GENETIC INFLUENCES ON SOCIAL COGNITION, EXECUTIVE FUNCTION, AND ASSOCIATED NEURAL NETWORKS By Reid Blanchett A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Genetics—Doctor of Philosophy 2022 ABSTRACT GENETIC INFLUENCES ON SOCIAL COGNITION, EXECUTIVE FUNCTION, AND ASSOCIATED NEURAL NETWORKS By Reid Blanchett Fundamental cognitive domains include executive function and social cognition. Both social cognition and executive functioning can be studied using neuroimaging techniques that allow direct observations to be made about brain structure and function. These techniques can also be applied to the study of brain development, revealing how circuits involved in executive function and social cognition change during important developmental periods such as infancy. Along with providing a window into brain maturation, neuroimaging can be used to study cases where cognitive domains are disrupted and make comparisons to learn about typical brain development and function. For my dissertation, I have explored these cognitive domains and associated neural circuits both in typically developing individuals and in individuals with Turner syndrome, a condition caused by the full or partial absence of the second sex chromosome. First, I used a classic twin design and demonstrated relatively low narrow-sense heritability estimates for neonatal resting-state functional connectivity phenotypes. I studied both between- and within- network connectivity in neonates and demonstrated that only 6 out of 36 phenotypes had heritability estimates greater than 0.10; no estimates were statistically significant. These within- and between-network phenotypes included networks heavily recruited for social cognition and executive functioning. I also showed statistically significant associations between neonate resting-state functional connectivity phenotypes and specific demographic and medical history variables. Second, I compared structural and functional connectivity between typically developing male and female infants and infants with Turner syndrome. I saw no differences between the three groups in integrity of the superior longitudinal fasciculus or reduced connectivity between the right precentral gyrus and brain regions in the occipital and parietal regions involved with social cognition, visuospatial reasoning, and executive function. Fronto-parietal connectivity and integrity of the superior longitudinal fasciculus are disrupted in older individuals with Turner syndrome and these results suggested that these changes emerge after the first year of life. I conducted a further exploratory analysis of 54 fiber tracts and showed significant group differences that primarily reflected masculinization of white matter microstructure in TS. Other differences may have arisen due to hemizygosity of the pseudoautosomal region. Finally, I developed a browser-based online testing platform targeting domains such as executive functioning and social cognition, which are often disrupted in Turner syndrome. I then validated the battery via administration to neurotypical males and females and to adult women with Turner syndrome, who performed more poorly on tests of executive function and visuospatial reasoning. Taken together, the results presented in this dissertation contribute greatly to our understanding of the role of genetics in social cognition, executive function, and their related neural networks. These results can be further utilized in longitudinal studies of brain development and in future cognitive testing research. ACKNOWLEDGEMENTS I would first like to thank my mentor, Dr. Rebecca Knickmeyer, for her unwavering support, constant encouragement, and dedication to helping me achieve my goals and dreams. I could never have even conceived of doing something this big without her on my team. I would like to acknowledge the members of the Knickmeyer lab as well who were there with me every step of the way, especially Dr. Ann Alex with whom I will always share a special bond. Additionally, I want to thank the Genetics and Genome Sciences Program for being so flexible and for its commitment to its students. My committee has also been a source of great support, and I would like to thank Drs. Alex Burt, Arjun Krishnan, Brian Schutte, and David Zhu for their time and investment in my future. Finally, I would like to thank the family that I have made along the way during my time in higher education. Jasmine Charter-Harris, Billie-Heard, my friends at MSU, and most especially the Santelli family who were there for every triumph and every tear on my adventure. iv TABLE OF CONTENTS LIST OF TABLES .......................................................................................................... viii LIST OF FIGURES ..........................................................................................................xi KEY TO ABBREVIATIONS ............................................................................................xv Chapter 1 ........................................................................................................................ 1 Introduction .................................................................................................................. 1 Neuroimaging Modalities Used in this Dissertation ................................................... 3 Resting state functional MRI .................................................................................. 3 Diffusion Tensor Imaging....................................................................................... 6 Executive Functions (EF) .......................................................................................... 9 Definition of EF and Component Processes .......................................................... 9 Relevance of EF to Psychiatry ............................................................................ 10 Neurobiology of EF .............................................................................................. 11 Development of EF and Associated Neural Circuits ............................................ 12 Genetic architecture of EF and associated neural circuits ................................... 14 Sex Differences in EF .......................................................................................... 15 Social Cognition (SC) ............................................................................................. 16 Definition of SC and Component Processes ....................................................... 16 Relevance of SC to Psychiatry ............................................................................ 17 Neurobiology of SC ............................................................................................. 18 Development of SC ............................................................................................. 20 Genetic architecture of SC .................................................................................. 22 Sex differences in SC .......................................................................................... 22 Turner syndrome as a model for understanding sex chromosome effects on EF and SC........................................................................................................................... 23 The importance of studying EF and SC circuits during infancy ............................... 26 Rationale for the dissertation .................................................................................. 28 Chapter 2 ...................................................................................................................... 31 Genetic and Environmental Factors Influencing Neonatal Resting-State Functional Connectivity ............................................................................................................... 31 Abstract .................................................................................................................. 32 Introduction ............................................................................................................. 33 Materials and Methods ........................................................................................... 37 Participants.......................................................................................................... 37 MRI Acquisition and Processing .......................................................................... 38 Statistical Analyses ............................................................................................. 41 Results .................................................................................................................... 43 Participants for Objective 1.................................................................................. 43 Intraclass correlations.......................................................................................... 44 Additive mixed-effects modeling .......................................................................... 47 v Participants for Objective 2.................................................................................. 48 Backwards elimination regression ....................................................................... 50 Mixed Linear Modeling ........................................................................................ 50 Power analysis .................................................................................................... 52 Discussion .............................................................................................................. 53 Chapter 3 ...................................................................................................................... 62 Anatomical and functional connectivity differences in the brains of infants with TS when compared to TD children .................................................................................. 62 Abstract .................................................................................................................. 63 Introduction ............................................................................................................. 64 Methods .................................................................................................................. 67 Participants.......................................................................................................... 67 rs-fMRI and DTI Acquisition and Processing ....................................................... 68 Image analysis (rs-fMRI) ..................................................................................... 69 Statistical Analyses (rs-fMRI) .............................................................................. 70 Image Analysis (DTI) ........................................................................................... 71 Statistical Analyses (DTI) .................................................................................... 72 Exploratory Analyses (DTI) .................................................................................. 73 Results .................................................................................................................... 73 Participants.......................................................................................................... 73 Connectivity Measures ........................................................................................ 74 DTI Measures ...................................................................................................... 77 Exploratory Analyses ........................................................................................... 78 Discussion .............................................................................................................. 84 Chapter 4 ...................................................................................................................... 96 An online platform for testing the disrupted cognitive domains of women with Turner syndrome ................................................................................................................... 96 Abstract .................................................................................................................. 97 Introduction ............................................................................................................. 98 Methods ................................................................................................................ 102 Participants........................................................................................................ 102 Cognitive Battery ............................................................................................... 104 Mental rotation task ........................................................................................ 104 Flanker task.................................................................................................... 105 Reading the Mind in the Eyes ........................................................................ 105 Continuous performance task ........................................................................ 105 Corsi block task .............................................................................................. 106 Digit span task................................................................................................ 106 Simple response time task ............................................................................. 107 Autism spectrum quotient ............................................................................... 107 ADHD report scale ......................................................................................... 107 Statistical Analysis ............................................................................................. 107 Results .................................................................................................................. 108 Mental rotation task ........................................................................................... 108 Flanker task ....................................................................................................... 111 vi Reading the mind in the eyes ............................................................................ 115 Continuous performance task............................................................................ 116 Corsi block task ................................................................................................. 116 Digit span task ................................................................................................... 118 Simple response time task ................................................................................ 120 Autism spectrum quotient .................................................................................. 121 ADHD report scale ............................................................................................ 122 Discussion ............................................................................................................ 122 Chapter 5 .................................................................................................................... 128 Conclusions and future directions ............................................................................ 128 Summary .............................................................................................................. 129 Rigor and reproducibility ....................................................................................... 133 Future work ........................................................................................................... 133 Longitudinal imaging of infants .......................................................................... 133 Utilization of sample-derived networks in the imaging of neonates ................... 134 The temporal relationships between volume, tract integrity, and resting state connectivity in TS .............................................................................................. 135 Implementation of the cognitive battery on a sample of TS women with sequenced exomes ........................................................................................... 136 Selection of appropriate data analysis approach for Aim 1 ................................... 137 Conclusion ............................................................................................................ 138 APPENDICES ............................................................................................................. 139 APPENDIX A: Supplemental data for Chapter 2 ...................................................... 140 APPENDIX B: Supplemental data for Chapter 3 ...................................................... 165 APPENDIX C: Supplemental data for Chapter 4 ...................................................... 190 REFERENCES ............................................................................................................ 203 vii LIST OF TABLES Table 1.1 The eight adult canonical resting-state networks and their subcomponents............................................................................................................. 5 Table 1.2 Major white matter tracts in the brain organized by fiber type. ................ 8 Table 2.1 Assigned networks numbers, network name, and network abbreviation for the eight assigned resting-state connectivity networks. ................................... 41 Table 2.2 Demographic and medical history variables for objective 1 participants. ...................................................................................................................................... 45 Table 2.3 Demographics and medical history of participants for objective 2 ........ 49 Table 3.1 Demographics and medical history for one-year-old infants with Turner syndrome and their typically developing male and female counterparts. ............. 75 Table 3.2 Pairwise comparison of the resting-state functional connectivity between the right precentral gyrus and the five listed regions............................... 76 Table 3.3 Pairwise comparison of the resting-state functional connectivity between the right precentral gyrus and the right calcarine and lingual cortices. . 76 Table 3.4 Paired comparisons between the three groups for structural connectivity of the left superior longitudinal fasciculus ......................................... 78 Table 3.5 Fasciculi with statistically significant FDR-corrected global p-values and post hoc FDR-correct p-values for individual diffusivity metrics ........................... 80 Table 3.6 Chromosomal hierarchies for each statistically significant tract shown for axial diffusivity, radial diffusivity, and fractional anisotropy............................. 84 Table 4.1 Demographic and medical history variables for Turner Syndrome participants. ............................................................................................................... 103 Table 4.2 Demographic variables for control participants .................................... 104 Table 4.3 ANOVA results for general rection time between the three groups on the Mental Rotation Task. ............................................................................................... 109 Table 4.4 Post hoc comparison of general reaction times over all rotational groups after outlier removal and log transformation on the Mental Rotation Task. .................................................................................................................................... 109 viii Table 4.5 ANOVA results for general accuracy between the three groups on the Mental Rotation Task ................................................................................................ 110 Table 4.6 Post hoc comparison of general accuracy over all rotational groups before outlier removal............................................................................................... 111 Table 4.7 ANOVA results for log transformed reaction time on congruent stimuli for the Flanker task ................................................................................................... 113 Table 4.8 ANOVA results for log transformed reaction time on incongruent stimuli for the Flanker task ................................................................................................... 114 Table 4.9 Post hoc comparisons for log transformed reaction time, congruent and incongruent trials, on the Flanker task both before and after outlier removal. ... 114 Table 4.10 ANOVA results between the three groups for Reading the Mind in the Eyes. ........................................................................................................................... 115 Table 4.11 ANOVA results between the three groups for the Corsi Block Task for length of blocks repeated. ........................................................................................ 118 Table 4.12 Post hoc comparisons for maximum blocks repeated on the Corsi Block Task before and after outlier removal. .......................................................... 118 Table 4.13 ANOVA results between the three groups for the Digit Span task for length of numbers repeated. .................................................................................... 119 Table 4.14 Post hoc comparisons for maximum number of repeated numbers on the Digit Span Task after outlier removal. ............................................................... 119 Table 4.15 ANOVA results between the three groups for the Simple Response Time Task for reaction time ...................................................................................... 120 Table 4.16 Post hoc comparisons for reaction time on the Simple Response Time Task before and after outlier removal. .................................................................... 121 Table A.1 The 90 regions from the neonate specific AAL atlas assigned to the eight intrinsic functional networks……………………………………………..............141 Table A.2 Differences in demographics and medical history variables between the two cohorts scanned on either the Allegra or Trio MRI for Objective 1 …………145 Table A.3 Differences in demographics and medical history variables between the two cohorts scanned on either the Allegra or Trio MRI for Objective 2……………………………………………………………………………………………..…149 Table A.4 Narrow-sense heritability estimates for between-network connectivity phenotypes…………………………………………………………………………………..152 ix Table A.5. Narrow-sense heritability estimates for within-network connectivity phenotypes…………………………………………………………………………………..153 Table A.6 Mixed linear modeling results after backwards elimination for between- network phenotypes………………………………………………………….…………….154 Table A.7 Mixed linear modeling results after backwards elimination for between- network phenotypes………………………………………………………………………..163 Table B.1 The 46 tracts that passed quality control with global FDR p-values given, along with local p-values in the estimation of axial diffusivity (AD), radial diffusivity (RD), and fractional anisotropy (FA)……………………………………….166 x LIST OF FIGURES Figure 2.1 Comparison of the functional connectivity values from three-minute and four-minute data from a subsample of 58 subjects. ......................................... 39 Figure 2.2 Twin-twin correlations of connectivity .................................................... 46 Figure 2.3 Narrow-sense heritability values for all networks and network pairs... 47 Figure 2.4 Between- and within-network narrow-sense heritability estimates for resting-state phenotypes............................................................................................ 48 Figure 2.5 Scatterplots of each continuous variable with statistical significance in the mixed modeling..................................................................................................... 51 Figure 2.6 Scatterplot of paternal education against the Z-score of the connectivity values in the dorsal attention network ................................................ 52 Figure 2.7 Power analysis using simulated data. The vertical line indicates the number of twin pairs contained in the current study. .............................................. 53 Figure 3.1 Models of DTI results for the right premotor portion of the corticothalamic tract (top), the left inferior longitudinal fasciculus (middle), and left inferior fronto-occipital fasciculus (bottom)....................................................... 83 Figure 4.1 Overall differences in reaction times between groups. ....................... 108 Figure 4.2 Overall differences in accuracy between groups on the Mental Rotation Task. ........................................................................................................................... 110 Figure 4.3 Differences in log transformed response times on congruent stimuli on the Flanker task. ........................................................................................................ 111 Figure 4.4 Differences in response times on incongruent stimuli on the Flanker task. ............................................................................................................................ 114 Figure 4.5 Differences in the score on the Reading the Mind in the Eyes task between the three groups. The ANOVA was not significant. .................................... 115 Figure 4.6 Differences in the length of blocks repeated on the Corsi Block Test. .................................................................................................................................... 117 Figure 4.7 Differences in the length of numbers repeated on the Digit Span Task.. .................................................................................................................................... 119 xi Figure 4.8 Reaction time between the three groups on the Simple Response Time Task.. .......................................................................................................................... 120 Figure 4.9 Score on the Autism Spectrum Quotient between the three groups. The ANOVA was not significant. ..................................................................................... 121 Figure 4.10 Score on the ADHD Report Scale between the three group. ............. 122 Figure B.1 Model of DTI results for the tapetum portion of the corpus callosum for measures of axial diffusivity………………………………………………………….178 Figure B.2 Model of DTI results for the left corticofugal tract for measures of fractional anisotropy……………………………………………………………………….179 Figure B.3 Model of DTI results for the left motor corticofugal tract for measures of axial diffusivity…………………………………………………………….....................180 Figure B.4 Model of DTI results for the left motor corticofugal tract for measures of radial diffusivity………………………………………………………………………….181 Figure B.5 Model of DTI results for the right optic tract for measures of fractional anisotropy……………………………………………………………………………………182 Figure B.6 Model of DTI results for right frontotemporal region of the arcuate fasciculus for measures of axial diffusivity……………………………….