural pathways of the brain. Our findings are of theoretical and practical relevance, giving new insight in the relationship between social health and white matter tracts, and contributing to the understanding of the mechanisms that lead to neuropsychiatric diseases, including dementia, in which social dysfunction is a common symptom.

      Acknowledgements and disclosures

      The Rotterdam Study is funded by Erasmus Medical Center and Erasmus University, Rotterdam, Netherlands Organization for the Health Research and Development (ZonMw), the Research Institute for Diseases in the Elderly (RIDE), the Ministry of Education, Culture and Science, the Ministry for Health, Welfare and Sports, the European Commission (DG XII), and the Municipality of Rotterdam. This work was partly supported by the JPND project Social Health And Reserve in the Dementia patient journey (SHARED) and financed through projects funded by the Netherlands Organisation for Health Research and Development (grant numbers 733051082 and 733050831).
      Wiro Niessen is founder, scientific lead and shareholder of Quantib BV. All remaining authors report no biomedical financial interests or potential conflicts of interest.
      The authors are grateful to the study participants, the staff from the Rotterdam Study and the participating general practitioners and pharmacists. The authors would also like to thank Henri Vrooman and Marius de Groot for help with the figures.

      References

        • Huber M.
        • Knottnerus J.A.
        • Green L.
        • Horst Hvd
        • Jadad A.R.
        • Kromhout D.
        • et al.
        How should we define health?.
        Bmj. 2011; 343: d4163
        • Vernooij-Dassen M.
        • Jeon Y.H.
        Social health and dementia: the power of human capabilities.
        Int Psychogeriatr. 2016; 28: 701-703
        • Droes R.M.
        • Chattat R.
        • Diaz A.
        • Gove D.
        • Graff M.
        • Murphy K.
        • et al.
        Social health and dementia: a European consensus on the operationalization of the concept and directions for research and practice.
        Aging Ment Health. 2017; 21: 4-17
        • Vernooij-Dassen M.
        • Moniz-Cook E.
        • Verhey F.
        • Chattat R.
        • Woods B.
        • Meiland F.
        • et al.
        Bridging the divide between biomedical and psychosocial approaches in dementia research: the 2019 INTERDEM manifesto.
        Aging Ment Health. 2019; : 1-7
        • Insel T.R.
        • Fernald R.D.
        How the brain processes social information: searching for the social brain.
        Annu Rev Neurosci. 2004; 27: 697-722
        • Porcelli S.
        • Van Der Wee N.
        • van der Werff S.
        • Aghajani M.
        • Glennon J.C.
        • van Heukelum S.
        • et al.
        Social brain, social dysfunction and social withdrawal.
        Neurosci Biobehav Rev. 2019; 97: 10-33
        • Holt-Lunstad J.
        The Major Health Implications of Social Connection.
        Current Directions in Psychological Science. 2021; 30: 251-259
        • Holt-Lunstad J.
        • Smith T.B.
        • Baker M.
        • Harris T.
        • Stephenson D.
        Loneliness and social isolation as risk factors for mortality: a meta-analytic review.
        Perspect Psychol Sci. 2015; 10: 227-237
        • Holt-Lunstad J.
        • Smith T.B.
        • Layton J.B.
        Social relationships and mortality risk: a meta-analytic review.
        PLoS Med. 2010; 7e1000316
        • Howick J.
        • Kelly P.
        • Kelly M.
        Establishing a causal link between social relationships and health using the Bradford Hill Guidelines.
        SSM Popul Health. 2019; 8100402
        • Malcolm M.
        • Frost H.
        • Cowie J.
        Loneliness and social isolation causal association with health-related lifestyle risk in older adults: a systematic review and meta-analysis protocol.
        Syst Rev. 2019; 8: 48
        • Holt-Lunstad J.
        Why Social Relationships Are Important for Physical Health: A Systems Approach to Understanding and Modifying Risk and Protection.
        Annu Rev Psychol. 2018; 69: 437-458
        • Holt-Lunstad J.
        The Potential Public Health Relevance of Social Isolation and Loneliness: Prevalence, Epidemiology, and Risk Factors.
        Public Policy & Aging Report. 2018; 27: 127-130
        • Ameis S.H.
        • Catani M.
        Altered white matter connectivity as a neural substrate for social impairment in Autism Spectrum Disorder.