…………...183 Figure B.7 Model of DTI results for the cingulum adjoining the hippocampus for measures of axial diffusivity ……………………………………………………………..184 Figure B.8 Model of DTI results for the cingulum adjoining the hippocampus for measures of radial diffusivity……………………………………..................................185 Figure B.9 Model of DTI results for the motor bundle of the corpus callosum for measures of axial diffusivity……………………………………………………………...186 Figure B.10 Model of DTI results for the left inferior fronto-occipital fasciculus for measures of fractional anisotropy………………………………………………………187 Figure B.11 Model of DTI results for the inferior longitudinal fasciculus for measures of fractional anisotropy………………………………………………………188 Figure B.12 Model of DTI results for the tapetum region of the corpus callosum for measures of fractional anisotropy…………………………………………………..189 Figure C.1 Density plot of reaction time in milliseconds for the Mental Rotation Task between the three groups………………………………………………………………….191 Figure C.2 Density plot of reaction time in milliseconds after log transformation for the Mental Rotation Task between the three groups…………………………………191 xii Figure C.3 Density plot of accuracy for the Mental Rotation Task between the three groups………………………………………………………………………………………..192 Figure C.4 Density plot of reaction time (ms) for the Flanker Task for congruent responses between the three groups…………………………………………………..192 Figure C.5 Density plot of reaction time (ms) after log transformation for congruent responses for the Flanker Task between the three groups…………………………193 Figure C.6 Density plot of reaction time (ms) for the Flanker Task for incongruent responses between the three groups………………………………………………….193 Figure C.7 Density plot of reaction time (ms) after log transformation for incongruent responses for the Flanker Task between the three groups………….194 Figure C.8 Density plot of accuracy for Reading the Mind in the Eyes between the three groups………………………………………………………………………………….194 Figure C.9 Density plot of accuracy for the Continuous Performance Task between the three groups before outlier removal………………………………………………..195 Figure C.10 Density plot of accuracy for the Continuous Performance Task between the three groups after outlier removal…………………………………………………..195 Figure C.11 Density plot of reaction time (ms) for the Continuous Performance Task between the three groups before outlier removal....................................................196 Figure C.12 Density plot of reaction time (ms) for the Continuous Performance Task between the three groups after outlier removal……………………………………….196 Figure C.13 Density plot of reaction time (ms) for the Continuous Performance Task for errors of commission between the three groups before outlier removal……………………………………………………………………………………….197 Figure C.14 Density plot of reaction time (ms) for the Continuous Performance Task for errors of commission between the three groups after outlier removal……………………………………………………………………………………….197 Figure C.15 Density plot of reaction time (ms) for the Continuous Performance Task for errors of omission between the three groups before outlier removal…………………………………………………………………….…………………198 Figure C.16 Density plot of reaction time (ms) for the Continuous Performance Task for errors of omission between the three groups after outlier removal……………………………………………………………………………………….198 Figure C.17 Density plot of the maximum blocks repeated for the Corsi Block Task between the three groups before outlier removal……………………………….……199 xiii Figure C.18 Density plot of the maximum blocks repeated for the Corsi Block Task between the three groups after outlier removal……………………………………….199 Figure C.19 Density plot of the maximum number length repeated for the Corsi Block Task between the three groups…………………………………………………..200 Figure C. 20 Density plot of the reaction time in milliseconds on the Simple Response Time Task between the three groups before log transformation……………………………………………………………………………....200 Figure C.21 Density plot of the reaction time in milliseconds on the Simple Response Time Task between the three groups after log transformation……………………………………………………………………………....201 Figure C.22 Density plot of the accuracy on the Autism Spectrum Quotient between the three groups…………………………………………………………………………….201 Figure C.23 Density plot of the score on the ADHD report scale between the three groups before outlier removal…………………………………………………………….202 Figure C.24 Density plot of the score on the ADHD report scale between the three groups after outlier removal………………………………………………………………202 xiv KEY TO ABBREVIATIONS SC: Social cognition EF: Executive function rs-fMRI: Resting-state functional magnetic resonance imaging DTI: Diffusion tensor imaging BOLD: Blood oxygen level dependent FP: Frontoparietal VA: Ventral Attention DA: Dorsal Attention Lim: Limbic SS: Somatosensory SC: Subcortical Vis: Visual DM: Default mode ADHD: Attention deficit hyperactivity disorder ASD: Autism spectrum disorder PFC: Prefrontal cortex FPN: Frontoparietal network SLF: Superior longitudinal fasciculus FA: Fractional anisotropy RD: Radial diffusivity MD: Mean diffusivity AD: Axial Diffusivity RDoC: Research domain criteria ToM: Theory of mind DMN: Default mode network xv ILF: Inferior longitudinal fasciculus TS: Turner syndrome GLM: General linear model MZ: Monozygotic DZ: Dizygotic EBDS: Early brain development study VR: Visual reasoning MRI: Magnetic resonance imaging TD: Typically developing UNC: University of North Carolina TR: Repetition time TE: Echo time QC: Quality control FADTTS: Functional analysis of diffustion tensor tract statistics PAR: Pseudoautosomal region IFOF: Inferior fronto-occipital fasciculus CHD: Congenital heart disease CPT: Continuous performance task SRT: Simple response time ASQ: Autism spectrum quotient CTSA: Clinical translational science award FDR: False discover rate xvi Chapter 1 Introduction 1 The objective of the presented dissertation is to better understand genetic influences on social cognition (SC), executive function (EF), and associated neural circuits. This objective was achieved via 3 specific aims: (1) identify genetic and environmental factors influencing neonatal resting-state functional connectivity including connectivity within and between the frontoparietal and default mode networks, which play critical roles in EF and SC respectively. (2) identify anatomical and functional connectivity differences in the brains of infants with Turner syndrome when compared to typically developing children, and (3) creating an online cognitive battery targeting cognitive domains that are often disrupted in Turner syndrome including SC and EF. The first two aims were carried out via neuroimaging techniques. Thus, the introduction to the dissertation begins by describing the two neuroimaging modalities utilized in this work: resting-state functional magnetic resonance imaging (rs-fMRI) and diffusion tensor imaging (DTI). Sections 2 and 3 will describe EF and SC and their relevance to psychiatry along with what is currently known about their underlying neurobiology, roles in development, their genetic architecture, and potential sex differences. Aims 2 and 3 focus on individuals with Turner syndrome, a developmental disorder that is caused by the complete or partial absence of the second sex chromosome and is often accompanied by deficits in both SC and EF. Thus, Section 4 of the introduction explains why Turner syndrome is a powerful model for understanding genetic influences on EF and SC. Finally, different developmental periods are presented with the challenges and advantages associated with each in terms of studying cognition and its coupling with neuroimaging. 2 Neuroimaging Modalities Used in this Dissertation Resting state functional MRI In Aims 1 and 2 of this dissertation I used resting state functional MRI (rs-fMRI) to better understand genetic influences on brain networks involved in SC and EF. The basis of fMRI is the observation of blood oxygen level dependent response (BOLD). The principle behind BOLD is that when a neuron is activated, there is increased blood flow with increased oxygen to that area of the brain as the glucose needed by the neuron is aerobically metabolized1 through cellular respiration. fMRI2 signals can be collected in either a task-based fMRI, or a resting-state fMRI but both are based in the principle of BOLD responses. Task-based fMRI has primarily been used to identify brain regions functionally involved in different cognitive processes, while rsfMRI explores the intrinsic functional segregation of brain networks by looking at the temporally correlated hemodynamic responses of the brain at rest3. rsfMRI is especially suitable for the study of infants as it requires no task to be performed by the participant. A sleeping infant is perfectly appropriate for rsfMRI and yields results that add to the growing body of literature concerning network development. rsfMRI data can be analyzed in multiple ways. In Aim 1 of this dissertation, I focus on eight intrinsic functional networks that were originally identified in adults. They include the frontoparietal (FP), ventral attention (VA), dorsal attention (DA), limbic (Lim), somatosensory (SS), subcortical (SC), visual (Vis), and default mode (DM) networks. To calculate functional connectivity within and between these networks, a neonate specific AAL atlas is used to define 90 regions covering the neonatal cerebral cortex. The average 3 BOLD time course is extracted from each region for each subject to construct a 90 x 90 correlation matrix. The correlation matrix is Fisher-Z transformed to be the functional connectivity matrix for each subject. The 90 regions are assigned to the eight networks as shown in Table 1.1. Subsequently, within- and between-network connectivity is calculated by averaging the FC values of the within-/between- networks for each subject. This method was most appropriate for testing my central hypothesis that measures of within network and between-network connectivity would be heritable, with the strongest genetic effect evident for early maturing networks involved in perception and movement and the weakest genetic effects evident in later maturing networks involved in EF and SC. I took a slightly different approach in Aim 2. Once again, a neonate specific AAL atlas is used to define 90 regions covering the neonatal cerebral cortex, but instead of creating a 90 x 90 correlation matrix, I focused on functional connectivity between the right precentral gyrus and five specific anatomical brain regions: right and left calcarine cortex, right and left lingual cortex, and right supramarginal cortex This greatly reduced the number of multiple corrections needed and hones in on the main hypotheses of the study. 4 Table 1.1 The eight adult canonical resting-state networks and their subcomponents. Canonical network Subcomponents Function Somatosensory Precentral gyrus A network of structures in Rolandic operculum the brain that interpret pain, Supplementary motor area body position, touch, and Postcentral gyrus temperature. Paracentral lobule Heschl gyrus Superior temporal gyrus Default Mode Superior frontal gyrus The default mode network4 Orbitofrontal cortex is associated with the Anterior cingulate gyrus neural correlates of the Posterior cingulate gyrus everyday mind at rest, Angular gyrus being anticorrelated with Precuneus task performances5. Middle temporal gyrus However, the DM network has shown activation in SC tasks6,7. Limbic Orbitofrontal cortex A network involved in Olfactory sense of smell, Rectus gyrus socioemotional Parahippocampal gyrus functioning8, and hedonic Temporal pole processing9. Frontoparietal Middle frontal gyrus Heavily involved in EF, the Orbitofrontal cortex FP network processes Inferior frontal gyrus goal-directed behavior and Inferior parietal lobule is involved in higher-order cognitive processing10 Ventral Attention Insula Recruited for stimulus- Middle cingulate gyrus driven attention control11 Supramarginal gyrus Subcortical Hippocampus A network involved in Amygdala emotional control, memory Caudate storage, planning, reward Putamen processing. Pallidum Thalamus 5 Table 1.1 (cont’d) Dorsal Attention Superior parietal gyrus Voluntary orientation of Inferior temporal gyrus attention11 Visual Calcarine cortex Involved in spatial Cuneus awareness, object Lingual gyrus recognition, and Superior occipital gyrus processing of visual Middle occipital gyrus information. Inferior occipital gyrus Fusiform gyrus Diffusion Tensor Imaging In Aim 2, in addition to assessing functional connectivity, I also used diffusion tensor imaging (DTI)12 to assess the integrity of white matter tracts involved in EF and SC. White matter tracts, or fasciculi, are bundles of axons connecting anatomically distinct regions of the brain. They can be visualized and characterized in either two or three dimensions. Using the principles of Brownian motion, which is based on the random movement of water molecules in this specific case, one can analyze diffusion of water molecules present in a voxel. A voxel, similar to a pixel but in three dimensions, is a unit of volume that can be measured by MRI. The distribution and displacement of the water molecules within this voxel conveys structural information about the brain as tissues may not share similar qualities in all directions, resulting in different and detectable patterns of motion. When the diffusion of water molecules is directional this is referred to as anisotropic diffusion. In contrast, isotropic diffusion is the same in all directions. In the case of white matter, diffusion in the direction of the fibers is faster than the perpendicular direction where it is restricted by cell membranes, the myelin sheath, and microfilament. It can then be assumed that the direction of the fastest diffusion would point to the overall 6 direction and orientation of the fibers. Consequently, one can use diffusion to reconstruct and visualize major fiber tracts within the brain including projection fibers, which connect the cortex to the lower parts of the brain and brainstem, association fibers, which connect different cortical regions in the same hemisphere, and commissural fibers, which are interhemispheric connections. Major white matter tracts in the brain are listed in Table 1,2 along with the brain regions they connect and the functions they support. Simplified, the diffusion of water within a voxel can be imagined as an ellipsoid. Axial diffusivity represents diffusivity along the principal axis while the minor axes represent radial diffusivity. Axial diffusivity has been related to axonal integrity and radial diffusivity to degree of myelination of the axons13. From axial and radial diffusivity, fractional anisotropy, which reflects the preferred orientation of water diffusion, and mean diffusivity, representing the mean amount of water diffusion, can be calculated. Structural disintegration has been associated with increases in mean and radial diffusivity while higher fractional anisotropy and axial diffusivity are related to preserved fiber integrity14. Quantitative analysis of these diffusivity indices along reconstructed fiber tracts is known as quantitative tractography. 7 Table 1.2 Major white matter tracts in the brain organized by fiber type. Tract/Fasciculus Connections Function Association Arcuate Temporal and Language processing15, parietal lobe to frontal executive control16, lobe attention Inferior fronto- Occipital lobe to Visuospatial processing, occipital frontal lobe planning, facial recognition 17 Superior Occipital and parietal EF including set shifting longitudinal cortices to frontal and working memory18, cortices visuospatial processing, attention19 Inferior longitudinal Occipital areas to Integration of visuo- anterior temporal behavioral processes, regions facial recognition, object recogntion20 Cingulum Frontal and parietal, Episodic memory as well as subcortical (hippocampal portion), structures to the attention, processing cingulate gyrus speed21 Uncinate Orbitofrontal cortex to Response inhibition, temporal lobe reward processing Fornix Hippocampus to Episodic memory, mammillary bodies of specifically recall 22 the hypothalamus, preoptic nuclei, ventral striatum, orbital cortex and anterior cingulate cortex Commissural Corpus callosum Cortices of left and Transferring sensory, right hemispheres motor, and cognitive information from both cerebral hemispheres to each other23 Optic tract Retina to visual Vision cortex in occipital lobes 8 Table 1.2 (cont’d) Optic radiation Geniculate nucleus of Vision thalamus to primary visual cortex Projection Corticospinal Motor and Voluntary motor function24 somatosensory cortices to brain stem Corticofugal Cerebral cortex to Modulation of sensory brainstem and spinal information25 cord Corticoreticular Primary motor cortex Gait function and postural to pontomedullary control26 junction Corticothalamic Cerebral cortex to Sends sensory thalamus information for relay into other portions of brain Executive Functions (EF) Definition of EF and Component Processes EFs are a group of top-down cognitive processes supporting goal-directed behavior. These processes include flexible thinking, working memory, and inhibitory control27,28. Flexibility refers to the ability of switching between tasks and changing behavior based on inputs from the environment. Common tasks meant to test this process are flanker-type tasks29 and Stroop30 tasks. The inhibitory aspect of EF also plays a role in flexibility, as inhibitory control allows individuals to ignore goal-irrelevant stimuli. In contrast to long-term memory, working memory consists of the small amount of information that can be held by the mind for cognitive functioning, which is required for both flexibility and inhibitory control. Working memory can be probed with tasks such as Corsi block31 or digit span32. When used together these functions go on to support higher 9 order executive functions such as planning, reasoning, and problem solving. Working memory and inhibitory control can be broken up into multiple subdomains. Working memory includes active maintenance (maintaining information over a delay), flexible updating (updating or replacing information), and interference control (selective attention to stimuli). Inhibitory control can be divided into goal selection (suppression of stimuli inappropriate for accomplishing a given goal) and response selection (choosing an appropriate behavior or movement and inhibiting those when inappropriate as a response)33. Relevance of EF to Psychiatry Atypical EF is a common thread uniting many psychiatric disorders. Attention deficit hyperactivity disorder34–36 (ADHD), depression37,38, schizophrenia39–41, and the autism spectrum disorders42–44 (ASDs) all show deficits in some or all aspects of EF. Deficits in EF play a central role in ADHD, a condition characterized by inattentiveness, hyperactivity, and impulsivity with a prevalence of 9.4% in children between the ages of 2-1745. The core diagnostic criteria for ADHD involves deficits in attention, working memory (particularly response selection), and flexible thinking, all important aspects of EF46. These individuals typically are known to struggle in shifting focus between tasks, selecting an appropriate behavioral response, and accessing their working memory 47. As research into ADHD continues to develop, there has been a shift in conceptualizing the disorder as merely a behavioral problem to a condition of EF deficits that result in the observed phenotype. 10 Neurobiology of EF The prefrontal cortex (PFC) plays a critical role in EF. A meta-analysis performed by Yuan and Raz48 found that larger and thicker prefrontal cortex volumes were associated with better EF performance. The PFC is integral for spatial attention 49 and mediates both top-down and bottom-up cognition50. The PFC is also considered the “circuit breaker”50 for response inhibition, turning on to suppress a response or partially activating to pause a response51. The parietal lobe additionally plays an instrumental role in EF52, coupling with the PFC to form the frontoparietal network (FPN) which is heavily involved in working memory50. The temporoparietal junction, a component of the FPN, is key for reorientation of attention when presented with a stimulus 53,54. Volumetric reductions in the parietal lobe additionally can help cause reduced response inhibition. When coupled with a volumetric reduction in the frontal cortex, and working memory overall is decreased55,56. The inferior and posterior portion of the parietal cortex additionally support the function of the PFC through their involvement in working memory57–60 and cognitive control61–63, respectively. The coupled frontal and parietal cortices essential for EF compose the FPN, one of the most highly globally connected networks in the brain64,65. EF has a strong and well- studied association with rs-functional networks, with the FPN at the forefront of investigation. Reineberg et al.66 demonstrated that individual differences in an overall and general EF measure correlated with variation in the frontoparietal network in its connection to other resting-state networks. The FPN is associated with flexible thinking67. Interestingly, the DMN shows an inverse level of activity during activation of the FPN. The DMN is characterized by low activity during focused attention and high activity when the 11 mind is not involved in behavioral tasks4,50,68. Disruption to this deactivation is a proposed basis for mental health disorders. Schizophrenia, for example, has been shown to have hyperconnectivity of the DMN with the FPN due to a lack of deactivation of the DMN, leading to deficits in EF68–73. Frontal and parietal regions involved in EF are anatomically connected by axonal bundles such as the superior longitudinal fasciculus (SLF) which is a key white matter tract involved in EF cognitive processes19,74,75. Inhibition has been shown to be inversely correlated with mean diffusivity in the SLF. Increased fractional anisotropy in the SLF is associated with increased skill level in set shifting, a key component of EF where the brain is able to unconsciously switch between tasks19. Working memory is also heavily influenced by SLF integrity, with increased FA associated with better working memory skills18,76,77. This pattern of altered diffusivity in the SLF is seen in schizophrenia 78 and ASD79, and is supported by lesion studies80,81. Development of EF and Associated Neural Circuits Development of the prefrontal cortex is protracted in comparison to other brain regions, continuing into the third decade of life 82. The brain develops in a “back-to-front” manner reflecting phylogeny, with the highly specialized frontal cortex developing last 83 and the more primitive portions developing first84. The frontal cortex itself also develops in a “back-to-front” manner. The PFC can grow and differentiate independently from other areas and in mice it has been shown that specific fetal growth factors regulate the regionalization85,86 even before innervation from afferent axons is made 85. The PFC delineates into distinct regions including the medial, lateral, and orbitofrontal aspects 87 12 beginning the third month prenatally88. In infancy, specifically between birth and two years of age, the prefrontal cortex develops rapidly. The development of EF abilities coincides with the development of the prefrontal cortex, with an extended development allowing for vulnerability to perturbation and environmental influences. Executive attention, or the ability to block interfering information from current attention, develops around four months of age 89. Other aspects of EF such as flexibility and set shifting emerge between five and eight months of age 90, with appreciable working memory developing just before the sixth month91. Though there is a detectible temporal pattern in emerging aspects of EF, this time period is especially sensitive to disturbances that have long lasting impacts. Preterm birth 92,93, influences from parenting90, maternal depression and anxiety94, and prenatal maternal stress95 can all be factors in determination of long-term EF outcomes. It his hypothesized that because the infant brain changes so dynamically during this early developmental changes, integration of aberrations can all have long-lasting, permanent effects. During childhood, the brain reaches its final adult volume at approximately 6 years of age96. The grey matter volume increases in the frontal lobe peaks at around 12 years of age, or just ahead of puberty, and is followed by a decline in adulthood 97. The grey matter development and subsequent pruning follows an “inverted-u shaped” trajectory98. This second surge of neuronal growth is one of the most dynamic periods of development only after infancy. The brain undergoes a rewiring until approximately 24 years old with drastic changes made to the prefrontal cortex in terms of synaptic pruning and myelination82. 13 The FPN is detectable prenatally but is the last of the canonical networks to show adult-like integration between itself and other networks. While adult brains rely on far- reaching connectivity functioning between networks, the connectivity of the brains of infants are mostly based on anatomical connections, or regions physically near each other99. Through time, segregation, or a general decrease in correlation strength in anatomically near regions, and integration, an increase in correlation between regions not necessarily anatomically connected, occurs99,100. This increase in between-network connections allow for a more efficient brain in terms of processing. Regions of higher order cognitive function, such as the FPN, maintain their own spatial and developmental independence moving through development101 and lag behind those more fundamental functions again defined by phylogeny. The SLF is a key connector that the brain heavily relies on for EF. It as well has protracted growth, not easily identifiable in its full form even after birth 102,103. It is still noticeably small in the developmental period of 3 to 12 months104 with lower FA up to 13- 24 months105. Elongation of major white matter tracts, such as the SLF, are generally not completed until around nine months, while the myelination process continues into late childhood and adolescence101,106. Genetic architecture of EF and associated neural circuits The heritability of EF has been estimated at 86%-100%107–109. A common, latent EF variable determined by Friedman and colleagues107 was estimated at 99%, with the subdomains of updating and set shifting having an estimated heritability of 56% and 42%, respectively. Working memory specifically has had reported heritability between 42% and 46%110 and inhibition is estimated at 60%111. Polygenic risk scores have established 14 relationships between EF domains and psychiatric disorders. Schizophrenia shows association with poor cognitive flexibility, ADHD and depression for inhibitory control, and bipolar disorder with working memory112. Genetic factors are also the main source for phenotypic correlations between flexible thinking, working memory, and schizophrenia 113– 115. These genetic factors driving the disorders manifest in neuroimaging endophenotypes that may allow for earlier diagnosis of psychiatric disease. These studies have primarily been conducted in adults. In the second chapter of this dissertation, we extend the current literature by estimating the heritability of resting-state networks involved in EF during early infancy. Sex Differences in EF In early childhood males perform more poorly than females in attention, working memory, and inhibitory control116–118. In a meta-analysis of sex-differences in adults in EF, heterogeneity in results was a major finding and was attributed to heterogeneity in the administration of the cognitive tests119. However, it was demonstrated consistently that males scored higher than females in working memory tasks and females performed better in response inhibition119. The prevalence of ADHD in the population is estimated anywhere from 4-15%120– 123 in grade school children with the ratio of males to females observed at 3:1 124,125. Different symptomology is associated with each sex, with females showing more inattentiveness and males having higher rates of impulsivity126–128. Females are also more likely to be diagnosed with anxiety and depression prior to being diagnosed with ADHD with females being 5.4 times more likely to be diagnosed with major depression before their ADHD diagnosis127,129,130. Females are also more likely to have comorbid psychiatric 15 disorders and less externalizing problems than males127,129,131,132. The mechanisms contributing to sex differences in EF and EF-related disorders are an area of ongoing debate. Chapters 3 and 4 of this dissertation will help address this question by (1) determining how the loss of a second sex chromosome influences white matter tracts and brain networks involved in EF in infancy, and (2) creating an online platform for testing EF in adult women with X monosomy. Social Cognition (SC) Definition of SC and Component Processes SC, or the “knowing of people”133, is the set of psychological processes that allow individuals to engage and take advantage of being part of a social group 134. The Research Domain Criteria (RDoC)33 breaks SC into multiple subdomains. These include attachment, perception and understanding of others, perception and understanding of self, and social communication which is further broken down into reception and production of facial and non-facial communication135. According to the RDoC, attachment is selective engagement in positive social interactions with other individuals which develops a social bond. Attachment involves the processing of social cues as well as social learning and memory that are associated with the forming and maintenance of relationships 33. Facial and non-facial communications involve one’s ability to both convey and perceive an emotional state of another nonverbally via facial expressions, or non-verbal gesture, affect, and body language33. Theory of Mind (ToM) is the understanding that others have beliefs, desires, and intentions that differ from an individual’s own and is also a key component of SC136. Measures of SC can include tests such as Reading the Mind in the 16 Eyes, a task that requires an individual to look at a set of eyes and select which emotion they believe the eyes are conveying137. This probes both ToM and facial communications. Additional testing measures include other emotion recognition tasks, such as the Penn Emotion Recognition Task138 and the Bell Lysaker Emotion Recognition Task139 which both have been evaluated to have useful psychometric properties. Tests for SC outside of the emotion recognition tasks include the Edinburgh Social Cognition Test140, which addresses ToM and an understanding of social norms, and The Awareness of Social Interference Test141. Relevance of SC to Psychiatry Social cognitive disturbances arise in numerous psychiatric disorders and sometimes form the foundation of the diagnostic criteria. This is the case for autism spectrum disorders (ASDs)142. Characterized by profound impairments in interpersonal interaction and communication, levels of SC are directly related to levels of social functioning 143. Though there are many changes in the degree of severity in certain symptoms of ASD over life, social difficulties tend to remain the most stable 144. Schizophrenia-spectrum disorders145,146 are also characterized by social cognitive deficits. While ASD is characterized by “insufficient” inferences, schizophrenia can be classified by its “excessive” mental attributions, leading to the hypothesis that the two are diametrically related disorders of SC147,148. In schizophrenia, levels of cognition and overall functioning are mediated by SC145,149, which has been reported as the greatest unmet treatment need in the disorder150. ADHD also has a component of social cognitive deficiency in childhood and adolescence, showing poorer abilities in recognizing emotion facially in others151–153. Like in schizophrenia, individuals with ADHD show increased functional impairments as 17 social cognitive abilities decrease154 taking the pattern of a loop in which improvement in one is needed to improve the other155. Neurobiology of SC In SC, ToM has been shown to be associated with the left medial prefrontal cortex156–158 and the amygdala, part of the limbic system, in facial expression recognition159–161. The orbitofrontal cortex is implicated in understanding and displaying appropriate social behaviours162–164. The temporoparietal junction is an additional area with strong support for its role in SC165 and is thought to control representation of the self or another individual166 153. ASD with its social cognitive deficits show volumetric abnormalities, specifically in the parietal lobe. Children with ASD show localized reduction in grey matter in the parietal networks167 which are implicated in various aspects of SC168. Specifically, research has shown a role for the inferior parietal lobule in SC169 in ASD, displaying a negative correlation between grey matter volumes and social cognitive scales. The DMN is a major component of the “social brain” and is involved in empathy, morality, ToM, and perceiving and interpreting other’s emotional states 170. Three main components of the DMN are the temporoparietal junction, the posterior cingulate cortex, and the medial prefrontal cortex. Within-network connectivity of the DMN shows an inverse correlation with the severity of autistic traits163171. Aberrant within-network functional connectivity is additionally observed in ADHD in the DMN172. Both ASD and ADHD are characterized by their unique social behaviors, and the DMN appears to have a key role in this system. The temporoparietal junction is also implicated in SC including 18 theory of mind173, empathy174, and moral judgements175 176, and is a necessary component for mentalizing based on lesion studies173. Depression and schizophrenia, which both involve disruptions in higher-level social functioning, show lower connectivity between the temporoparietal junction and the posterior cingulate cortex and overall show reduced resting-state functional connectivity between the temporoparietal junction and other key regions of social processing177. Aberrant functional connectivity was additionally found between the temporoparietal junction and key areas that control behavior and SC in a study comparing individuals with major depressive disorder to controls178. The integrity of specific white matter pathways is also important to the function of the “social brain”. The inferior longitudinal fasciculus, the inferior fronto-occipital fasciculus, and the superior longitudinal fasciculus are all linked to facial recognition and processing179. ASD is often accompanied by disrupted white matter microstructure180–182, and lower fractional anisotropy of the superior longitudinal fasciculus has shown to be positively correlated with social interaction scores in ASD183–185 The inferior longitudinal fasciculus (ILF) is also implicated in SC as it is heavily involved with emotional facial recognition20, with lower FA being reported in disorders of SC183,186. The ILF is directly connected to the amygdala whose relation to SC has been previously touched upon. The arcuate fasciculus, which connect the temporoparietal junction to the frontal cortex, showed reduced white matter integrity in a study of high-functioning autism that simultaneously showed weaker functional connectivity between the temporoparietal junction and frontal areas187. 19 Development of SC Like in EF, the prefrontal cortex is integral for the development of SC. The parietal networks, inferior parietal lobule, and temporoparietal junction all play a role in SC as well. The prefrontal cortex has been discussed in terms of prenatal development above. The parietal lobe can be defined visually between gestational weeks 14 and 26, with the formation of the parieto-occipital sulcus188,189. In infancy at approximately 2 months of age the inferior parietal areas are used in facial recognition 190. In the second year of life, the inferior parietal cortices show the fastest rate of surface area expansion compared to other regions191. The grey matter of the parietal lobe increases in childhood and peaks at 10.2 years for females and 11.8 years for males, declining in adolescence 97. This pattern is similar to what occurs in the frontal lobe, but the decline is more substantial in the parietal lobe97. Aspects of SC are measurable even from the early stages of infancy. Between two and three months infants are able to distinguish between two facial expressions 192. This finding should be interpreted with caution though because although a discrimination is able to be made, understanding of the emotional meaning of the facial expression has not developed. Theory of Mind, or at least its precursors, are evident in infancy during the 12-18 month period193as evidenced by “mind reading”, or the ability to predict the goals of others194. Over development, understanding of disgust and anger in facial expressions increases drastically, whereas recognition of fear and happiness show stability from early childhood195. Between the ages of three and six, considered the preschool years, understanding of Theory of Mind exhibits conceptual changes such as understanding of false-beliefs and the beliefs of others196–198. 20 Key white matter tracts that influence SC include the inferior longitudinal fasciculus, the inferior fronto-occipital fasciculus, and the superior longitudinal fasciculus. Fetal development of white matter tracts occur in four phases, with the superior and inferior longitudinal fasciculi developing in the second wave (approximately 14 weeks gestation) and the inferior-occipital fasciculus in the fourth wave (approximately 20 weeks gestation)103. These association fibers can be identified with tractography in neonates but lack sufficient development to be detected by tractography at 20 weeks gestation199,200. During infancy the number of discernable tracts, their length and volume, and fractional anisotropy increase with age, with a plateau between ages 3 and 5201. The default mode network is a primary focus in the study of functional networks involved in SC. Functional networks can be observed in utero as early as the second trimester202–204. In a study in utero examining functional connectivity in the fetal brain, statistically significant overlap was found between adult and prenatal default mode network modules (average age 33.5 weeks gestation)202. Over gestation, these modules gain more long-range connections and the default mode network becomes more specialized205,206, with the posterior cingulate cortex acting as a hub of the network207. In full-term babies, a “proto-default network” has been described, with only a portion of the adult resting-state connections observed in these infants208. At approximately two years of age, the network gains more adult-like topology, integrating the anatomical portions missing from the network in neonates209. However, research in early school age children still show only sparse functional connectivity, though over development integration with other networks occurs210. 21 Genetic architecture of SC Vasopressin and oxytocin receptor genes playing a part in SC is a recently growing area of research, with specific relevance to schizophrenia as well as ASD211–213. For example, genetic variants in the oxytocin repressor gene show deficits in global social functioning as well as in social affect recognition213,214. The genetic underpinnings of SC have been studied widely, with multiple association studies revealing SNPs related to social functioning215–217. A study by St. Pourcain and colleagues218 estimated the heritability of social communication at 18% which was replicated by an additional study by the group215 but shown to vary with age219. Studies focusing on the broad general concept of SC have reported heritability of 68%220, with attention to specific subdomains such as emotion identification being estimated at 36%221. Emotion identification specifically has been shown to associate with polygenic risk for schizophrenia222, a disorder that has been previously discussed for its great influence on SC. Studies involving neuroimaging endophenotypes as markers for these genetic factors have primarily been conducted in adults. Chapter 2 of this dissertation will extend the current literature by estimating the heritability of resting-state networks involved in SC during early infancy. Sex differences in SC Sex-differences in social cognitive testing are very dependent on the test administered223. However, typically developing females have been shown to have a greater general ability to decipher facial expressions than in typically developing males 224– 226, have better nonverbal communication227,228, and are better at expressing emotional 22 experiences229. Women also tend to score higher than men on empathy ratings and emotional regulation230. The autism spectrum disorders additionally show a sex-bias. The ratio of diagnosis in males to females is estimated at 3:1231,232 to 4:1233,234 with prevalence in the population observed at approximately one percent234–236. Females are typically diagnosed later than males237 and have more pronounced difficulties with social communication and interaction238,239 at a young age compared to the male ASD population. However, girls with ASD that lack cognitive impairment are consistently better at camouflaging, or using compensatory behaviors, to mask ASD’s associated social challenges 240–242. Another area of observed differences in clinical presentation of ASD between the sexes is in cases of intellectual impairment. With intellectual disability, the symptomology of ASD in girls is more severe243 and the sex ratio becomes closer to 2:1 M:F whereas the ratio climbs to approximately 11:1 M:F in individuals without intellectual impairment 244,245. The mechanisms contributing to sex differences in SC and SC-related disorders are an area of ongoing debate and that Chapters 3 and 4 of this dissertation will help address this question by (1) determining how the loss of a second sex chromosome influences white matter tracts and brain networks involved in SC in infancy, and (2) by creating an online platform for testing SC function in adult women with X monosomy. Turner syndrome as a model for understanding sex chromosome effects on EF and SC Described by Henry Turner in 1938246, Turner syndrome (TS) or X monosomy occurs in approximately 1 in 2000 live female births and results from a complete or partial absence of the second sex chromosome247. Mosaicisms may occur, in which different 23 cells in the body contain different chromosomal arrangements. Common phenotypes of TS include congenital heart defects, renal abnormalities, liver disorders, and gonadal dysgenesis248,249. Girls and women with TS often exhibit a unique cognitive profile with normal or above average intelligence and deficits in three specific cognitive domains. These include EF, SC, and visuospatial processing. These skills are essential for everyday functioning and influence short-term memory recall, social interactions, and an understanding of space and measurement. These neurocognitive domains affected in TS shows great relevancy to both ASD and ADHD as they share similar deficits250,251. TS can also be comorbid with ASD and ADHD. TS individuals are more likely to be diagnosed with these disorders than XX females. In one early study, Turner syndrome girls had a risk of developing ASD 300 times higher than typically developing girls252. As a note, the study had a very small sample size and worked under the assumption that the population prevalence of ASD in females was 1:10,000, the accepted prevalence at the time. A more updated study put the estimation at four times more likely and used a registry system, yielding a much larger sample size253. In a separate study, an 18-fold increase in the prevalence of ADHD in TS girls compared with girls in the general population was found, equating to a prevalence of approximately 25% in TS girls254. The ADHD presentation is also more similar to males with ADHD than females, with higher hyperactivity scores being reported for females with TS255. However, the aforementioned registry study did not find any association between TS and ADHD, though this could be due to its retrospective design and the underdiagnosis of ADHD in TS253. Elevated risk for male-based neurodevelopmental conditions in phenotypically female individuals with TS may have a genetic explanation. 24 Females with TS are more similar to typically developing males in the sense that 1) both are hemizygous for genes that escape X inactivation (approximately 15%256), and 2) there is only the presence of one X chromosome, uncovering X chromosome mutations. Differences in neuroanatomy and brain connectivity have also been observed in TS and align with the known neurocognitive challenges. For example decreased grey matter volumes are consistently reported in the parieto-occipital cortex 257–261, which plays a key role in visuospatial processing. Consistently, reduction in volume of the parietal lobe and increase in volume of the amygdala and orbitofrontal cortex has been reported257,259,262,263. White matter volume increases have also been consistently observed in the temporal lobe in TS which is implicated in language and SC 257–259.. TS has a very distinct structural connectivity profile as well which is supported by the literature. Multiple studies have reported reduced FA in the superior longitudinal fasciculus (SLF) with others reporting more global reductions258,264,265. As discussed previously, the SLF is a part of brain networks that process working memory, language, visuospatial attention, and numerical tasks76,266,267. Functional connectivity studies have also contributed to our understanding of the neurocognitive profile of X monosomy. Numerous lines of investigation have demonstrated altered functional connectivity in TS,268 which have implications to the symptomology of the disorder, including the neurocognitive domains269–271. Although the work was done using task-based fMRI, Bray and colleagues271 were the first to demonstrate abnormalities in the parieto-occipital and parieto-temporal pathways, which implies a root to the deficits in visuospatial processing. A recent study of infants with TS indicates that volume differences are already present at 25 one year of age272, but no studies have looked at the DTI or resting-state MRI phenotypes in TS infants. This dissertation will address this gap in Chapter 3. There is significant heterogeneity within TS in terms of the cognitive phenotype, which remains unexplained. Thorough cognitive testing is needed to understand qualitatively and quantitatively the unique domains affected in TS. This ability is currently limited by both access to resources such as clinical testing centers and licensed software as well as geographical constraints based on location of the testing facility. A browser- based online platform is needed to facilitate large-scale studies of how genetic and clinical factors influence EF and SC variation within TS. A solution to this challenge is approached in Chapter 4 of this dissertation. The importance of studying EF and SC circuits during infancy To date, the vast majority of imaging genetics has been performed on adults and adolescents, missing a critical window in which vulnerability to psychiatric disorders is established in early life. A high-familial risk study by Auerbach and colleagues273, with groups defined by severity of paternal presence of ADHD, shows ADHD may also begin to emerge in infancy as developmental pathways are enacted 274,275. ASD is another disorder that’s descent into clinical presentation is focused on infancy and early childhood276,277, but evidence of abnormal developmental changes are discernable prenatally as well. ASD has recently been considered prenatally derived due to the exponential proliferation of cortical neurons between ten and twenty weeks of gestation that result in the hallmarked overgrowth of the cerebral cortex later in life278. Additionally, most ASD risk genes are expressed prenatally in higher levels than postnatal development278,279. Altered prenatal gene expression for ADHD has also been reported, 26 with differentially methylated genes also playing a role280,281. Overall, the fetal and infant brain are growing at an exceptionally high rate, making it potentially vulnerable to both genetic and environmental disruptions. Prenatal stress, which includes physical stress, nutrition, hormonal, substance use, infection, and psychological stress, can have long lasting effects on the brain, specifically in cognition and susceptibility to the future development of psychiatric disorders. Famine and maternal malnutrition during pregnancy has been shown to result in higher levels of schizophrenia in offspring 282–284. Maternal psychiatric stress has also been shown to increase the probability of the presentation of anxiety and depression in offspring later in adolescence and adulthood 285– 287. In addition, there are specific challenges in using an adult population to study the genetic influences on brain anatomy and function. Adults carry with them years of behavioral-environmental interactions and, in the individuals with chronic psychiatric illnesses, the added complication of prolonged use of psychotropic medication. For example, the use of lithium in bipolar disorder is characterized by increases in grey matter volumes as part of the treatment response288. Conversely, the use of antipsychotics is linked to decreased grey matter volumes in individuals taking these medications 289. Examining the influence of genetic risk for bipolar disorder on brain development in presymptomatic individuals, beginning in infancy, may provide clearer insights into etiology. TS, as previously mentioned, is characterized by short stature and gonadal dysgenesis, among other defining features. As a result, most girls go through some type of hormone therapy to ameliorate the repercussions of the illness290. Individuals may also take medications for heart problems, renal problems, or take psychotropics 291. 27 Medications, including hormone therapy, could cause changes in brain structure and function in these individuals that are not directly related to X chromosome loss. Infant neuroimaging could also facilitate the development and implementation of early interventions, which would increase quality of life for affected individuals. In the work of Emerson et al.292, the group was able to use functional connectivity measures from six- month-old infants as a predictor for the diagnosis of autism spectrum disorder with a positive predictive value of 100%. This demonstrates that at-risk individuals can be identified based on their infant neuroimaging phenotypes. The outcomes of autism spectrum disorder are heavily reliant on early detection and intervention 292. Dawson and colleagues293 demonstrated that specialized early intervention significantly improved IQ and adaptive behavior over a two-year period beginning in early childhood in comparison to children receiving only standardized care from a physician. Rationale for the dissertation The overarching goal of this dissertation is to determine how genetic variation influences neural circuits and cognitive domains relevant to psychiatric illness with a particular focus on EF and SC. In this introduction, I have demonstrated that EF and SC are heritable, disrupted in multiple psychiatric disorders, differ between males and females, and linked to brain regions and neural circuits with protracted developmental trajectories that can be studied via neuroimaging. I have also identified specific research gaps that will be addressed by my dissertation studies. Specifically, in Aim 1 I will identify genetic and environmental factors influencing the FPN, DMN, and other resting-state networks in infancy, a period that has been relatively understudied. In Aim 2, I will determine if differences in functional connectivity and white matter integrity, identified in 28 adults and adolescents with TS and linked to EF and SC deficits, are also present in infancy. Finally, in Aim 3 I will develop a browser-based online cognitive testing platform targeting EF, SC, and other areas of challenge in TS. This will facilitate large-scale studies of how genetic variation influences SC and EF in TS. In the following paragraphs, I will explain each aim in greater detail. Together, the successful completion of the three aims provides novel information about genetic influences on EF, SC, and connectivity in associated neural circuits. Ultimately, understanding the genetic basis for variation in EF and SC could open up new opportunities for identifying at-risk individuals in early life and lead to the identification of treatment targets. In my first aim I sought to identify genetic and environmental factors influencing neonatal resting-state functional connectivity including connectivity within and between the frontoparietal and default mode networks, which play critical roles in EF and SC respectively. I hypothesized that measures of within-network and between-network connectivity would be heritable, with the earliest maturing networks having the strongest genetic effect. Conversely, I hypothesized that the later maturing networks, FP and DM for example, would be under the least amount of genetic control in this developmental period. Additionally, I hypothesized that maternal psychiatric history would be associated with within- and between-network functional connectivity in the limbic network, which also plays a role in SC. Accomplishment of this aim defines an incredibly early timepoint in the determination of genetic control over the brain’s connectivity profile. This timepoint links prenatal findings with those in older infants, helping to create a more complete map of the genetic architecture of the developing human brain. 