        Cortex. 2015; 62: 158-181
        • Wang Y.
        • Metoki A.
        • Alm K.H.
        • Olson I.R.
        White matter pathways and social cognition.
        Neuroscience & Biobehavioral Reviews. 2018; 90: 350-370
        • Wang Y.
        • Olson I.R.
        The Original Social Network: White Matter and Social Cognition.
        Trends in Cognitive Sciences. 2018; 22: 504-516
        • van der Velpen I.F.
        • Melis R.J.F.
        • Perry M.
        • Vernooij-Dassen M.J.F.
        • Ikram M.A.
        • Vernooij M.W.
        Social Health Is Associated With Structural Brain Changes in Older Adults: The Rotterdam Study.
        Biol Psychiatry Cogn Neurosci Neuroimaging. 2021;
        • Pettersson-Yeo W.
        • Allen P.
        • Benetti S.
        • McGuire P.
        • Mechelli A.
        Dysconnectivity in schizophrenia: where are we now?.
        Neurosci Biobehav Rev. 2011; 35: 1110-1124
        • Jalbrzikowski M.
        • Villalon-Reina J.E.
        • Karlsgodt K.H.
        • Senturk D.
        • Chow C.
        • Thompson P.M.
        • et al.
        Altered white matter microstructure is associated with social cognition and psychotic symptoms in 22q11.2 microdeletion syndrome.
        Front Behav Neurosci. 2014; 8: 393
        • Jung S.
        • Kim J.H.
        • Sung G.
        • Ko Y.G.
        • Bang M.
        • Park C.I.
        • et al.
        Uncinate fasciculus white matter connectivity related to impaired social perception and cross-sectional and longitudinal symptoms in patients with schizophrenia spectrum psychosis.
        Neurosci Lett. 2020; 737135144
        • Li Y.
        • Zhou Z.
        • Chang C.
        • Qian L.
        • Li C.
        • Xiao T.
        • et al.
        Anomalies in uncinate fasciculus development and social defects in preschoolers with autism spectrum disorder.
        BMC Psychiatry. 2019; 19: 399
        • Roelofs E.F.
        • Bas-Hoogendam J.M.
        • van Ewijk H.
        • Ganjgahi H.
        • van der Werff S.J.A.
        • Barendse M.E.A.
        • et al.
        Investigating microstructure of white matter tracts as candidate endophenotypes of Social Anxiety Disorder - Findings from the Leiden Family Lab study on Social Anxiety Disorder (LFLSAD).
        Neuroimage Clin. 2020; 28102493
        • Filley C.M.
        Social Cognition and White Matter: Connectivity and Cooperation.
        Cogn Behav Neurol. 2020; 33: 67-75
        • Saito J.
        • Hori M.
        • Nemoto T.
        • Katagiri N.
        • Shimoji K.
        • Ito S.
        • et al.
        Longitudinal study examining abnormal white matter integrity using a tract-specific analysis in individuals with a high risk for psychosis.
        Psychiatry Clin Neurosci. 2017; 71: 530-541
        • Noonan M.P.
        • Mars R.B.
        • Sallet J.
        • Dunbar R.I.M.
        • Fellows L.K.
        The structural and functional brain networks that support human social networks.
        Behav Brain Res. 2018; 355: 12-23
        • Molesworth T.
        • Sheu L.K.
        • Cohen S.
        • Gianaros P.J.
        • Verstynen T.D.
        Social network diversity and white matter microstructural integrity in humans.
        Soc Cogn Affect Neurosci. 2015; 10: 1169-1176
        • Spreng R.N.
        • Dimas E.
        • Mwilambwe-Tshilobo L.
        • Dagher A.
        • Koellinger P.
        • Nave G.
        • et al.
        The default network of the human brain is associated with perceived social isolation.
        Nature Communications. 2020; 11: 6393
        • Ikram M.A.
        • Brusselle G.
        • Ghanbari M.
        • Goedegebure A.
        • Ikram M.K.
        • Kavousi M.
        • et al.
        Objectives, design and main findings until 2020 from the Rotterdam Study.
        European Journal of Epidemiology. 2020; 35: 483-517
        • Radloff L.S.
        The CES-D Scale:A Self-Report Depression Scale for Research in the General Population.
        Applied Psychological Measurement. 1977; 1: 385-401
        • Ikram M.A.