29 In the second aim, I compared the anatomical and functional connectivity differences in infants with TS to typically developing children of the same age. I hypothesized that infants with TS would have aberrant diffusivity in the SLF as well as reduced functional connectivity between the precentral gyrus and regions involved in basic visual processing (calcarine cortex), SC (supramarginal gyrus and lingual cortex), and EF (supramarginal gyrus). Successful completion of this aim will have a positive impact on the field by having the potential to show whether differences in structural and functional connectivity observed in older children and adults with TS are present in infancy, versus emerging at a later developmental stage. This has implications for early intervention strategies and therapies for TS that could help ameliorate neurocognitive deficits in the disorder. In my final aim, I created an online browser-based cognitive battery targeting domains that are often disrupted in TS including EF and SC. I validated the battery by comparing females with TS, XX females, and XY males. I hypothesized that, overall, the TS group would perform significantly worse on the tests in all domains, with longer response times and lower accuracy on the tests. At the completion of this aim, I will have a validated set of tasks that can be used to measure cognition in TS on an online platform that can be done from anywhere with an internet connection and circumvents the need for in person testing. In person testing limits the geographical area a researcher can pull from; thus, this objective allows for an expansion of women with TS that can be reached. My work provides the foundation for a future large-scale study of genetic influences on cognition in TS. 30 Chapter 2 Genetic and Environmental Factors Influencing Neonatal Resting-State Functional Connectivity 31 Abstract Functional magnetic resonance imaging has been used to identify complex brain networks by examining the correlation of blood-oxygen-level-dependent signals between brain regions during the resting state. Many of the brain networks identified in adults are detectable at birth, but genetic and environmental influences governing connectivity within and between these networks in early infancy has yet to be explored. We investigated the genetic influences on neonatal resting-state connectivity phenotypes by generating intraclass correlations and performing mixed effects modeling to estimate narrow-sense heritability on measures of within network and between-network connectivity in a large cohort of neonate twins. We also used backwards elimination regression and mixed linear modeling to identify specific demographic and medical history variables influencing within and between network connectivity in a large cohort of typically developing twins and singletons. Of the 36 connectivity phenotypes examined, only 6 showed narrow-sense heritability estimates greater than 0.10, with none being statistically significant. Demographic and obstetric history variables contributed to between- and within-network connectivity. Our results suggest that in early infancy genetic factors minimally influence brain connectivity. However, specific demographic and medical history variables such as gestational age at birth and maternal psychiatric history may influence resting-state connectivity measures. 32 Introduction Historically, functional magnetic resonance imaging (fMRI) studies of the brain have focused on localizing functions to specific cortical and subcortical areas without addressing the functional integration of different brain areas, which is essential for complex behavior. More recently, resting-state functional connectivity studies have provided new insights into large-scale brain organization by examining the temporal correlation of blood-oxygen-level-dependent (BOLD) signals between brain regions that are not necessarily anatomically connected 294. Regions with strongly correlated BOLD activity are thought to represent functional networks. Around ten major canonical networks have been identified in adults 295, including the default mode, somatosensory, frontoparietal, visual, dorsal attention, ventral attention, limbic, and subcortical networks. Understanding individual differences in resting-state connectivity is of interest because these networks play critical roles in perception 296–299 and high-level cognition 67,300,301, encompassing social cognition 302–304, executive functioning 66,305, and attention 306,307. Furthermore, abnormalities in these resting-state connectivity networks are associated with the development of psychiatric diseases. Schizophrenia 308–310, obsessive compulsive disorder 311, and autism spectrum disorders 312 all show distinct patterns of within- and between-network connectivity when compared to typically developing individuals. The prenatal and early postnatal period represents a critical epoch in the establishment of resting state networks. In utero, the intrinsic connectivity of the entire human brain increases over time, with the greatest period of change observed between 26- and 29-weeks gestation. This period of rapidly increasing connectivity occurs first in 33 occipital cortex, and then in temporal, frontal, and parietal cortex 313. Modularity is present in the fetal human brain, but in contrast to canonical brain networks measured in adulthood, which often include spatially distant brain areas, these fetal modules are spatially restricted. Modularity is defined in the context of networks as separate clusters of function within the brain as a whole that interact more with themselves than in the network community. Some of these modules may represent very early versions of adult canonical networks including the sensorimotor network and the default mode network. Interestingly, modularity decreases during the third trimester, suggesting more integration between different brain regions 205. Neuroimaging of neonates reveals that many resting- state networks are detectable shortly after birth 101,208,209,314–317. Some of these networks already demonstrate an adult-like topology, including the sensorimotor network, the auditory/language network, and primary visual networks. Others, involving high-order association cortices show a less mature topology. This includes the default mode network, the lateral visual/parietal network, the salience network, and frontoparietal networks involved in executive function 318. The maturation of canonical networks during the first two years of life follows a specific temporal sequence, beginning with primary sensorimotor/auditory networks, proceeding to visual networks, then to salience and default-mode networks, and finally to executive control networks. During this same period, and later in childhood, within-network connectivity strengthens while between-network connectivity wanes 319. Individual differences in the development of resting-state connectivity could be due to both genetic and environmental factors. In adults, resting state connectivity metrics are under modest‐to‐moderate genetic control 320,321, with stronger heritability observed in 34 within- as opposed to between-network connectivity 322. The default mode network, for example, has been reported to have a heritability of 0.424 for functional connectivity measures 320, with an average heritability of connection strengths in adults for a BOLD time course at 15-18% 321. In adolescents, longitudinal age-related changes in both between- and within-network connectivity are pervasive, with up to 53% of the variation attributable to heritability 323. Studies on genetic contributions to resting-state connectivity in infancy are extremely limited. In a 2014 paper, Gao and colleagues 324 used voxel-wise general linear model (GLM) regression analysis to test the effect of genetic relatedness on voxel-wise functional connectivity patterns in neonates, one-year-old infants, and two- year-old infants. In neonates, genetic effects were observed in the posterior cingulate cortex, cuneus, fusiform gyrus, and bilateral middle and inferior frontal gyri. In one-year- olds, genetic effects were observed in the paracentral lobule, posterior cingulate cortex, bilateral lingual gyrus, right middle occipital gyrus, right angular gyrus/parietal-temporal- occipital junction, bilateral middle and inferior frontal gyri, and ventromedial prefrontal cortex. In two-year-olds, genetic effects were observed in the left middle temporal gyrus and left supramarginal gyrus. The authors also reported modest genetic effects on within- network connectivity that varied with age but did not examine between network connectivity. Because infancy is a critical period in the establishment of functional brain networks, the pre- and perinatal environment could also be important factors in explaining inter-individual variation in brain connectivity. Two of the most studied factors to date include maternal depression and prenatal drug exposure. Regarding maternal depression, high prenatal maternal depressive symptoms are associated with greater 35 functional connectivity between the amygdala and an array of brain regions involved in emotional regulation including the left insula and bilateral anterior cingulate, medial orbitofrontal and ventromedial prefrontal cortices 325. Interestingly, similar patterns of altered connectivity have been observed in adults and adolescents with major depressive disorder 326–328. Brain networks involved in emotional regulation also appear to be particularly sensitive to prenatal drug exposures including cocaine and marijuana 329–332. Further, prenatal cocaine exposure is associated with hyperconnectivity between the thalamus and the frontal regions of the brain 333, while multiple drugs are associated with a hypo-connectivity between the thalamus and motor-related regions. Many other aspects of the pre- and perinatal environment, which are known to have major effects on neonatal brain structure 334, have yet to be examined in relation to functional connectivity. The main objectives of the current study are to (1) investigate genetic influences on neonatal resting-state connectivity phenotypes using twin data and (2) understand how major demographic and medical history variables affect neonatal resting-state connectivity phenotypes using a large, prospective cohort of singletons and twins. We address the first objective by generating intraclass correlations and performing mixed effects modeling on measures of within network and between-network connectivity. We address the second objective via linear mixed modeling. Based on the literature reviewed above, we hypothesized that measures of within network and between-network connectivity would be heritable, with the strongest genetic effect evident for early maturing networks involved in perception and movement. As discussed above, maternal depression has been linked to changes in limbic connectivity in offspring, therefore we further hypothesized that maternal psychiatric history would be associated with functional 36 connectivity within the limbic network and between the limbic network and other networks. Additionally, we hypothesized that gestational age at birth would be associated with connectivity within the somatosensory network and between the somatosensory network and other networks. We based this hypothesis on studies showing that disrupted development of the somatosensory network in preterm infants is associated with downstream cognitive deficits 335,336. Materials and Methods Participants Participants were 268 neonates (average gestational age of 261 days), both male and female, including singleton, unpaired, monozygotic (MZ) and dizygotic (DZ) twins. The current sample represents a subset of 1329 infants enrolled in the Early Brain Development Study (EBDS) 337–339 at the University of North Carolina at Chapel Hill that were imaged via MRI around 1 month post-birth and had usable resting state fMRI scans. Exclusion criteria for the parent study were set at enrollment and included the presence of abnormalities on fetal ultrasound or major medical illness in the mother. Additional exclusion criteria for the current analysis included gestational age at birth 32 weeks or fewer and missing medical/demographic information. Opposite-sex twin pairs were also excluded. Demographic variables including maternal and paternal age, education, and ethnicity, household income, maternal smoking during pregnancy, and paternal psychiatric history were collected via maternal report. Income was factorized by dividing it into three categories, each defined as falling into either at or below 200%, between 200% and 400%, and above 400% of the federal poverty level. The infant’s date of birth 37 was used to identify the federal poverty level for each specific case while also taking into account family size. Maternal psychiatric history was assigned based upon both maternal report and review of medical records. Maternal and paternal psychiatric history were considered binary variables. A positive psychiatric history included a diagnosis in any of the following DSM-V categories: autism spectrum disorders, Tourette’s syndrome, attention-deficit hyperactivity disorders, obsessive-compulsive and related disorders, anxiety disorders, mood disorders, or schizophrenia spectrum and other psychotic disorders. Medical history variables (gestational age at birth, birthweight, APGAR scores, gestation number, delivery method, and stay in neonatal intensive care unit over 24 h) were collected from maternity and pediatric medical records 337. MRI Acquisition and Processing A detailed description of image acquisition and processing can be found in 340. Briefly, functional MRI data was acquired using a T2*-weighted EPI sequence: TR=2 s, TE = 32 ms, 33 slices, and 4mm isotropic resolution, with 150 volumes acquired over 5 min. Similar scan durations have been used in prior infant imaging studies and minimize data loss due to infants waking up during the scan (Gao, Gilmore et al. 2013, Alcauter, Lin et al. 2014, Gao, Alcauter et al. 2015, Gao, Lin et al. 2017). Structural images were acquired using a 3D MPRAGE sequence: TR = 1820 ms, TE = 4.38 ms, and 1 mm isotropic resolution. Two different scanners were used: a 3T head-only Siemens Allegra during the early years of the project, and a 3T Siemens TIM Trio (Siemens Medical Supplies, Erlangen, Germany) during later years. There were 42 infants imaged on the TIM Trio and 226 on the Siemens Allegra. All infants were in a natural sleep state during the imaging session. 38 Functional data were preprocessed using the FMRIB (for Functional MRI of the Brain) Software Library (FSL; version 6.0; 341) and Analysis of Functional NeuroImages (AFNI; 342). Steps included discarding the first 3 volumes (6s), slice-timing correction, rigid-body motion correction, bandpass filtering (0.01–0.08 Hz), and removing the nuisance signals including the signals of white matter, CSF, the 6 motion parameters and their derivative, quadratic terms. The 6 motion parameters were estimated using rigid body registration for each volume with the first volume as the reference image. The nuisance signals were band-pass filtered before regression to match the frequency of the blood oxygen level-dependent signal. Data scrubbing was implemented with scrubbing criteria at global signal changes >0.5% or framewise displacement >0.3mm 343–345 with continuous data points no less than 3. This approach is commonly used in infant resting state fMRI studies in the field 346 and has been shown to minimize motion artifacts. Participants with less than three minutes (90 data points) of functional data after scrubbing were excluded and all infants’ functional data were truncated into 90 volumes to ensure consistency across subjects. We chose the three-minute threshold to balance sample size and duration of available time series after scrubbing. This threshold is widely Figure 2.1 Comparison of the functional connectivity values from three-minute and four-minute data from a subsample of 58 subjects. The left panel is the mean functional connectivity (FC) of the 58 subjects with truncated 3-minute (3min) data, the middle panel is the mean FC of the same 58 subjects with their 4-minute (4min) data; the 39 Figure 2.1 (cont’d) right panel is the scatter plot of the mean FC between 3min data and 4min data, as well as their correlation. used in the field 318,346–348, and previous test-retest studies confirm a three-minute duration provides reasonable test-retest reliability349–351. Further, when we compare a functional connectivity matrix generated using the three-minute threshold to one created using a four-minute threshold in a subset of 58 infants, we find they are almost identical (Figure 2.1), supporting the validity of a three-minute threshold. After scrubbing, a single global signal regression step was performed to remove the global signal re-introduced by scrubbing. A template-based skull-stripping method was used the extract the brain from T1 image. After performing a rigid-body registration between the functional data and T1-weighted structural images of the same subject, a nonlinear registration (fNIRT in FSL) was done between individual T1-weighted images and the neonatal template images 352. The combined transformation field was used to warp the preprocessed rs-fMRI functional images to the neonate template 353. Finally, the images were spatially smoothed with Gaussian kernel (full width at half maximum = 6mm). As outlined in Gao et al. 101, a neonate specific AAL atlas was used to define 90 regions covering the neonatal cerebral cortex, as defined by Tzourio-Mazoyer et al and Shi et al. 353,354 for infants. The average BOLD time course was extracted from each region for each subject to construct a 90 by 90 correlation matrix. The correlation matrix was Fisher-Z transformed to be the functional connectivity matrix for each subject. The 90 regions from neonate specific AAL atlas were assigned to eight intrinsic functional networks 355 based on spatial overlapping (i.e., winner-take-all approach to 40 assign each region to the network with the highest level of overlap in volume) to group all connections as either within one (both nodes within one network) or between two networks (each node belonging to a different network) (Table A.1). Subsequently, within- and between-network connectivity was calculated by averaging the FC values of the within-/between- networks for each subject. The following networks were examined: default mode, sensorimotor, frontoparietal, visual, dorsal attention, ventral attention, limbic, and subcortical network. Each network was given a number and an abbreviation as shown in Table 2.1. Table 2.1 Assigned networks numbers, network name, and network abbreviation for the eight assigned resting-state connectivity networks. Network Abbreviation 1 Visual Vis 2 Somatosensory SS 3 Dorsal Attention DA 4 Ventral Attention VA 5 Limbic Lim 6 Frontoparietal FP 7 Default Mode DM 8 Subcortical SC Statistical Analyses To investigate genetic and environmental influences on neonatal resting-state connectivity phenotypes, we calculated intra-class correlations in both MZ and DZ twin pairs with correction for residual framewise displacement, which indicates subject motion. Analyses were performed with R statistical software 4.0.1 356 using the ‘irr’ package 357 (version 0.84.1). To estimate narrow sense heritabilities we fitted mixed-effects models to data from MZ and DZ twins jointly using the nlme R-package. The model was as follows: 𝑦𝑖𝑗 = 𝜇 + 41 𝑥𝑖𝑗 𝛽𝑖 + 𝜖𝑖𝑗 where 𝑦𝑖𝑗 is an imaging phenotype (pre-adjusted by the effect of the scanner, motion correction, and sex) for the jth twin of the ith twin-pair, 𝜇 is an intercept, 𝑥𝑖𝑗 took value 1 for MZ twins and √0.5 for DZ twins, 𝛽𝑖 is a raondom effect common to a twin pair, assumed to be normally distributed with mean 0 and variance 𝜎𝛽2 ,and 𝜖𝑖𝑗 is an error term, assumed to be normally distributed, independent across subjects, with mean zero and a variance that varied by zygosity 𝑉𝑎𝑟(𝜖𝑖𝑗 ) = 𝜎𝜖2𝑀𝑍 and 𝑉𝑎𝑟(𝜖𝑖𝑗 ) = 𝜎𝜖2𝐷𝑍 for MZ and DZ twins, respectively. Note that with this setting the covariance between twins is 𝐶𝑜𝑣(𝑥𝑖𝑗 𝛽𝑖 , 𝑥𝑖𝑗′ 𝛽𝑖 ) = 𝑥𝑖𝑗 𝑥𝑖𝑗′ 𝜎𝛽2 which is equal to 𝜎𝛽2 for MZ twins and equal to 0.5𝜎𝛽2 for DZ twins. Therefore, 𝜎𝛽2 represents the additive genetic variance for the trait. For MZ twins the error variance represent the variance of non-genetic effects (on the other hand, for DZ twins the error variance also includes the variance due to mendelian sampling). Thus, 2 𝜎𝛽 the heritability of the trait was estimated using 2 . This method was also used to 𝜎𝛽 +𝜎𝜖2𝑀𝑍 estimate the heritability of the motion variable. Confidence intervals were calculating using Bootstrap, and a permutation analysis was performed to generate p-values. To understand how major demographic and medical history variables affect neonatal resting-state connectivity phenotypes, we performed AIC based backward- elimination regression of the demographic and medical history variables described in the “Participants” section. We then ran mixed effects models including the selected variables for effect size estimation and significance testing and to estimate r2 values. A Bonferroni correction for the number of mixed effect models run was applied (0.05/36 = .00139). To account for familial relatedness within MZ and DZ twins, we incorporated a standard ACE model as described in Xia et al. 358. In brief, the twin-based ACE model heritability of each 42 connectivity measure was estimated by fitting a regression model 𝑦𝑖 = 𝜇 + 𝑥𝑖 𝛽 + 𝑎𝑖 + 𝑐𝑖 + 𝑒𝑖 , where 𝑦𝑖 is the connectivity value of individual 𝑖 (𝑖 =1, …, n), 𝜇 is the overall mean, 𝑥𝑖 are covariates, and 𝑎𝑖 , 𝑐𝑖 , and 𝑒𝑖 denote the (random) additive genetic, shared environmental, and residual effects. This is done under the assumption that the three random terms are mutually independent and normally distributed with mean 0 and variances 𝜎𝑎2 , 𝜎𝑐2 , and 𝜎𝑒2 . A power analysis was performed in R by creating a function to fit a mixed-effects additive model using restricted maximum likelihood and then applying it to simulated data. The simulation included narrow-sense heritabilities of 0.1, 0.15, 0.25, and 0.5 with a simulated number of twin pairs, assuming an equal number of MZ and DZ pairs, of 25, 50, 75, 100, and 150. Because this is an additive model of inheritance, dizygotic twins were given a coefficient of relatedness of 0.5 and monozygotic twins a coefficient of 1.0. Results Participants for Objective 1 Descriptive statistics for the demographic and medical history variables can be found in Table 2.2. These statistics only include MZ and DZ twins and apply to the first aim of the current study: determination of intraclass correlations and estimation of heritability for each neuroimaging phenotype. Of the 126 participants, 64 are DZ and 62 are MZ, representing a total of 63 twin pairs. Twin pairs were scanned using the same scanner type. Delivery method, household income, and maternal age were significantly different between the cohorts used on the separate scanners (Table A.2). 43 Intraclass correlations Figure 2.1 shows the calculated intra-class correlations. Seven of the between- network pairs are located in quadrant one (Figure 2.a), indicating positive correlations between both MZ and DZ twins. None of the within network connectivity measures were located in quadrant one (Figure 2.2b). Quadrant two indicates that the ICC coefficient was 44 Table 2.2 Demographic and medical history variables for objective 1 participants. Continuous Variables Average SD Min Max Birth weight (g) 2475 441 1470 3650 Gestational age at birth (days) 254 12 224 273 Gestational age at MRI (days) 292 14 248 348 5 min APGAR score 8 0.75 4 10 Maternal education (years) 15 3. 6 24 Paternal education (years) 15 3 6 24 Maternal age (years) 29 5 16 42 Paternal age (years) 32 6 20 49 Residual framewise displacement 0.11 0.02 0.07 0.17 Categorical variables N % Sex Male 60 47% Female 66 52% Delivery Method Vaginal 40 31% C-section 86 68% Household income High 46 36% Mid 26 20% Low 54 42% Maternal ethnicity White 96 76% Black 28 22% Asian 4 3% Native American 0 0% Paternal Ethnicity White 82 65% Black 38 30% Asian 6 4% Native American 0 0% Maternal psychiatric history No 98 73% Yes 28 26% Paternal psychiatric history No 120 95% Yes 6 4% Scanner Allegra 100 79% TIM Trio 26 20% Maternal smoking No 122 96% Yes 4 3% NICU Stay No 124 98% Yes 2 1% 45 negative in MZ twins, but positive in DZ twins. Phenotypes falling in this quadrant include the FP, DA, DM, and Lim networks, and the DM-DA, Lim-DA, SC-VA, DA-Vis, VA-Vis, and FP-VA network pairs. Quadrant three displays coefficients that are negative both in the MZ and DZ twins for that phenotype. This includes the Vis and SC networks and SC- DA, DM-Lim, VA-SS, and VA-DA network pairs. Finally, quadrant four displays MZ coefficients that are positive while the DZ twins’ coefficients are negative. The networks and network pairs included in quadrant four are the SS and VA networks and Lim-VA, SC-Lim, DA-SS, FA-DA, DM-Vis, DM-SS, FP-Lim, DM-FP, SC-SS, SC-Vis, FP-Vis between-network pairs. We used the correlations as an exploratory analysis and chose to use mixed effects modeling for the heritability estimation because it is a more powerful model and allows estimation of heritability using all data from MZ and DZ twins together. Figure 2.2 Twin-twin correlations of connectivity. A) Between-network functional connectivity measures. B) Within network connectivity measures. Network numbers are as follows: 1: Vis, 2: SS, 3: DA, 4:VA, 5: Lim, 6: FP, 7:DM, 8: SC. 46 Additive mixed-effects modeling None of the phenotypes examined showed an h2 greater than 0.25. Six between- network pairs (SC-FP, SC-Vis) showed an h2 greater than 0.1 (Table A.3, Figures 2 and 3). The narrow-sense heritability estimation of motion as a heritable trait yielded a value of 0.208 (CI = (0,0.05); p-value = 0.076). Figure 2.3 Narrow-sense heritability values for all networks and network pairs. Between network pairs (left) and networks (right) display the narrow sense heritability with the 95% confidence interval included. 47 Figure 2.4 Between- and within-network narrow-sense heritability estimates for resting-state phenotypes. The narrow-sense heritability estimates are represented both by the size of the dot and the gradient of the color. Participants for Objective 2 The demographic and medical history statistical values can be found in Table 2.4. These statistics include MZ and DZ twins, singletons and unpaired twins and apply to the second aim of the current study: determination of demographic and medical history variables that may contribute to the observed phenotypes. Maternal and paternal age, household income, gestation number, and smoking status were all significantly different between scanner type (Table A.1). Correlations above 0.7 between the continuous predictor variables were seen in birth weight and gestational age, maternal and paternal education, and age, and then a correlation between education and age (Figure A.1). 48 Table 2.3 Demographics and medical history of participants for objective 2. Continuous Variables Average SD Min Max Birth weight (g) 2797 649 840 4820 Gestational age at birth (days) 261 17 210 295 Gestational age at MRI (days) 295 14 248 348 5 min APGAR score 9 0.809 4 10 Maternal education (years) 15 3 6 24 Paternal education (years) 15 3 6 24 Maternal age (years) 30 6 16 44 Paternal age (years) 32 7 18 64 Duration in NICU (days) 1 6 0 60 Residual framewise displacement 0.113 0.023 0.061 0.192 Categorical variables N % Sex Male 134 50% Female 134 50% Delivery Method Vaginal 120 44% C-section 148 55% Household income High 84 31% Mid 71 26% Low 113 42% Maternal ethnicity White 207 77% Black 55 20% Asian 3 1% Native American 3 1% Paternal Ethnicity White 184 68% Black 74 27% Asian 9 3% Native American 1 0% Maternal psychiatric history No 189 70% Yes 79 29% Paternal psychiatric history No 230 85% Yes 38 14% Scanner Allegra 226 84% TIM Trio 42 15% Maternal smoking No 244 91% Yes 24 8% Gestation Number Twin 125 46% Singleton 143 53% NICU Stay No 244 91% Yes 24 8% 49 Backwards elimination regression The covariates selected in the backwards elimination of each phenotype are shown in Table A.3. Overall, technical variables (type of scanner, motion), gestation number, gender, gestational age at birth, birth weight, gestational age at MRI, and presence of maternal psychiatric history were selected for 25% or more of the phenotypes examined. Mixed Linear Modeling The phenotypes for the between-network pairs included five statistically significant associations that passed Bonferroni correction (Table A.4). Gestational age at MRI was found to positively affect the SS-Vis network-pair (p = 0.0004; explains 5.20% of the variance) and gestational age at birth was an important negative predictor for both VA- DA (p = 0.0009, explains 10.40% of variance) and DM-FP (p = 0.0004, explains 4.90% of the variance) connectivity (Figure 2.4). Presence of a maternal psychiatric history was positively and significantly associated with FP-VA (p = 0.0012, explains 3.66% of the variance) connectivity. Finally, the scanner type used to collect the MRI data was found to be significantly and positively associated with the VA-Vis network pair (p = 0.0009, explains 3.46% of the variance). For the within-network phenotypes, two variables passed Bonferroni correction to yield statistically significant associations with connectivity in the DA network. Gestational number was positively associated (p = 0.00016) with and paternal education at enrollment was negatively associated (p = 0.00049) (Figure 2.5) with the DA network (Table A.4), explaining 8.47% and 10.80% of the variance respectively. 50 Figure 2.5 Scatterplots of each continuous variable with statistical significance in the mixed modeling. A) Gestational age at MRI plotted against the Z-score of the connectivity for the SS-Vis between-network pair. B) Gestational age at birth plotted against the Z-score of the connectivity for the VA-DA between-network pair. C) Gestational age at birth plotted against the Z-score of the connectivity for the DM-FP between-network pair. The plots are fitted with a regression line with a 95% confidence interval. 51 Figure 2.6 Scatterplot of paternal education against the Z-score of the connectivity values in the dorsal attention network. The regression line is plotted along with a 95% confidence interval. Power analysis The results of the power analysis indicate that this study is well powered (80%) to detect heritabilities of 0.5 and greater using our current number of twin pairs (Figure 2.6). We have moderate power to detect heritabilities between 0.25 and 0.45 and low power to detect heritabilities less than 0.25. 52 Figure 2.7 Power analysis using simulated data. The vertical line indicates the number of twin pairs contained in the current study. Discussion To our knowledge, this study is the first to examine both genetic and environmental influences acting upon resting-state functional connectivity in a large sample of typically developing neonates. Overall, our twin study suggests that genetic factors have relatively little influence on within and between network connectivity in early infancy, while our backward elimination regression and mixed-modelling analyses identified specific demographic and medical history variables that may influence infant functional connectivity measures, including gestational age at birth and maternal psychiatric history. 53 Our original hypothesis that measures of within-network and between-network connectivity would be heritable, with the strongest genetic effect evident for early maturing networks involved in perception and movement, was not well supported. Of the 36 connectivity phenotypes examined, only six showed narrow-sense heritability estimates greater than 0.10, though none were statistically significant. In contrast, heritability estimates in children for between-networks of the Lim-FP and SC-FP can exceed heritability of 0.50 and 0.60, respectively 359. To demonstrate significance of a heritability estimate in this range, over 2000 twin pairs would be needed based on our power analysis. Our conclusion that genetic factors play a relatively minor role in explaining individual variation in neonatal measures of within and between network connectivity is supported by results from the intraclass correlations. Our results suggest that the genetic architecture of within and between network connectivity is quite different at different life stages. One possible explanation for the observed patterns is the potential role of experience-dependent learning in shaping inter-individual differences in infancy. For example, the development of the frontal cortex is heavily dependent on early-life experiences, with socioeconomic stressors playing a large role in the progression of structure and function in the first year of life 360. Though our study may not provide strong evidence that resting-state connectivity phenotypes are heritable in early infancy, we can’t rule out that significant heritability could be detected in a larger study. Interestingly, our intraclass correlation analyses revealed many phenotypes which showed negative correlations between DZ twins. A few phenotypes showed negative correlations between MZ twins as well. Low or negative correlations between DZ twins have been observed for some other phenotypes, especially parent ratings of 54 temperament 361. In that particular case, low DZ resemblance is thought to arise from parents exaggerating the differences between DZ twins, a phenomenon known as the contrast effect. The current study did not involve parent ratings, but direct measures of physiology. However, it is possible that parents might treat members of a DZ pair differently in such a way that the development of functional connectivity is affected. Furthermore, both MZ and DZ twins may compete for restricted resources such as prenatal nutrition, postnatal nutrition, or parental attention, that may offset genetic factors. Through our backward elimination and mixed modeling approach, we were able to identify multiple demographic and obstetric history variables that contribute to between- and within-network connectivity phenotypes including gestational age at birth, gestational age at MRI, and presence of maternal psychiatric history. As regards gestational age at birth, our initial hypothesis that the somatosensory network would be impacted by gestational age at birth was partially supported. While gestational age at birth was not selected for within network connectivity of the somatosensory network, it was selected for 4 between network pairs involving somatosensory cortex including SS-Vis, DA-SS, DM- SS, and FP-SS. Gestational age at birth was also selected for eight other between network pairs and was especially strongly associated with connectivity between the default mode network and the frontoparietal, and ventral attention and dorsal attention networks, explaining about 10.40% and 4.90% of the individual variability, respectively. Interestingly, in most cases gestational age at birth was associated with decreased between-network connectivity. This was unexpected as the final trimester of pregnancy is generally characterized by greater integration between distant brain regions 205. Our finding is in contrast to a small-scale machine learning study of gestational age effects on 55 resting state connectivity which demonstrated that connections both within and between networks are typically reduced in preterm infants 362. However, research exploring brain volumes in neonates has reported increased brain volumes in earlier born children 337, indicating perhaps that these children undergo accelerated brain growth due to factors such as experiencing a richer sensory environment outside the womb, or as a protective adaptation resulting in faster brain growth to compensate for the physical delays in pre- term and early birth. Gestational age at MRI was selected as an important variable for 8 between- network pairs (SS-Vis, VA-Vis, Lim-VA, Lim-SS, FP-SS, DM-Lim, DM-FP, SC-SS) and for functional connectivity within the visual network. Functional connectivity within the visual network increased with age. Some between network pairs showed decreasing connectivity over age, while other showed increasing connectivity. The strongest relationship we observed with age was connectivity between the visual and somatosensory cortex, which increased over development. Age explained approximately 5.20% of the variance in this particular phenotype. These are both early maturing networks and clearly play important roles in how the very young infant interacts with and learns about the world. We note that the proportion of variance explained by age in our functional connectivity phenotypes is substantially less than we have observed for structural MRI phenotypes in this same age-range. For example, age explains 51% of the variation in intracranial volume during early infancy, 55% of the variation in surface area, and 18% of the variation in cortical thickness 337. As regards maternal psychiatric history, we originally hypothesized that it would preferentially affect connectivity phenotypes involving the limbic system. Our data did not 56 support this hypothesis, but maternal psychiatric history was selected as a potentially important predictor for 8 between-network phenotypes (DA-Vis, Lim-SS, Lim-VA, FP-Vis, FP-SS, FP-VA, DM-SS, SC-FP) and 1 within network phenotype (FP). Of these the FP- VA and FP-Vis network pairs were significantly associated with maternal psychiatric history in our mixed modeling analysis. The frontoparietal network and its resting-state connectivity are heavily associated with executive functioning later in life. Reineberg et al. 66 demonstrated that individual differences in an overall and general executive functioning measure are based on variation in the frontoparietal network. Individual variation in set-shifting, a subcomponent of executive function, was significantly associated with the ventral attention network. The ventral attention network was additionally shown to have great influence on set-shifting in a later work by Reineberg et al. 305. We note that most of the women with a positive psychiatric history in our study had been diagnosed with a mood and/or anxiety disorder. Consequently, our results are relevant to a growing body of literature addressing potential relationships between maternal depression and anxiety and offspring executive function. Interestingly, multiple epidemiological studies have observed associations between maternal depression and offspring risk for ADHD 363–368. These relationships could arise due to shared genetic vulnerabilities for depression and ADHD, but could also reflect the impact of in utero exposure to physiological features associated with depression (high levels of glucocorticoids, pro-inflammatory cytokines, and altered function of serotonin systems). The current study suggests that functional connectivity between the frontoparietal and 57 other networks may play a role in the mechanistic pathways that link maternal depression to attention problems in offspring. Focusing specifically on the within-network resting-state connectivity phenotypes, paternal education at enrollment was selected as a statistically significant contributor to the dorsal attention network in our second objective. Interestingly, the relationship is negative, suggesting that higher paternal education is associated with delayed development of the dorsal attention network. This may seem surprising given previous studies showing that higher parental education is associated with better cognitive function 369–371 and reduced risk for ADHD 372. No study has specifically examined whether a faster or slower maturational rate of this network in infancy is associated with later cognitive outcomes or risk for attention problems, although a study by 373 demonstrated that increased connectivity between specific components of the dorsal network correlates with sustained attention in females between 4 and 7 years of age. We note that prior research conducted by our group found a negative association between paternal education and cortical thickness in the frontal lobe during early infancy 334. This may be relevant to the current findings, as the dorsal attention network incorporates frontal regions 374. Gestation number was also selected for this network. It has been documented that a majority of multiples are born with low birth weight, which is a major predictor of hyperactivity and attention problems in later years, including the development of attention deficit hyperactivity disorder 375–378. However, in our backward selection analysis, gestation number was selected as an important predictor of within network connectivity in the dorsal attention network, while birthweight itself was not. This suggests that the 58 association between gestation number and connectivity within this network is not simply a function of low birthweight. Alternative explanations include obstetric complications and nutrition competition between twins. Intrauterine competition is present in the gestation of multiples. Examples include unequal placental sharing, chorionic circulatory imbalance, and the physiological challenges of providing resources to nurture more than one fetus 379,380. Technical variables such as scanner type and subject motion were also frequently selected in our backwards elimination regression analyses, although the only relationship meeting our criteria for significance in the mixed models was between scanner type and connectivity between the ventral attention and visual networks. Independent groups have demonstrated the impact of motion artifacts on fMRI 344,381,382. We addressed this known challenge in several ways including: (1) regressing out 6 motion parameters during preprocessing, (2) implementing the widely accepted data scrubbing approach with scrubbing criteria at 0.5% signal change and 0.5mm framewise displacement, and (3) including a measure of subject motion in our analyses. We note that motion shows a strong genetic component in both children and adults 383,384. In adults, three measures of head motion during the fMRI (mean translation, maximum translation, and mean rotation) have heritabilities ranging from 37 to 51% 383. Our work adds to that literature by showing that motion during neonatal scans has low-to-moderate heritability. The current study had many strengths including the use of a prospective cohort design, state-of-the-art methods for the acquisition and analysis of infant imaging data, and collection of rich metadata. One potential limitation is the use of sleeping subjects. Functional connectivity in sleeping infants more closely resembles adult slow wave sleep 59 rather than the patterns of awake adults 385, and connectivity measures can be influenced by stage of the sleep cycle 386. We were not able to track when in the sleep cycle our infant subjects were scanned, and this could have introduced noise and reduced the power of our analyses. This may be one reason we did not see strong positive correlations between our MZ and DZ twins. However, use of sleeping subjects is widely accepted for infant imaging studies as it minimizes subject motion, as compared to awake scanning 208,385,387–389, and does not require anesthesia, which at such a young age causes aberrant brain development 390. In addition, some of the variables examined in Objective 2 should not be conceptualized as exclusively environmental. For example, gestational age at birth is influenced by both fetal and maternal genetic factors in addition to environmental factors 391. In a study by Wu and colleagues, the broad-sense heritability for gestational age at birth was estimated at 24.45%, with 60.33% attributed to environmental influences such as maternal stress, exposure to tobacco, and possible infections 392. Birthweight is also a heritable trait heavily influenced by the environment. Heritability of weight increases during the second and third trimesters, culminating in a heritability estimate of 26% for singletons and 29% for twins at birth 393. Furthermore, one must consider the distinction between objective and effective environments. Objective environments refer to environments as measured by the researcher, while effective environments are defined by the outcomes they produce. It is possible for an objectively shared event (e.g. gestational age at birth) to have effectively non-shared, or child-specific, effects (e.g. one child in a pair might remain in the NICU longer after being born early or suffer more severe neurodevelopmental effects) 394. Finally, we acknowledge that classic twin studies rely on 60 certain fundamental assumptions including (1) findings from twins are generalizable to the rest of the population, (2) in utero environment is identical with no chorion or amnion differences, (3) environment affects MZ and DZ twins equally, and (4) there are no gene by environmental correlations or interactions. An alternative approach that could be used would be to define functional networks based on neonatal data as opposed to the adult map. Argument for either method is depended on the question one seeks to answer. The present study aimed to examine how canonical functional networks are impacted by genetic and environmental factors so adult-defined networks were used to facilitate interpretation of the findings. Our utilization of the adult networks allows us to directly compare to adult resting-state networks and make comparisons to the literature. In conclusion, our twin study suggests that genetic factors do not play a major role in explaining individual variation in resting-state connectivity measures in early infancy, though the low power of the study does not let us directly conclude a failure to reject the null hypothesis. Future studies will be needed with larger participant numbers to corroborate our findings. However, specific demographic and medical history variables were identified that may influence functional connectivity measures. Future research could map the development of heritability in a longitudinal study, observing the maturation of the connections into adolescence where connectivity has been demonstrated to be mostly under genetic control. 61 Chapter 3 Anatomical and functional connectivity differences in the brains of infants with TS when compared to TD children 62 Abstract Turner syndrome (TS), a condition caused by complete or partial loss of an X- chromosome, is often accompanied by deficits in specific cognitive domains. Magnetic resonance imaging studies of adults and children with TS suggest these deficits may reflect differences in anatomical and functional connectivity. However, no imaging studies have explored connectivity in infants with TS. Consequently, it is unclear at what point in development connectivity differences emerge. To address this, we compared functional and structural connectivity of one-year-old infants with TS to typically developing one- year-old boys and girls. We examined functional connectivity between the right precentral gyrus and five regions that show reduced volume in 1-year old infants with TS compared to either male or female controls and found no differences. To assess anatomical connectivity, we examined diffusivity indices along the superior longitudinal fasciculus (SLF). An exploratory analysis of 54 additional white matter tracts was also performed. TS and control groups did not differ along the SLF. However, exploratory analyses revealed significant group differences in nine tracts. The results suggest the first year of life is a window in which interventions might prevent connectivity differences observed at later ages, and by extension, some of the cognitive challenges associated with TS. 63 Introduction Described by Henry Turner in 1938 246, Turner syndrome (TS) is caused by the partial or complete loss of an X-chromosome. The condition occurs in approximately 1 in 2000 live female births 247, making it one of the most common aneuploidies. TS represents a unique population for studying X chromosome effects on human development 268,269,395, because females with TS are hemizygous for many genes in the pseudoautosomal regions (PAR) of the X chromosome, when compared to both XX females and XY males 396. For the 15% of genes outside the PAR that escape X-inactivation 397, females with TS have reduced gene dosage compared to XX females, but are similar to XY males. The loss of the second sex chromosome produces multi-systemic effects. TS is often accompanied by gonadal dysgenesis, congenital heart defects, renal abnormalities, and liver disorders 398, as well as a unique neurocognitive profile. Deficits in social cognition (SC) 399–401, executive functioning (EF) 270,402,403, and visuospatial reasoning (VR) 404,405 are often present. Furthermore, individuals with TS appear to be at increased risk for male-biased neurodevelopmental disorders including Autism Spectrum Disorders (ASDs) 252,253 and Attention Deficit Hyperactivity Disorder (ADHD) 268. Structural magnetic resonance imaging (MRI) has been used to identify neuroanatomical features of TS that may explain the unique cognitive profile and increased risk for male-biased neurodevelopmental disorders. To date, most neuroimaging studies of TS have been performed on adults and adolescents, who show structural changes consistent with the observed cognitive challenges: for example, decreased grey matter volume in parieto-occipital regions implicated in visuospatial reasoning 257–261. Consistently, white matter volume increases in TS have been observed 64 in the temporal lobe, which is implicated in language and social cognition 257–259. The first quantitative neuroimaging study of infants with TS observed many structural features consistent with those present in adolescents and adults with TS, suggesting that many of the neuroanatomical phenotypes in TS are established early and persist into adulthood 272. High-level cognitive processes disrupted in TS, such as executive function and social cognition, require coordination between structurally segregated brain regions. Thus, studies of anatomical and functional brain connectivity, using diffusion tensor imaging (DTI) and resting-state functional magnetic resonance imaging, provide additional insights into the neurological basis of the TS cognitive profile. While DTI allows researchers to examine the integrity of axonal pathways connecting different brain regions, functional connectivity analyses of rs-fMRI imaging data provide insight into the organization of large-scale brain networks supporting these processes. In older children, adolescents, and adults, TS has a distinct structural connectivity profile which is well supported by the literature. Multiple studies have reported reduced fractional anisotropy (FA) in the superior longitudinal fasciculus (SLF) with others reporting more global reductions 258,264,265. The SLF facilitates working memory, language, visuospatial attention, and numerical tasks 76,266,267, all domains affected in TS individuals. The age at which these differences in white matter integrity emerge is currently unknown. Reduced functional connectivity at rest has been reported in the frontoparietal and dorsal attention networks in girls with TS, which may explain the increased prevalence of ADHD in the TS population 268. Whole-brain reduction in functional connectivity strength 65 was identified in the postcentral gyrus/intraparietal sulcus, angular gyrus, and cuneus and the right cerebellum in girls with TS 269. This study also showed a relationship between these connectivity deficits and cognitive domains such as working memory and visuospatial reasoning. Using task-based fMRI, Bray and colleagues 271 were the first to demonstrate abnormal connectivity in the parieto-occipital and parieto-temporal pathways, which could explain deficits in visuospatial processing. As previously stated, most neuroimaging studies of TS have been performed on adults and adolescents, leaving the question of when in development these observed differences in structural and functional connectivity arise unanswered. Consequently, it is unclear whether observed differences are a direct consequence of hemizygosity of X- chromosome genes during the prenatal and early postnatal period, when axonal pathways and functional networks are first established, or occur later. If these difference do arise later in childhood, they could reflect common postnatal clinical experiences such as treatment with growth hormone and estrogen replacement therapy, which are commonly used to increase adult height and initiate puberty, respectively 290,406. Individuals with TS may also take medications for the heart and renal problems previously mentioned, undergo surgery with general anesthesia, or take psychotropics 291 with potential consequences for brain development. Finally, postnatal deficiency in gonadal steroid hormones could also be involved 407–409. The current study focuses on one-year-old infants that have not yet been exposed to growth hormone and estrogen replacement therapy. Our main objective was to determine whether anatomical and functional connections between frontal and parietal/occipital cortices are altered in one-year-old infants with TS. We focus on brain 66 regions that showed volume differences in a prior study of the same cohort 272. We addressed this objective through a cross-sectional case-control study of TS and typically developing (TD) infants with DTI and rs-fMRI scanning data. Based on the literature summarized above, we hypothesized that infants with TS would have aberrant diffusivity in the superior longitudinal fasciculus (SLF) and reduced functional connectivity between the precentral gyrus and regions involved in basic visual processing (calcarine cortex), social