        • van der Lugt A.
        • Niessen W.J.
        • Koudstaal P.J.
        • Krestin G.P.
        • Hofman A.
        • et al.
        The Rotterdam Scan Study: design update 2016 and main findings.
        Eur J Epidemiol. 2015; 30: 1299-1315
        • Vrooman H.A.
        • Cocosco C.A.
        • van der Lijn F.
        • Stokking R.
        • Ikram M.A.
        • Vernooij M.W.
        • et al.
        Multi-spectral brain tissue segmentation using automatically trained k-Nearest-Neighbor classification.
        NeuroImage. 2007; 37: 71-81
        • de Boer R.
        • Vrooman H.A.
        • van der Lijn F.
        • Vernooij M.W.
        • Ikram M.A.
        • van der Lugt A.
        • et al.
        White matter lesion extension to automatic brain tissue segmentation on MRI.
        NeuroImage. 2009; 45: 1151-1161
        • Koppelmans V.
        • de Groot M.
        • de Ruiter M.B.
        • Boogerd W.
        • Seynaeve C.
        • Vernooij M.W.
        • et al.
        Global and focal white matter integrity in breast cancer survivors 20 years after adjuvant chemotherapy.
        Human Brain Mapping. 2014; 35: 889-899
        • Leemans A.
        • Jeurissen B.
        • Sijbers J.
        • Jones D.K.
        ExploreDTI: a graphical toolbox for processing, analyzing, and visualizing diffusion MR data.
        Proc Intl Soc Mag Reson Med. 2009; : 3537
        • de Groot M.
        • Ikram M.A.
        • Akoudad S.
        • Krestin G.P.
        • Hofman A.
        • van der Lugt A.
        • et al.
        Tract-specific white matter degeneration in aging: The Rotterdam Study.
        Alzheimer's & Dementia. 2015; 11: 321-330
        • Behrens T.E.
        • Berg H.J.
        • Jbabdi S.
        • Rushworth M.F.
        • Woolrich M.W.
        Probabilistic diffusion tractography with multiple fibre orientations: What can we gain?.
        Neuroimage. 2007; 34: 144-155
        • Jenkinson M.
        • Beckmann C.F.
        • Behrens T.E.
        • Woolrich M.W.
        • Smith S.M.
        Fsl. Neuroimage. 2012; 62: 782-790
        • Thomason M.E.
        • Thompson P.M.
        Diffusion imaging, white matter, and psychopathology.
        Annu Rev Clin Psychol. 2011; 7: 63-85
        • Beekman A.T.
        • Deeg D.J.
        • Van Limbeek J.
        • Braam A.W.
        • De Vries M.Z.
        • Van Tilburg W.
        Criterion validity of the Center for Epidemiologic Studies Depression scale (CES-D): results from a community-based sample of older subjects in The Netherlands.
        Psychol Med. 1997; 27: 231-235
        • Wittchen H.U.
        • Lachner G.
        • Wunderlich U.
        • Pfister H.
        Test-retest reliability of the computerized DSM-IV version of the Munich-Composite International Diagnostic Interview (M-CIDI).
        Soc Psychiatry Psychiatr Epidemiol. 1998; 33: 568-578
        • Sidak Z.
        Rectangular Confidence Regions for the Means of Multivariate Normal Distributions.
        Journal of the American Statistical Association. 1967; 62: 626-633
        • Buyanova I.S.
        • Arsalidou M.
        Cerebral White Matter Myelination and Relations to Age, Gender, and Cognition: A Selective Review.
        Front Hum Neurosci. 2021; 15662031
        • Nakagawa S.
        • Takeuchi H.
        • Taki Y.
        • Nouchi R.
        • Sekiguchi A.
        • Kotozaki Y.
        • et al.
        White matter structures associated with loneliness in young adults.
        Scientific Reports. 2015; 517001
        • Shen X.
        • Reus L.M.
        • Cox S.R.
        • Adams M.J.
        • Liewald D.C.
        • Bastin M.E.
        • et al.
        Subcortical volume and white matter integrity abnormalities in major depressive disorder: findings from UK Biobank imaging data.
        Scientific Reports. 2017; 7: 5547
        • Borys S.
        • Perlman D.
        Gender Differences in Loneliness.
        Personality and Social Psychology Bulletin. 1985; 11: 63-74