cognition (supramarginal gyrus and lingual cortex), and executive function (supramarginal gyrus). To our knowledge, this study is the first of its kind to utilize rs-fMRI and DTI together in TS to answer questions of brain structure and function at this developmental stage. Our results have the potential to show whether differences in structural and functional connectivity observed in older children and adults with TS are present in infancy, versus emerging at a later developmental stage. This has implications for early intervention strategies and therapies for TS that could help ameliorate neurocognitive deficits in the disorder. Methods Participants Imaging data from 26 females with X monosomy (6 mosaic, and 20 with complete X monosomy), 39 typically developing males, and 47 typically developing females were used in this study. All participants were approximately one year of age and had resting- state connectivity data. Usable DTI scans were available for 24 females with X monosomy, 31 TD males, and 36 TD females. TS participants were recruited through the University of North Carolina’s (UNC) Pediatric Endocrinology Department, UNC’s Turner Syndrome Clinic, national support groups, and heath care providers from across the 67 United States. Typically developing subjects represent a subset of the Early Brain Development Study (EBDS) cohort at the University of North Carolina at Chapel Hill 337– 339. Exclusion criteria for both groups included substance abuse or major health problems in the mother during pregnancy, major psychiatric illness in either parent, extreme prematurity of the child, and any congenital abnormality in the subject not associated with TS. rs-fMRI and DTI Acquisition and Processing Imaging data were acquired over a ten-year period on either a Siemens Allegra head-only or a Siemens Tim Trio scanner (Siemens Medical System, Inc., Erlangen, Germany) which replaced the Allegra part way through the study. There were 32 infants imaged on the TIM Trio and 81 on the Siemens Allegra. All infants were in a natural sleep state during imaging. Functional imaging was performed using a T2-weighted EPI sequence: TR=2 s, TE = 32 ms, 33 slices, and 4mm isotropic resolution, while structural images were acquired using three-dimensional magnetization-prepared rapid acquisition with gradient echo (MPRAGE) sequence, which is a T1-weighted imaging technique. The sequence is as follows: TR = 1820 ms, TE 4.38 ms, and 1 mm isotropic resolution. These methods are described in 353. A 6‐direction protocol with the following parameters was used to collect DTI data on the Allegra during the initial years of the study (81 infants total): Repetition Time (TR)/ Echo Time (TE) = 5,200/73 ms, slice thickness = 2 mm, and in‐plane resolution = 2 × 2 mm2, with a total of 45 slices in 6 unique directions using b value of 1,000 s/mm2 and 1 baseline image (b value = 0) per sequence. In total, 35 DWIs were generated per subject 68 by repeating the sequence five times to improve the signal-to-noise ratio. Later imaging on the Allegra used 42 directions of diffusion sensitization with a b value of 1,000 s/mm2 in addition to seven baseline images, which generated a total of 49 DWIs. The parameters for the 42‐direction data were as follows: TR/TE/Flip angle = 7,680/82/90°, slice thickness = 2 mm, and in‐plane resolution = 2 × 2 mm2, with a total of 60–72 slices. The remainder of subjects were scanned on the Tim Trio following the same parameters as the 42-direction Allegra protocol. These methods are described in 410. Image analysis (rs-fMRI) Functional data were preprocessed using the FMRIB (for Functional MRI of the Brain) Software Library (FSL; version 6.0; 341) and Analysis of Functional NeuroImages (AFNI; 342). Steps included discarding the first 3 volumes (6s), slice-timing correction, rigid-body motion correction, bandpass filtering (0.01–0.08 Hz), and regression of white matter, CSF, the 6 motion parameters and their derivative, quadratic terms. The nuisance signals were band-pass filtered before regression to match the frequency of the blood oxygen level-dependent signal. A single global signal regression step was performed after the scrubbing to eliminate global signal re-introduced by scrubbing. Data scrubbing was implemented with scrubbing criteria at global signal changes >0.5% and framewise displacement >0.3mm 343 with continuous datapoints no less than 3. Participants with less than three minutes (90 datapoints) of functional data after scrubbing were excluded. Then, to make the same length of data, all infants’ functional data were truncated into 90 volumes. Finally, the images were spatially smoothed with Gaussian kernel (full width at half maximum = 6mm). 69 After performing a rigid-body registration between the functional data and T1- weighted structural images of the same subject, a nonlinear registration (fNIRT in FLS) was done between individual T1-weighted images and infant age specific AAL template images 352. The combined transformation field was used to warp the preprocessed rs- fMRI functional images to the neonate template 353. The structural image skull-stripping was done using a template-based method. After performing a rigid-body registration between the functional data and T1-weighted structural images of the same subject, a nonlinear registration (fNIRT in FLS) was done between individual T1-weighted images and an infant-specific AAL template 352 which includes 90 regions of interest. The combined transformation field was used to warp the preprocessed rs-fMRI functional images to the group template 353. For the current analysis, we extracted the average BOLD time course from the right precentral gyrus, right and left calcarine cortex, right and left lingual cortex, and right supramarginal cortex (regions which showed reduced volumes in infants with TS). Correlations between the right precentral gyrus and the other five regions were then calculated for each subject. The correlations were then Fisher-Z transformed for statistical analysis. Statistical Analyses (rs-fMRI) All statistical analyses were performed using R statistical software 4.0.5 411 using base functions. 70 Two-sided Fisher’s exact tests were used to evaluate group differences in categorical variables. A two-sided Kruskal-Wallis H-test was run on each of the continuous variables with a Dunn test used for post hoc comparisons. A one-way ANCOVA was used to test differences in functional connectivity between the three groups (TS females, XX females, XY males) with post-hoc FDR- corrected pairwise comparisons. Five outcome variables were examined: (1) connectivity between right precentral gyrus and right calcarine cortex, (2) connectivity between right precentral gyrus and left calcarine cortex, (3) connectivity between right precentral gyrus and right lingual cortex, (4) connectivity between right precentral gyrus and left lingual cortex, and (5) connectivity between right precentral gyrus and right supramarginal cortex. To reduce the variance, covariates were chosen that are variables previously associated with imaging outcomes. In this case we used the scanner type, birth weight, maternal and paternal education, and gestational age at MRI. Image Analysis (DTI) All image analysis steps for DTI are described in Girault et al. 410 and represent a neonate-specific pipeline adapted from the general UNC-Utah NA-MIC DTI pipeline 412. Quality control (QC) is an automated protocol using DTIPrep 413. DTIPrep detects slice- wise and gradient-wise intensities and corrects for motion artifacts and eddy current effects 413. Images with large motion artifacts and aberrant gradients are excluded. Weighted least squares fitting was used to estimate the diffusion tensors. Additional expert-guided QC was performed using 3D Slicer. Skull stripping was performed as specified in Verde et al. 412. A DTI atlas derived from one-year-olds was used to map the images. Successful registration was confirmed by visually comparing the warped DTI 71 images and the atlas. Atlas fibers were then mapped onto the individual subject space, and DTIAtlasFiberAnalyzer was used to extract the fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD) and axial diffusivity (AD) profiles. A correlation analysis was then made for all individuals of the plot of the atlas against the four metrics. Profiles with a correlation value above 0.7 were considered well mapped to the atlas space; profiles that did not meet this criterion were excluded. Statistical Analyses (DTI) As described in Jha et al. 414, FADTTS, or functional analysis of diffusion tensor tract statistics 415, was used to test group differences in FA, AD, and RD along the SLF white matter tract. Mean diffusivity (MD) was not included as it is an overall, general measure of diffusivity as a whole and based on AD and RD. FADTTS gives a global test statistic, and local test statistics along the white matter fiber tracts. We included DTI protocol, gender, ICV, postnatal age at MRI, gestational age at birth, and two motion measures as covariates. The first motion measure was calculated by taking the sum of the slices and volumes excluded due to artifacts along with the number of excluded volumes due to residual motion. The second measurement is the number of volumes exceeding the threshold of the rotational angle (angle > 1 degree) and translational volume (translation > 2mm). Within each tract, local p-values are corrected for multiple comparisons with FDR. The local p-values are then merged with the test statistics onto the corresponding fiber locations for visualization. 72 Exploratory Analyses (DTI) The processes for DTI image and statistical analyses are the same as those described for the SLF. A total of 54 tracts were generated which include, bilaterally, the arcuate fasciculus (frontoparietal, frontotemporal, and temporoparietal segments), anterior portion of the cingulum, posterior cingulum adjoining the hippocampus, corpus callosum (body, genu, motor, parietal, premotor, rostrum, splenium, and tapetum segments), corticofugal (motor, parietal, prefrontal, premotor segments), corticoreticular, corticospinal, corticothalamic (motor, parietal, prefrontal, premotor, superior segments), fornix, fronto-occipital fasciculus, inferior longitudinal fasciculus, optic tract, optic radiation, superior longitudinal fasciculus, and uncinate fasciculus. However, only 46 tracts passed quality control and had an n greater than or equal to 20 for each group (Table B.1). Tracts excluded from the statistical analyses include: the right temporoparietal portion of the arcuate fasciculus, left corticospinal, left optic tract, right superior longitudinal fasciculus, right premotor and prefrontal corticofugal tracts, and left premotor and prefrontal corticofugal tracts. Results Participants Descriptive statistics for the demographic and medical history variables of the participants in this study are in Table 3.1. Significant differences were found between the three groups in gestational age at birth, birthweight, scanner type, and maternal ethnicity. On average, children with TS were born earlier than control participants and had the 73 lowest birthweight. The children with TS were more likely to be scanned on the Tim Trio than the Allegra and were more likely to be White. Connectivity Measures No statistically significant relationships were found after FDR correction (Table 3.2). However, the following connectivity pairs had uncorrected p values below 0.05 and FDR correct p-values below 0.10: right precentral gyrus to right calcarine cortex and right precentral gyrus to right lingual cortex. We proceeded with pairwise comparisons for these two phenotypes, and found significant differences in the XX-X0 group for both the right calcarine cortex (p=0.029) after FDR correction and right lingual cortex (p=0.0187) (Table 3.3). 74 Table 3.1 Demographics and medical history for one-year-old infants with Turner syndrome and their typically developing male and female counterparts. Female Male Variable TS control Control p Mean Mean N (SD) N (SD) N Mean (SD) range range range Gestational 268 age at birth (11) 277 (9) 47 (days) 246- 259-295 274 (10) 241- 26 286 39 289 0.006 Birth weight 26 2803 3380 (grams) (421) (428) 47 2155- 2340- 3386 (469) 3925 4414 39 2375-4562 <.0001 Age at MRI 26 389 (days) (16) 382 (22) 47 359- 339-439 382 (18) 343- 419 39 422 0.357 Maternal 25 29 (6) 30 (5) 47 age (years) 21-42 20-40 39 30 (4) 20-41 0.531 Paternal 25 32 (7) 32 (6) 47 age (years) 23-50 21-50 37 32 (4) 22-39 0.842 Maternal 25 15 (3) education 47 16 (3) 9- 6-20 (years) 23 39 16 (3) 10-22 0.235 Paternal 24 14 (3) 16 (3) 8- education 3-23 22 (years) 46 39 16 (3) 9-22 0.093 Total 24 76094.5 68280 household (69420) (46264) 73562 income 0- 0- (47217) (dollars) 330000 45 205000 37 0-195000 0.913 N Percent N Percent N Percent Maternal ethnicity 0.0181 White 23 92.0% 32 68.1% 35 89.7% Black 2 8.0% 13 27.7% 2 5.1% Asian 0 0.0% 2 4.3% 2 5.1% Paternal ethnicity 0.056 White 23 92.0% 32 68.1% 33 86.8% Black 2 8.0% 13 27.7% 3 7.9% Asian 0 0.0% 2 4.3% 2 5.3% 75 Table 3.1 (cont’d) Smoking 0.843 Yes 1 3.8% 3 6.4% 1 2.6% No 25 96.2% 44 93.6% 38 97.4% Scanner 0.0185 Trio 13 50.0% 11 23.4% 7 17.9% Allegra 13 50.0% 36 76.6% 32 82.1% Allegra directionality A06 5 45% 9 35% 9 35% 0.834 A42 6 55% 17 65% 17 65% Table 3.2 Pairwise comparison of the resting-state functional connectivity between the right precentral gyrus and the five listed regions. Region LS (SE) LS (SE) LS (SE) Uncorrected FDR TS Female Male p-value Corrected p-value Right calcarine -0.358 -0.018 -0.151 0.0397 0.099 cortex (0.170) (0.15) (0.153) Left calcarine -0.312 0.003 -0.123 0.59 0.59 cortex (0.170) (0.15) (0.153) Right lingual -0.3920 -0.067 -0.184 0.025 0.099 cortex (0.152) (0.134) (0.137) Left lingual -0.268 -0.019 -0.133 0.099 0.165 cortex (0.150) (0.132) (0.135) Right 0.25 0.095 0.033 0.299 0.37 supramarginal (0.188) (0.166) (0.169) gyrus Table 3.3 Pairwise comparison of the resting-state functional connectivity between the right precentral gyrus and the right calcarine and lingual cortices. Region Comparison Estimate SE p-value Right calcarine cortex XX-XY 0.133 0.105 0.421 XX-X0 0.340 0.131 0.029 XY-X0 0.208 0.126 0.231 Right lingual cortex XX-XY 0.0118 0.094 0.425 XX-X0 0.325 0.117 0.0187 XY-X0 0.207 0.112 0.0163 76 DTI Measures In our primary analysis, which focused on the superior longitudinal fasciculus, we did not observe statistically significant differences between individuals with TS and male or female controls (Table 3.4). The right superior longitudinal fasciculus was left out of the results due to the previously described cutoff for subject numbers after quality control. 77 Table 3.4 Paired comparisons between the three groups for structural connectivity of the left superior longitudinal fasciculus. Axial diffusivity (AD), radial diffusivity (RD), and fractional anisotropy (FA) are represented in the post hoc analysis and show no statistical significance. FDR-corrected p- values Superior Longitudin FDR- al Uncorrecte correcte Fasciculus, d global p- d global Left value p-value AD RD FA Female control, Male 0.65 control 0.099 0.171 0.068 0.31 4 Female control, 0.29 Turner syndrome 0.585 0.621 0.427 0.515 6 Male control, Turner 0.20 syndrome 0.092 0.165 0.212 0.077 5 Exploratory Analyses Of the white matter tracts examined, we found significant differences between TS participants and controls in nine white matter tracts (Table 3.5): the frontotemporal segment of the right arcuate fasciculus, the right posterior cingulum adjoining the hippocampus, the motor and tapetum segments of the corpus collosum, the left motor segment of the corticofugal tract, the premotor segment of the right corticothalamic tract, the left inferior fronto-occipital fasciculus, the left inferior longitudinal fasciculus, and the right optic tract. AD was statistically significant in the right frontotemporal segment of the arcuate fasciculus (p=0.009 female control and Turner syndrome), the right posterior cingulum 78 adjoining the hippocampus (p=0.004 female control and male control, p=0.001 female control and Turner syndrome), the motor segment of the corpus callosum (p=0.024, female control and Turner syndrome), the tapetum segment of the corpus callosum (p=0.042, female control and Turner syndrome), and the left motor segment of the corticofugal tract (p=<0.0001 in both female control and male control, and female control and Turner syndrome, p=0.004 for male control and TS). Additionally, the premotor portion of the right corticothalamic tract (p<0.0001 female control and male control, and p=0.005 for female control and Turner syndrome), the left inferior fronto-occipital fasciculus (p=0.003 male control and Turner syndrome) all present with statistically significant FDR-corrected p values. RD showed statistical significance in the right posterior cingulum adjoining the hippocampus (p=0.002 female control and Turner syndrome, and p<0.0001 female and male controls), the left motor section of the corticofugal tract (p=0.024 female and male control; p<0.0001 female control and Turner syndrome), and the left inferior longitudinal fasciculus (p=0.044 female control and Turner syndrome). Finally, FA was statistically significant in the tapetum of the corpus collosum (p=0.011, female control and Turner syndrome), the left motor segment of the corticofugal tract ( p=0.021, female control and Turner syndrome; p<0.0001, male control and Turner syndrome), the left inferior fronto-occipital fasciculus (p=0.046, female control and Turner syndrome; 0.012 male control and Turner syndrome), the left inferior longitudinal fasciculus (p=0.039, female control and Turner syndrome), and the right optic tract (p=0.007, male control and Turner syndrome). 79 Table 3.5 Fasciculi with statistically significant FDR-corrected global p-values and post hoc FDR-correct p-values for individual diffusivity metrics. AD indicates axial diffusivity, RD indicates radial diffusivity, and FA indicates fractional anisotropy. Post hoc FDR-corrected p-values Global FDR- Globa correct l p- ed p- value value AD RD FA Arcuate fasciculus, right, frontotemporal Female control, 0.026 0.076 Male control 0.062 0.013 0.255 Female control, 0.007 0.033 Turner syndrome 0.009 0.224 0.078 Male control, 0.387 0.455 Turner syndrome 0.329 0.725 0.319 Cingulum adjoining the hippocampus, right Female control, <0.00 <0.000 <0.00 Male control 01 92 0.004 01 0.526 Female control, <0.00 <0.000 Turner syndrome 01 92 0.001 0.002 0.197 Male control, Turner syndrome 0.466 0.522 0.608 0.87 0.899 Corpus callosum, motor Female control, Male control 0.013 0.054 0.099 0.863 0.641 Female control, Turner syndrome 0.007 0.033 0.027 0.426 0.261 Male control, Turner syndrome 0.607 0.644 0.419 0.584 0.477 Corpus callosum, tapetum Female control, 0.104 Male control 0.312 0.403 0.84 0.941 Female control, Turner syndrome 0.004 0.022 0.042 0.241 0.011 Male control, Turner syndrome 0.051 0.1104 0.178 0.505 0.09 80 Table 3.5 (cont’d) Corticofugal, left, motor Female control, <0.00 <0.000 <0.00 <0.000 0.654 Male control 01 92 01 1 Female control, <0.00 <0.000 <0.00 <0.000 0.021 Turner syndrome 01 92 01 1 <0.00 <0.000 0.004 0.087 01 92 Male control, <0.00 Turner 01 syndrome Corticothalamic, right, premotor Female control, <0.00 <0.000 <0.00 0.008 Male control 01 92 01 0.338 Female control, 0.003 0.018 Turner syndrome 0.005 0.311 0.129 Male control, 0.452 0.515 Turner syndrome 0.319 0.574 0.626 Inferior fronto- occipital fasciculus, left Female control, 0.029 0.080 0.077 Male control 0.075 0.378 Female control, 0.032 0.083 Turner syndrome 0.088 0.14 0.046 Male control, 0.001 0.007 Turner syndrome 0.003 0.066 0.012 Inferior longitudinal fasciculus, left Female control, 0.017 0.065 Male control 0.045 0.801 0.313 Female control, 0.001 0.007 Turner syndrome 0.201 0.044 0.039 Male control, 0.028 0.078 Turner syndrome 0.339 0.247 0.073 Optic tract, right Female control, 0.024 0.073 Male control 0.098 0.084 0.018 Female control, 0.035 0.087 Turner syndrome 0.406 0.211 0.388 Male control, 0.008 0.035 Turner syndrome 0.145 0.06 0.007 81 We classified group differences into six categories based on both global and local test statistics: XX>X0=XY, X0=XY>XX, X0>XY=XX, XX=XY>X0, X0=XX>XY, and XY>X0=XX. We refer to both XX>X0=XY and X0=XY>XX as “masculinization” patterns because the TS females are similar to TD males but differ from TD females. We refer to both X0>XY=XX and XX=XY>X0 as “PAR” patterns because the suggest a role for gene in the pseudoautosomal regions of the sex chromosomes. Finally, we refer to both X0=XX>XY and XY>X0=X as “sex difference” patterns because phenotypic females (TS and TD) differ from TD males. The primary pattern for each diffusivity measure and each tract that showed significant group difference in post-hoc analyses is shown in Table 3.6. Figure 3.1 illustrates how these three different patterns manifest in the right premotor portion of the corticothalamic tract, the left inferior longitudinal fasciculus, and the left inferior fronto-occipital fasciculus. All other fasciculi can be seen in Figures B.1-B.12. 82 0.0 19 - 0.01 0.01 9 - 0.01 0.01 9 - 0.0 Figure 3.1 Models of DTI results for the right premotor portion of the corticothalamic tract (top), the left inferior longitudinal fasciculus (middle), and left inferior fronto-occipital fasciculus (bottom). In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. Here, the corticothalamic tract is an example of a masculinization, the inferior longitudinal fasciculus an example of the effects from the PAR, and the anterior portion of the inferior fronto-occipital fasciculus an example of a sex difference. 83 Table 3.6 Chromosomal hierarchies for each statistically significant tract shown for axial diffusivity, radial diffusivity, and fractional anisotropy. XX indicates typically developing female, X0 indicates TS female, and XY indicates typically developing male. Classifications in this table represent the dominant pattern we observed. However, examination of local tract statistics often revealed regional complexities, which we describe in more detail in the discussion section. Fasciculus Axial Diffusivity Radial Diffusivity Fractional Anisotropy Arcuate, right, XX> X0 = XY frontotemporal Cingulum adjoining XX > X0 = XY XX > X0 = XY hippocampus, right Corpus callosum, X0 = XY > XX motor Corpus callosum, XX = XY > X0 XX = XY > X0 tapetum Corticofugal, left, XX > X0 = XY XX > X0 = XY X0 > XY = XX motor Corticothalamic, XX > X0 = XY right, premotor Inferior fronto- XX= X0 > XY X0 = XX > XY occipital fasciculus, left Inferior longitudinal XX = XY > X0 X0 > XX = XY fasciculus, left Optic tract, right X0 = XY > XX Discussion The current work is, to our knowledge, the first to use rs-fMRI and DTI data to explore brain connectivity in Turner syndrome in infancy. Overall, our study suggests a unique developmental pattern in Turner syndrome in which structural and functional connectivity differences arise at varying times in the life span. Our primary analyses indicate that diffusivity differences in the SLF are not present at one year of age and presumably emerge during early childhood, as they have been reported in older children and adults. Similarly, differences in functional connectivity between frontal and parietal/occipital regions were not detected in the current study but are evident later in 84 life. Our exploratory DTI analyses revealed nine fasciculi that differed between TD and TS infants. Our original hypothesis that infants with TS would have reduced functional connectivity between the precentral gyrus and our five selected regions came from previous work showing volumetric differences in the brains of the same cohort of individuals 272. The five selected regions also contribute to cognitive functions that are often disrupted in TS, supporting visual processing (left and right calcarine cortex) executive functioning (supramarginal gyrus) and social cognition (left and right lingual cortex). It is intriguing that there are such pronounced volumetric differences in this cohort of infants, but no differences were observed in functional connectivity, especially as disrupted connectivity between frontal and parietal regions has been reported in older individuals with TS74,416–418 . For structural connectivity we hypothesized there would be diffusivity differences between TS infants and TD infants in the SLF, a tract that connects frontal and parietal brain regions and is involved in executive functioning, attention, and working memory 74,416–419. This hypothesis was based on the extant literature in adults 258,264,265 and prepubescent girls with TS 265. However, we did not observe significant differences between groups in the current study. Integrating our results with the existing literature suggests the SLF develops normally in TS until at least 1 year of age but deviates during the toddler and preschool years. Null findings for fronto-parietal connectivity, assessed with both DTI and rs-fMRI, parallel findings on early cognitive development in TS. Infants with TS show a relatively normal cognitive profile at 12 months of age with a potentially slower rate of progression 85 in visual reception and fine motor skills from 12 to 24 months420. If differences in fronto- parietal connectivity arise after 12 months of age, there may be an opportunity to prevent those changes through early intervention, thereby preventing or ameliorating some of the cognitive impairments observed in school-aged children with TS. Historically, diagnosis of TS was made when a female failed to go through puberty or fell two or more standard deviations below the mean height for their age group, resulting in relatively late diagnosis and reduced intervention opportunities 421,422. However, characteristics such as lymphedema in infancy, presence of cardiac abnormalities which occur in approximately 50% of the TS population 421,423,424, and the advancement of prenatal diagnostics such as genetic amniocentesis, sonograms, and circulating cell-free DNA have contributed to earlier diagnoses. The most recent clinical practice guidelines for TS encourage annual developmental and behavioral screenings for TS and recommend academic adjustments to accommodate learning and performance issues 398. They also suggest that evidence- based interventions for cognitive or psychosocial problems in other populations be adapted to meet the needs of girls/women with TS. The possible benefits of administering early interventions in a proactive manner, before cognitive deficits emerge, is not addressed. Early interventions clearly improve cognitive and psychosocial outcomes for individuals with ADHD 425–427 and ASDs 428,429, conditions that share key characteristics with TS. Especially relevant to the current study, it has been shown that white matter diffusivity measurements improve with therapy in toddlers and preschoolers with ASD 430, and both functional and structural connectivity have been proposed as potential biomarkers for monitoring early interventions 431. 86 This leads directly into our exploratory DTI analysis, which revealed that some white matter tracts already display statistically significant differences at one year of age between the three groups. The majority of these exploratory findings followed a pattern in which TS infants differed from XX females but were similar to XY males. In other words, they suggest masculinized development of these tracts in TS. In tract-wide comparisons, this pattern was observed for specific diffusivity metrics in the right arcuate fasciculus (AD), right posterior cingulum (AD & RD), the motor segment of the corpus callosum (AD), the left corticofugal motor tract (AD, RD), and the right premotor corticothalamic tract (AD). Examination of local statistics revealed that this pattern was also present in portions of the left corticofugal motor tract (FA), left IFOF (AD & FA), left ILF (RD), and right optic tract (FA). Two possible explanations exist for this pattern, one being the relative lack of postnatal estrogen exposure in TD males and TS females. Another possibility is that these tracts are influenced by genes that lie outside the pseudoautosomal region but escape X- inactivation. Like TD males, TS girls are expected to be hemizygous for these genes. Interestingly, this pattern was generally observed in relation to AD and AD was generally lower in TS females and XY males. AD is known to decrease sharply across the first year of life 432. Thus, one could interpret high AD as delayed maturation and low AD as accelerated maturation. Alternatively, lower AD in infants with TS may reflect axonal damage (Winklewski et al. 2018), but this seems an unlikely explanation for the lower AD observed in XY males as compared to XX females. The second most common pattern we observed was one in which TS infants differed from both XX females and XY males, who were similar to each other. In tract- wide comparisons, this pattern was observed for the tapetum (AD & FA), left ILF (RD, 87 FA)), left corticofugal motor tract (FA), and left IFOF (FA). Examination of local statistics revealed that this pattern was also present for AD in a posterior area of the left IFOF. This pattern suggests a role for dosage sensitive genes in pseudoautosomal regions of the sex chromosomes, as TD males and females have two copies and TS females have one. However, this pattern could also reflect the influence of clinical characteristics that are more common in infants with TS such as congenital heart disease (CHD) and exposure to general anesthesia during surgeries to correct CHD, kidney abnormalities, or frequent ear infections. Interestingly, this pattern was seen most frequently for FA with TS females having higher FA than TD males or females. FA increases over development such that high FA can be interpreted as faster maturation and low FA as slower maturation. FA does not capture a single biological process, but reflects increasing myelination and axonal integrity 433. The least frequently observed pattern was one in which TS infants were similar to XX females and both differed from XX males. We refer to this as the “sex difference” pattern and it was observed in several regions of the left IFOF (FA) and the right optic tract (FA). This could indicate a role for testosterone in the development of these tracts. Testosterone is significantly higher in XY males as compared to XX females during mid gestation and in the first 6 months of postnatal life 434–439. Testosterone levels in TS are expected to be similar to XX females during gestation and even lower than XX females postnatally. It could also indicate a role for Y chromosome material in the development of these tracts. The relative lack of sex differences is in keeping with prior literature on the development of white matter integrity in infancy440. In subsequent paragraphs we discuss 88 the pattern of results for each tract in more detail and discuss the potential functional and clinical implications of the patterns we observed. The right frontotemporal segment of the arcuate fasciculus connects the middle and inferior temporal gyri with the precentral gyrus and posterior portion of the inferior and middle frontal gyri in the right hemisphere. This tract is involved in visuospatial processing and some aspects of language processing, especially prosody. Damage or underdevelopment of the right arcuate fasciculus is also associated with poorer ability in facial expression-based Theory of Mind (Nakajima et al. 2018), a subdomain of social cognition in which girls and women with TS are often deficient. We observed a pattern in which XX females had greater AD than females with TS, who were similar to XY males, which suggests that both TS females and XY males have more mature tracts than XX females. This seems slightly paradoxical given that this tract is involved in cognitive functions that are often disrupted in TS. Perhaps the development of this tract slows after one year of age in TS females or it may asymptote at a lower level than XX females. The right cingulum originates in the orbital frontal cortices, travels along the dorsal surface of the corpus callosum, then down the temporal lobe to the entorhinal cortex. In addition to linking frontal, parietal, and medial temporal sites, it also connects subcortical nuclei to the cingulate gyrus. We observed masculinization of both AD and RD in the posterior portion of the cingulum adjoining the hippocampus. Like AD, RD decreases over development. While changes in AD are thought to reflect increased numbers of fibers and/or axonal caliber, increased RD is thought to reflect myelination 14. As with the right frontotemporal segment of the arcuate fasciculus, our results suggest more rapid development of this tract in both XY males and TS females. 89 The right cingulum is known to be involved in general cognition and has been linked to a myriad of psychiatric disorders including ASD and ADHD, with lower FA than the TD population being observed 441,442. The cingulum as a whole is one of the last fasciculi to develop fully, with changes seen in the tract until nearly the third decade of life 443,444. Because the developmental period is so long, this allows for more susceptibility to damage or altered growth as there are more opportunities for disruption in development. Based on the ROIs of the atlas presented in Tamnes et al., the statistical significance we found is localized to the parahippocampal segment of the cingulum adjoining the hippocampus 443. It does not appear that the right cingulum adjoining the hippocampus has been studied specifically in development, and this study, to our knowledge, is the first to look in TS in comparison to a normative sample. The parahippocampal segment of the cingulum stems from the temporal lobe and fans out into the occipital lobe 445 and is involved in recognition memory 446. The parahippocampal subdivision also includes projections from the amygdala, a structure typically enlarged in TS 262,447. Specifically, the left amygdala was observed to be larger in TS. The left amygdala functions in fear recognition in facial expressions, which is an area of deficit in TS 448. The motor portion of the corpus callosum, connecting the primary motor cortices in both hemispheres, also presents a masculinizing effect (X0 = XY > XX) at the global level, but in this case, AD is greater in TD males and TS females than in TD females, leading us to the opposite conclusion; there is more mature growth of this fasciculus in TD females than TS females or TD males. Examination of local statistics shows that the overall masculinizing pattern largely reflects difference near the center of the tract. There is also a small region near the left hemisphere periphery that follows a PAR pattern. 90 Individuals with TS have documented difficulties with motor functions in terms of speed and number of motions required for a task 449,450. The greater the spatial processing on these tasks, the poorer performance observed in motor control 451. However, the deficits in motor function do not appear to be linked to the cognitive profile observed in TS 452. The tapetum is a segment of the corpus callosum that extends laterally on either side into the temporal lobe 453. The tapetum has been reported in the literature for TS as having aberrant diffusivity 265,454. This aligns with the cognitive profile of TS as the tapetum is an important contributor to visuospatial functioning 455, and social and communicative functioning 456. In general, our results for FA are in keeping with the existing literature with TD females having greater overall fiber integrity than TS females at multiple locations along the tract. In the literature, lower FA has been associated with TS in the tapetum 264,265. However, there were two locations, one in the far left periphery of the tract and one in the far right periphery of the tract, where TS females had higher FA than the control groups. For AD, examination of local statistics revealed two regions where females with TS differed significantly from XX females. In one region AD was higher in TS females, suggesting delayed development; in the other AD was lower in TS females, suggesting faster development. Although not statistically significant, similar differences were observed when comparing TS females to XY males, suggesting PAR-mediated effects on both FA and RD in this tract. Corticofugal tracts include corticoefferent and corticopetal fiber groups that interconnect the cerebral cortex, corona radiata, internal capsule, cerebral peduncles, pontine nuclei and/or the brainstem. For AD and RD, the left motor corticofugal tract follows the same pattern we see in the right cingulum adjoining the hippocampus, with 91 TD females showing slower maturation than TD males and females with TS. Significant group differences were also observed for FA. However, the primary pattern of differences for FA is more in keeping with a PAR effect than a masculinization effect with TS females having higher FA than both TD males and TD females. Existing literature does not report aberrant DTI measures for the corticofugal tracts in TS, suggesting these differences do not persist. The corticofugal tracts also travel through the internal capsule, which shows lower FA in TS girls than TD girls 258,265. Into middle age, the internal capsule shows higher FA in males than females, demonstrating that the patterns continue to differentiate through time 457. The relationship we observed in the left motor corticofugal tract could also be a contributor to the motor phenotype observed in TS. Corticothalamic fiber tracts are white matter pathways that radiate from the thalamus to the cerebral cortex, via the internal capsule and corona radiata. They are also known as thalamic radiations. Beginning in the thalamus, the premotor corticothalamic tract travels to the premotor regions of the frontal cortex which function in the planning and organization of movements. Again, showing an indication of earlier maturation, the masculinization of the right premotor corticothalamic tract (XX > X0 = XY; AD) follows the previously mentioned patterns. As previously discussed for the corticofugal tract, lower FA in the internal capsule was observed in young girls with TS than TD girls 258,265. The left inferior fronto-occipital fasciculus (IFOF) connects orbitofrontal cortex and the inferior frontal gyrus to the occipital lobe. We observed substantial regional heterogeneity in how group membership affected diffusivity in the IFOF. For AD, we observed a potential sex difference in the anterior IFOF with TS females having 92 significantly lower AD than XY males, but similar AD to females, suggesting faster development of this area in XY males. In the posterior IFOF, we identified a region where AD was increased in TS compared to both XX females and XY males, indicating a PAR effect. For FA, we observed potential sex differences in the anterior frontal lobe and midsection where females (both XX and XO) had greater tract integrity than XY males. There was also a section in the posterior frontal lobe that followed a PAR pattern with TS showing more integrity than both TD males and females. Finally, we observed a possible masculinization effect on a section of the posterior IFOF where TS females had greater tract integrity compared to XX females, but were similar to XY males. The IFOF is involved in attentional control and may contribute to attentional difficulties in ADHD 458, so the fact that we observed delayed maturation of AD in this fasciculus in TS could partially explain the increased vulnerability of girls with TS to ADHD. The left inferior longitudinal fasciculus connects the temporal and occipital lobes. Group differences in this tract for both FA and RD seemed to reflect hemizygosity in the PAR with TS individuals exhibiting greater myelination and higher tract integrity compared to both males and females. Typically in ASD a reduced FA in the ILF 459–461 and visual processing deficits are observed, which is the opposite of what we have found for TS. In a study in girls with TS from ages 7-14, a reduction in FA was observed in the ILF 264. The ILF is also implicated in visuospatial working memory 462 which would suggest a reduction in FA in our TS group, but it could be that our participants with TS undergo a stagnation in the development of the fasciculus after infancy that then becomes apparent with age. Finally, the right optic tract transports visual information from the optic chiasm to the right lateral geniculate body as a part of the visual pathway. Global test statistics 93 suggest that group differences in primarily reflect a sex difference in FA with both TS females and XX females differing significantly from XY males, but not differing from each other. Examination of local test statistics reveals additional complexity. FA differences in the anterior part of this tract are challenging to classify into one of the three patterns with significant differences between TS females and XY males (X0 > XY), but small and non- significant differences between TS and XX females and between XX females and XY males. The posterior part of this tract shows a pattern of masculinized FA (X0 = XY > XX). To our knowledge this tract has not been examined in adolescence or adulthood in TS. Nor are we aware of any studies reporting a sex difference in this tract. The current study has many strengths including the use of advanced image acquisition and analysis techniques optimized for studies of the infant brain. The inclusion of both male and female control groups provides unique insights into possible underlying mechanisms. The developmental period we have chosen also alleviates the issue of the treatment of TS with hormone therapy used to induce puberty which has been demonstrated to cause changes in connectivity 463 . A possible limitation to the study is our moderate sample size which may cause us to be underpowered to detect subtle differences in anatomical and functional connectivity. In conclusion, this study provides novel information about anatomical and functional connectivity in infants with TS. By comparing our results to the existing literature in older children, adolescents, and adults with TS, we have begun to construct a developmental model of this condition. Our results indicate that disrupted connectivity between frontal cortex and parieto-occipital cortex, as indexed by diffusivity measures in the SLF and resting-state connectivity between the right precentral gyrus and right and 94 left calcarine cortex, right and left lingual cortex, and right supramarginal cortex, is not present in late infancy. Early intervention could potentially prevent these brain phenotypes from emerging and lessen the cognitive impact of this genetic disorder in later stages of development. However, our exploratory studies revealed diffusivity differences in other white matter tracts during infancy. Some of these differences appear to persist into adulthood, while others appear to be specific to this developmental period. It is suggested that future studies look longitudinally at structural and functional connectivity in girls with TS in order to better understand brain development in this group, identify biomarkers for later cognitive challenges, and identify critical periods for intervention. 95 Chapter 4 An online platform for testing the disrupted cognitive domains of women with Turner syndrome 96 Abstract Turner syndrome is a relatively rare genetic disorder resulting from the partial or complete loss of the second X chromosome in phenotypically female individuals. Turner syndrome is accompanied by a wide array of multisystemic challenges that include a unique cognitive profile. This profile includes disruptions in executive functioning, visuospatial reasoning, and social cognition. Though the disorder is one of the most common chromosomal aneuploidies, girls and women with Turner syndrome are spread sparsely across the country. Current research to understand cognitive challenges in Turner syndrome requires in-person participation in a clinical setting that thus puts geographic and resource constraints on scientists trying to understand the disorder. To counter this problem, we have developed an online browser-based cognitive testing battery that circumvents the geographical challenges and requirement of trained clinicians for test administration. The online battery targets the three neurocognitive domains that are consistently disrupted in TS through the use of seven videogame-like tests and two traditional surveys. We administered the battery to a cohort of women with TS, with neurotypical men and women as controls. Our results showed discrimination between the three groups on all but one task in the battery, validating the battery for use in other studies. 97 Introduction X monosomy, or Turner syndrome (TS), is one of the most common chromosomal aneuploidies, occurring in approximately 1 in 2000 live female births 247. The disorder is caused by the partial or complete loss of the X-chromosome, making TS a uniquely powerful model for studying X-chromosome effects on human development268,269,395,464. TS is characterized by multisystemic challenges such as gonadal dysgenesis, renal abnormalities, congenital heart defects, and liver disorders398. TS is often characterized by a unique cognitive profile, with high-level processes such as executive functioning (EF)270,402,403 and social cognition (SC)399–401 being affected, along with visuospatial reasoning (VR)404,405. These domains affect skills essential for everyday functioning and influence cognitive flexibility, social interactions, and an understanding of space and measurement. This cognitive profile also includes a higher risk for autism spectrum disorders (ASD) and attention-deficit/hyperactivity disorder (ADHD) which share deficits in these neurocognitive domains250,251. These comorbid disorders are male-biased in the general population464, with a higher prevalence seen in males than females. However, ASD and ADHD have an even higher prevalence in the TS population than the male population. Though a small sample size was used, Turner syndrome girls had a risk of developing ASD 300 times higher than typically developing girls252. In a separate study, an 18-fold increase in the prevalence of ADHD was found in TS girls compared with girls in the general population, equating to a prevalence of approximately 25% in TS girls 254. 98 The genetics of TS are what make it a compelling model to explore these sex-biased disorders. Females with TS are hemizygous for genes in the pseudoautosomal regions of the sex-chromosomes which makes them different from both typically developing XX females and XY males396. However, X0 females can also be conceptualized as partially masculinized at the genetic level as they are expected to be hemizygous for the 15% of genes outside the pseudoautosomal region that escape X-inactivation397, when compared to XX females, but would have similar gene dosage to XY males. The ability to qualitatively and quantitatively investigate the unique domains affected in TS involve a wide array of tests, tasks, and surveys. To explore the cognitive profile of TS, typically a geographically centered cohort is needed, and testing involves an in-person visit to a research facility and the utilization of generally proprietary software. This makes it very difficult to reach a large geographic sample for a relatively rare disorder. Additionally, it would be impossible to conduct this important research without a team of trained specialists with access to subscriptions to the proprietary software, excluding many researchers from exploring questions about cognition in TS due to resource restraints. The current research seeks to ameliorate this large barrier in TS research capabilities by creating an online browser-based cognitive battery that can be taken on a computer anywhere in the United States. This would allow for larger cohorts of TS women to be included in studies without the geographical and resource constraints. In this study we utilize women with TS, and XX females and XY males as controls from across the United States. The objective of this study was to validate the battery as a usable metric 99 for collecting cognitive testing data. We utilize seven tasks and two traditional surveys in the battery, selected for their sensitivity in each cognitive domain. To investigate EF, we used a continuous performance test (CPT), a flanker task, a Corsi block test, and a digit span test. The CPT test requires the participant to push the space bar when the stimulus “X” is present and to ignore all other stimulus letters. Omission errors are then interpreted a measure of sustained attention and errors of commission are viewed as a measure of impulsivity. For the CPT we hypothesized that women with TS would make more errors of commission and omission and have longer reaction times and less accuracy than the controls465. The flanker task requires participants to indicate which direction a stimulus arrow is pointed while surrounded by congruent or incongruent errors. This test is conceptualized as a measure of both selective attention and inhibitory control. TS participants are expected to have longer response times and lower accuracy than the controls in the flanker task, with this pattern being more pronounced in the incongruent tests466. The Corsi block tasks requires a pattern to be memorized and repeated back by selecting blocks on the screen in the order they were presented. It measures visuospatial working memory. In comparison to the normative sample, participants with TS are expected to have a shorter length in pattern recall in the Corsi block test due to documented deficits in both working memory and visuospatial skills467. The final measure of EF, the digit span test, requires the participant to repeat a set of numbers shown briefly to them in a specific order. It is a non-visuospatial test of working 100 memory. We hypothesized that the lengths repeated by the TS group would be significantly less than the control groups468,469. A mental rotation task was used to explore VR in the TS population The mental rotation task takes a three-dimensional polygon and the participant is asked to select the polygon in an array that is the stimulus polygon, but rotated. We hypothesized that TS participants would have poorer accuracy and longer response times than controls. Reading the Mind in the Eyes, a test of facial emotion recognition developed by Simon Baron-Cohen, was used to evaluate social cognition. It involves being presented with a set of eyes and being asked to choose which of four emotions you believe the eyes to be conveying. For Reading the Mind in the Eyes the working hypothesis is that TS participants will have lower accuracy than controls465,468. The final video-game like task was a simple response time (SRT) test designed to measure processing speed. It asks the participant to react when an “X” appears on the screen. We hypothesized that women with TS would exhibit more anticipations on the simple response time task and a longer response time due to an average slower processing time than controls and higher levels of impulsivity470. The two traditional surveys used in the battery include the ADHD self-report scale and the Autism Spectrum Quotient (ASQ). We hypothesized that for the ASQ, scores would be higher for women with TS than the normative sample. We expected more TS women to score above threshold on the ADHD report scale than controls due to the higher prevalence of ADHD in TS than the general population254. Further, the presence of ADHD 101 based on the report scale will have a significant relationship with higher errors and response time and lower accuracy on the CPT. Methods Participants Women with Turner syndrome (TS) over the age of 18 were contacted via emails provided through the Turner Syndrome Society of the United States (TSSUS) that were solicited after an IRB approved message was sent to members across the country asking for participation. Thirty women that responded to the call for participants also finished the cognitive battery in its entirety (Table 4.1). The control group consisted of 34 men and 61 women without TS between the ages of 18-60 and without any health issues or medication usage as reported to the ResearchMatch database, which was used to gather participants for the control portion of the study (Table 4.2). ResearchMatch is a national health volunteer registry that was created by several academic institutions and is supported by the U.S. National Institutes of Health as part of the Clinical Translational Science Award (CTSA) program. ResearchMatch has a large population of volunteers who have consented to be contacted by researchers about health studies for which they may be eligible. Review and approval for this study and all procedures was obtained from Michigan State University’s Institutional Review Board. 102 Table 4.1 Demographic and medical history variables for Turner Syndrome participants. Diagnosed condition N % Variable N % Mean (SD) range X0 karyotype of those that reported 7 58% Ethnicity Aortic dissection 1 3% Asian 1 3% Aortic enlargement 2 7% White 25 83% Arthritis 4 13% Native American 1 3% Chronic ear infections 11 37% Age of diagnosis Chronic kidney disease 0 0% Prenatal 1 3% Coronary artery disease 0 0% Less than 1 8 27% Craniofacial abnormality 1 3% Childhood 20 67% 10.6 (5.1) 1-18 Curvature of spine 4 13% Greater than 20 1 3% Treated with estrogen replacement Diabetes 3 10% therapy 22 73% Gastrointestinal Age of estrogen problems 3 10% replacement therapy 14.6 (2) 10-18 Currently receiving estrogen replacement Glucose intolerance 2 7% therapy 18 60% Treated with growth hormone Hearing impairment 16 53% therapy 14 47% Heart abnormality 7 23% High blood pressure 9 30% High cholesterol 11 37% Kidney abnormality 7 23% Learning disorder 4 13% Liver disease 3 10% Osteoporosis 7 23% Ovarian failure 20 67% Seizures 0 0% Stroke 1 3% Thyroid disease 10 33% Visual impairment 12 40% 103 Table 4.1 (cont’d) Vitamin D deficiency 9 30% Table 4.2 Demographic variables for control participants. Variable Female Male Mean (SD) Mean (SD) N % range N % range 40 (15) 18- 41 (13) 22- Age 61 60 34 60 Race White 48 79% 24 71% Black 4 7% 4 12% Asian 4 7% 2 6% Mulit- racial 1 2% 0 0% Other 4 7% 1 3% Ethnicity Hispanic 3 5% 4 12% Not Hispanic 58 95% 30 88% Cognitive Battery The cognitive battery was designed and hosted on RedCap471,472, a browser-based software for creating surveys and managing databases. Designed in collaboration with the Michigan State University Biomedical Research Informatics Core (BRIC), the battery consisted of two traditional surveys and seven interactive game-like tests. The metrics included in the full battery are as follows: mental rotation task, flanker task, Reading the Mind in the Eyes, continuous performance task, Corsi block task, digit span task, simple response time task, the autism spectrum quotient, and the ADHD report scale. Mental rotation task Three dimensional polygons were used as the stimuli for the mental rotation task and were acquired from a validated dataset in a previously published work that created 104 the stimulus set473. The stimulus three-dimensional polygon was shown at the top of the screen, followed by five three-dimensional polygons beneath which were the stimulus polygon but rotated 50, 100, and 150 degrees in an order set by a random number generator. Only one of the five polygons was the stimulus polygon merely rotated, as opposed to mirrored, or mirrored and rotated. There were 36 different sets of polygons given to the participant. Participants had unlimited time to respond. Reaction time in milliseconds and accuracy were recorded. Flanker task The participant is given a set of five arrows. The middle arrow is randomly pointed left or right, and the four flanking arrows (two on each side of the middle arrow) are randomly facing left or right as well. Sometimes the middle arrow is congruent with the flanking arrows, and sometimes it is incongruent. There were 96 trials given. We measured accuracy and response time stratified by congruent and incongruent stimuli, Reading the Mind in the Eyes This test assesses the participant’s ability to use non-verbal cues to determine an emotion. Individuals are shown a pair of eyes and are then asked to pick from four choices which emotion they think best fits the stimulus. A definition was provided with each emotion. This test was developed in 1997474, with an updated version published in 2001137. Outcome was the number of stimuli identified correctly. Continuous performance task Continuous performance tasks are designed to measure both sustained attention and inhibitory control475. Our continuous performance task (CPT) is exactly 14 minutes in 105 length. The goal of the task is to press the space bar on a computer’s keyboard when a letter other than “X” appears on the screen. There are 360 trials with 36 X’s randomly appearing within 324 non-X stimuli. Randomly, there are 1, 2, or 4 seconds between presentation of the letters. Outcome measures included the number of commissions (pressing the space bar when “X” is shown on the screen), number of omissions (not pressing the space bar when a letter other than “X” is on the screen), reaction time for each trial, and overall accuracy. Corsi block task Nine identical and spatially separated blocks are shown on the screen which light up in a specific order. Participants were then asked to repeat this order back by clicking on the boxes in the sequence originally presented. Longest length of pattern and reaction time, defined as the amount of time a participant took to repeat back the sequence, were measured. The pattern sequence increases by an additional block each turn and the task ended when an incorrect response was returned. Digit span task A list of numbers is given one by one which had to be repeated back by the participant in the same order it was given by typing the numbers into the program. Participants were allowed three errors total throughout the task. The number pattern increased by one each trial until the participant returned an incorrect input three times. The length of numbers repeated back was the outcome measure. An ANOVA was performed between the three groups to identify differences in working memory capacity. 106 Simple response time task The simple response time task (SRT) is a reaction time task in which a single stimulus (an ‘X’) appears at a specifiable delay 1-3 seconds from the previous response. Two hundred trials were performed. Anticipations, or the number of times the spacebar is pushed before the stimulus appears on the screen, were measured along with response time which was defined as the time between when the stimulus was shown and when the spacebar was pressed. Autism spectrum quotient The autism spectrum quotient (ASQ) is a self-assessment tool for screening autism spectrum disorders (ASD). Each question gets one point if answered disagree/agree or strongly disagree/strongly agree. There are 50 questions total with 25 eliciting “disagree” responses and 25 meant to elicit “agree” responses. A cutoff of 32+ distinguishes well between individual with ASD and controls476. ADHD report scale A self-assessment tool for screening ADHD. The first six questions were used for scoring. A point was given if sometimes/often/very often was chosen for the first three questions, and an additional point is given for an often/very often selection on the three questions following. A score of four or more is indicative of possible ADHD. Statistical Analysis All group comparisons were done with an ANOVA. Post hoc analysis was performed using a Benjamini-Hochberg for multiple corrections. All analyses were run both with and 107 without outliers, which were defined as data points more than two standard deviations above and below the mean for the combined groups. Results Mental rotation task Looking at overall differences in reaction time, the data were heavily skewed to the right (Figure C.1). Thus, a log transformation was performed on the data (Figure C.2). The ANOVA between the three groups showed statistical significance (p<2x10 -16; Figure 4.1a; Table 4.3). After outlier removal the relationship remained significant (p<2x10 -16; Figure 4.1b). The post hoc test revealed statistical significance between X0 and XY, and X0 and XX (p<2x10-16; p<2x10-16) in which the X0 group had longer response times, but no difference between control groups (Table 4.4). Figure 4.1 Overall differences in reaction times between groups. A) Differences in reaction times before outlier removal. Based on the ANOVA, p<2x10-16. B) Differences in reaction time after outlier removal at two times the standard deviation above and below the mean. Based on the ANOVA, p<2x10-16. 108 Table 4.3 ANOVA results for general rection time between the three groups on the Mental Rotation Task. A) Log transformed differences in reaction times before outlier removal. B) Log transformed differences in reaction time after outlier removal. Sum Sq Mean Sq F p-value (A) Group 79.7 39.83 55.24 <2x10-16 (B) Group 66.3 33.13 54.16 <2x10-16 Table 4.4 Post hoc comparison of general reaction times over all rotational groups after outlier removal and log transformation on the Mental Rotation Task. Comparison Adjusted p-value General reaction time X0-XX <2x10-16 X0-XY <2x10-16 XX-XY 0.76 109 The data for general accuracy showed a normal distribution (Figure C.3), so an ANOVA was run. When general accuracy between the three groups was queried a statistically significant relationship was found (p=2.8x10-7; Figure 2a; Table 4.5) before but not after outlier removal (p=0.143 ; Figure 4.2b; Table 4.5). Post hoc comparisons again showed the pattern of significance between X0 and controls, but not between the controls themselves with X0 showing the lowest scores (Table 4.6). Figure 4.2 Overall differences in accuracy between groups on the Mental Rotation Task. A) Differences in reaction times before outlier removal. Based on the ANOVA, p=2.8x10-7. B) Differences in reaction time after outlier removal at two times the standard deviation above and below the mean. Based on the ANOVA, p=0.143. Table 4.5 ANOVA results for general accuracy between the three groups on the Mental Rotation Task. A) Before outlier removal. B) After outlier removal. Sum Sq Mean Sq F p-value (A) Group 571 25.67 15.72 2.8x10-7 (B) Group 13.4 6.72 1.96 0.143 110 Table 4.6 Post hoc comparison of general accuracy over all rotational groups before outlier removal. Comparison Adjusted p-value General accuracy X0-XX 1.5x10-6 X0-XY 1.5x10-6 XX-XY 0.58 Flanker task The reaction time data were distributed in a skewed manner thus, a log transformation was performed on the data (Figures C.2 and C.3). A comparison between the three groups in an ANOVA yielded a statistically significant relationship for response time on congruent patterns of the flanker both before and after removal of outliers (p<2x10-16; p<2x10-16; Figure 4.3; Table 4.7) with each pairwise comparison also showing statistical significance both before and after outlier removal (Table4. 9). X0 had the longest response time followed by XY and XX. Figure 4.3 Differences in log transformed response times on congruent stimuli on the Flanker task. A) Differences in response times after log transformation but before outlier removal. Based on the ANOVA, p<2x10-16 B) Differences in response time after 111 Figure 4.3 (cont’d) log transformation and outlier removal at two times the standard deviation above and below the mean. Based on the ANOVA, p<2x10-16. 112 Table 4.7 ANOVA results for log transformed reaction time on congruent stimuli for the Flanker task. A) Before outlier removal. B) After outlier removal. Sum Sq Mean Sq F p-value (A) Group 57 28.5 340.9 <2x10-16 (B) Group 103.3 51.66 261.2 <2x10-16 Although a relationship in the response time exists between the three groups, there was no statistical significance between TS and controls on accuracy for congruent markers (p=0.18). For incongruent trials, the reaction time data did not have a normal distribution so a log transformation was performed (Figure C.5 and C.6). The same pattern was observed for questions on the Flanker task with incongruent arrows; significant group differences were found in response time (p<1x10-15; Figure 4.4), but not accuracy (p=0.25). The pairwise comparison of the three groups on incongruent response time showed significant differences between all three groups both pre- and post-outlier removal (Table 4.9). Like the congruent questions, response time was greatest in X0, followed by XY and XX. 113 Figure 4.4 Differences in response times on incongruent stimuli on the Flanker task. A) Differences in response times before outlier removal. Based on the ANOVA, p<1x10-15. B) Differences in response time after outlier removal at two times the standard deviation above and below the mean. Based on the ANOVA, p<1x10 -15. Table 4.8 ANOVA results for log transformed reaction time on incongruent stimuli for the Flanker task. A) Before outlier removal. B) After outlier removal. Sum Sq Mean Sq F p-value (A) Group 155.2 77.6 420.8 <2x10-16 (B) Group 74.8 37.9 295.4 <2x10-16 Table 4.9 Post hoc comparisons for log transformed reaction time, congruent and incongruent trials, on the Flanker task both before and after outlier removal. Congruency Comparison Adjusted p-value Congruent pattern X0-XX <2x10-16 X0-XY <2x10-16 XX-XY 6.1x10-9 Congruent pattern (outliers removed) X0-XX <2x10-16 X0-XY <2x10-16 XX-XY 3.1x10-10 Incongruent pattern X0-XX <2x10-16 X0-XY <2x10-16 XX-XY 6.7x10-6 114 Table 4.9 (cont’d) Incongruent pattern (outliers removed) X0-XX <2x10-16 X0-XY <2x10-16 XX-XY 2x10-7 Reading the mind in the eyes The data in the Reading the Mind in the Eyes had a normal distribution for all three chromosomal arrangements (Figure C.7). Based on the ANOVA, no statistical significance was found between the three groups in their accuracy on the 36-question test (p = 0.068; Figure 4.5; Table 4.10). The comparison between X0 and XX had a p- value of 0.062 and is noted in Figure 4.5. Figure 4.5 Differences in the score on the Reading the Mind in the Eyes task between the three groups. The ANOVA was not significant. Table 4.10 ANOVA results between the three groups for Reading the Mind in the Eyes. Sum Sq Mean Sq F p-value Group 83.5 41.77 2.757 0.068 115 Continuous performance task All data were deemed normally distributed after outlier removal, with the exception of errors of omission, which was skewed slightly to the right (Figures C.8-C.15) No difference was found between the groups on reaction time both before and after outlier removal (p=0.38; p=0.31). Based on the ANOVAs, the three groups did not show statistically significant differences in the number of correct responses both before and after outlier removal (p=0.8; p=0.42). No differences were found between groups on errors of commission before and after outlier removal (p=0.8; p=0.5), nor on errors of omission (p=0.797; p=0.855). Corsi block task The data were normally distributed both before and after outlier removal (Figures C.17 and C.18). Length of pattern repeated was measured, and an ANOVA calculated between the XX, XY, and X0 groups. The ANOVA showed a statistically significant relationship (p=1.37x10-5; Figure 4.6a; Table 4.11). Post hoc comparisons showed the difference being driven by significant differences between X0 and XX, and XY and X0, but not XX and XY (Table 4.2). After outlier removal, the ANOVA was still significant (p=4.8x10-5; Figure4.6b; Table 4.11) and post hoc group differences were similar (Table 4.12). 116 Figure 4.6 Differences in the length of blocks repeated on the Corsi Block Test. A) Before outlier removal. The ANOVA is significant at p =1.37x10-5. B) After outlier removal. The ANOVA is significant at p=4.8x10-5. 117 Table 4.11 ANOVA results between the three groups for the Corsi Block Task for length of blocks repeated. A) ANOVA results before the removal of outliers. B) ANOVA results after the removal of outliers. Sum Sq Mean Sq F p-value (A) Group 1600.6 80.29 12.32 1.37x10- 5 (B) Group 13.91 6.95 4.62 4.8x10-5 Table 4.12 Post hoc comparisons for maximum blocks repeated on the Corsi Block Task before and after outlier removal. Comparison Adjusted p-value Max blocks X0-XX 4.6x10-5 X0-XY 4.6x10-5 XX-XY 0.71 Max blocks (outliers removed) X0-XX 0.0003 X0-XY 6.5x10-5 XX-XY 0.22 Digit span task The data were deemed normally distributed (Figure C.19). ANOVA indicated significant differences between the three groups when the entire sample was evaluated (p=0.041; Figure 4.7a) and when outliers were removed (p=0.011; Figure 4.7b) While post-hoc pairwise comparisons were not significant in the full sample, when outliers were removed post- hoc analysis revealed that XX women and XY men could recall longer numbers strings than women with TS (Table 4.14). 118 Figure 4.7 Differences in the length of numbers repeated on the Digit Span Task. A) Before outlier removal. The ANOVA is significant at p =0.041. B) After outlier removal. The ANOVA is significant at p=0.011. Table 4.13 ANOVA results between the three groups for the Digit Span task for length of numbers repeated. A) ANOVA results before the removal of outliers. B) ANOVA results after the removal of outliers. Sum Sq Mean Sq F p-value (A) Group 36.4 18.179 3.293 0.041 (B) Group 23.14 11.569 4.68 0.011 Table 4.14 Post hoc comparisons for maximum number of repeated numbers on the Digit Span Task after outlier removal. Comparison Adjusted p-value Max number repeated (outliers removed) X0-XX 0.013 X0-XY 0.013 XX-XY 0.56 119 Simple response time task The response time data were heavily skewed to the right, and thus log transformed (Figures C.20 and C..21). Based on the ANOVA, there is a statistically significant difference between the three groups on the SRT (p=<2x10-16; Figure 4.8a; Table 4.14), which became stronger after removal of outliers (p<<2x10 -16; Figure 4.8b; Table 4.15). Table 4.16 displays the post hoc relationships, which show that X0 participants had a longer reaction time than XY participants, and XY had longer reaction times than XX. Figure 4.8 Reaction time between the three groups on the Simple Response Time Task. A) Before outlier removal. The ANOVA is significant at p <2x10 -16. B) After outlier removal. The ANOVA is significant at p<2x10-16. Table 4.15 ANOVA results between the three groups for the Simple Response Time Task for reaction time. A) ANOVA results before the removal of outliers. B) ANOVA results after the removal of outliers. Sum Sq Mean Sq F p-value (A) Group 42.9 21.4 193 <2x10-16 (B) Group 42.2 21.1 269.6 <2x10-16 120 Table 4.16 Post hoc comparisons for reaction time on the Simple Response Time Task before and after outlier removal. Comparison Adjusted p-value Reaction time X0-XX <2x10-16 X0-XY 0.17 XX-XY <2x10-16 Reaction time (outliers removed) X0-XX <2x10-16 X0-XY 1.7x10-12 XX-XY <2x10-16 Autism spectrum quotient The data were normally distributed (Figure C.22). The ANOVA between the three groups did not yield a statistically significant relationship (p=0.285; Figure 9) when total score was measured. A score of thirty-two or above is indicative of the presence of some autistic traits. Figure 4.9 Score on the Autism Spectrum Quotient between the three groups. The ANOVA was not significant. 121 ADHD report scale The data were deemed to be normally distributed after outlier removal (Figures C.23 and C.24). No statistical significance was found between the three groups on the ADHD report scale in terms of total score (p=0.137; Figure 4.10a) both before and after outlier removal (p=0.98; Figure 4.10b). Figure 4.10 Score on the ADHD Report Scale between the three groups. A) The ANOVA was not significant for the relationship before outlier removal. B) The ANOVA was not significant for the relationship after outlier removal. Discussion To our knowledge, we are the first group to have created and validated a browser- based online cognitive testing platform that is able to show differences between women with TS and male and female controls across all three cognitive domains affected in TS. Overall, all but the two surveys, Reading the Mind in the Eyes, and the CPT in the battery showed separation between X0 and controls in some aspect, indicating the effectiveness and sensitivity of the battery. 122 Regarding the tests of EF, we did not see differences in accuracy on the flanker task. This is not unexpected, as accuracy on flanker tests is typically very high and in our own data, most individuals performed at/near ceiling. Reaction time data revealed that women with TS responded significantly more slowly than both control females and control males on both congruent and incongruent trials. Slower reaction time for congruent trials could represent an impairment in sustained attention in women with TS, while slower reaction time on incongruent trials could reflect deficits in selective attention and inhibitory control 398, which rely on activation of the fronto-parietal network477, specifically the anterior cingulate cortex (ACC)478,479. Studies of sustained attention have shown conflicting results in TS. In contrast, studies of inhibitory control have been more consistent in showing impairments in TS. Conditional accuracy functions, which show how accuracy varies as a function of reaction time, can also be estimated on the flanker task. The conditional accuracy functions demonstrate that accuracy for congruent and incongruent responses approach the same level when more time is taken for incongruent responses480. Individual differences in conditional accuracy are a marker of inhibitory control and selective attention with the flankers being processed first, followed by the middle arrow481. It could reasonably be assumed that in the case of TS participants, who have a deficit in response inhibition and selective attention 403, would have lower conditional accuracy than controls. We saw no statistical significance on any of the four aspects of the CPT (score, reaction time, commissions and omissions). This could be due to the test’s length (14 minutes) and the fact that the test was not taken in a clinical setting void of most distractions. However, it has been shown previously that there is no statistically significant 123 difference between at-home and in-clinic testing when using a cognitive battery, though the CPT was not part of this cited battery482. Typically, brain networks implemented when taking the CPT include the default mode network483 which is involved in activities of sustained attention484,485. The previously mentioned ACC486 is also involved in sustained attention which is the goal measurement of the CPT. The simple response time task showed statistical significance between the three groups. The SRT is a measure of processing speed 487 and the observed pattern of X0 having the longest reaction times followed by XY aligns with our hypotheses. Because the SRT is a measure of processing speed, these results should be taken into consideration in the context of the rest of our findings on the battery that rely on measurements of reaction time. The processing speed is an underlying factor that influences any and all of the reaction time measurements taken in this battery. General processing speed has been shown to recruit the frontal, parietal, and temporal cortices, which are connected via the superior and inferior longitudinal fasciculi19,488. The superior longitudinal fasciculus plays a large role in cognition, specifically EF. The Corsi block and digit span tasks compose our exploration into the EF subdomain of working memory. We were able to reject the null hypothesis in both cases as both tasks showed statistical significance in distinguishing the three groups based on length of blocks or numbers repeated. The Corsi block test has been shown in the literature to be an effective tool in discriminating between TS and TD individuals in children489. The same situation is found for the digit span, in which the literature shows a statistically significant difference between TS and TD children490 . These results validate our battery’s ability to function as a method to test working memory in these two cases, 124 and situate it in a successful position to test measurements of EF. In terms of the visuospatial component of the Corsi block task, forming the spatial map required to repeat the sequence recruits the temporo-parietal junction for maintenance of this map 491. Both the occipital and parietal lobe are recruited for this task491,492. Visuospatial reasoning was tested via the mental rotation task. The 1980 paper by Rovet and Netly493 concerning the mental rotation task in women with TS demonstrated poorer accuracy and poorer reaction time in the TS group which has been shown across the literature399,494. We are now able to add to this rich literature base with our own results that are able to reject the null hypothesis in terms of seeing both a difference in accuracy and in reaction time. This validates an additional cognitive domain disrupted in TS in our battery. Previous work has shown the degree of activation in the superior parietal cortex is proportional to the degree in which the object is rotated495. We were unable to reject the null hypothesis in our measure of social cognition. Reading the Mind in the Eyes showed no differences between the groups. It has been shown that typically developing females generally have an advantage on the Mind in the Eyes test496 over TD males.. It has been previously demonstrated that fronto- temporoparietal connectivity497 is a key circuit involved in the processing required for this task. Specifically, the temporoparietal junction498 is a main structural neural substrate that may be required for this task. Levels of activation in the orbitofrontal cortex have also been associated with performance on the task499. Though both Likert-style surveys did not display any statistical significance when comparing the three groups, the pattern observed for both the ASQ and ADHD report scale match what we would have expected to see; control males have higher scores of 125 symptomologies for both ASD and ADHD than control females before outlier removal. TS participants showed greater scores of symptomologies than female participants, which aligns with the observed higher prevalence of ADHD and ASD in the TS population, however, control males scored higher than the TS participants which we hypothesized would have been greater or equal to the control males. This may be due to lack of sufficient sample size, but could also be a function of selection bias. The control participants chosen for the study were selected using filters that removed any medical conditions and medication use, which reduces the potential for ASD and ADHD diagnoses to come through. Additionally, the participants who completed the study, both TS and controls, were able to finish an hour-long battery, a feat that possibly would be unattainable with a more severe presentation of ADHD or ASD traits. Limitations to this study include a relatively small sample size and possible factors introduced by not performing the battery in a clinical environment such as interruptions, multitasking, and the possible presence of other persons. Though fewer than thirty participants from the controls and TS groups were involved in the study, we were still able to show high levels of statistical significance across all but the CPT, Reading the Mind in the Eyes, and the two surveys. As previously noted, it has been shown that there is no statistically significant difference between a cognitive battery given at home, or in a clinic. The fact that the battery is able to be applied with merely the need for an internet connection and does not require travel to a clinical setting or staffing professionals in the clinic makes it an incredibly useful tool that has the potential to have more far-reaching applications than a traditional battery. Uninhibited by geographical location, the presented battery could potentially be used across the country. 126 In future iterations of the presented cognitive battery the CPT should be removed. No statistically significant differences were found between the three groups. Additionally, zeroes were scored which indicates that either the instructions were not clear, or that the test was not taken with the individual’s fullest attention and effort. The CPT is a very long test at 14 minutes, and may be entirely too long to have on the battery when discrimination between the groups is based on data that show the test was not taken properly. In conclusion, the presented study has shown validity in our testing method, paving the way for the battery to be expanded to encompass larger populations of women with TS and including more items. The creation of this cognitive battery gives researches to opportunity to expand research in TS to explore associations with brain structure, brain function, and perhaps the genetic underpinnings that drive neurodysgenesis. 127 Chapter 5 Conclusions and future directions 128 Summary The research presented in this dissertation broadens our understanding of the genetic underpinnings of executive function, social cognition, and associated neural networks. In Aim 1 (Chapter 2), I used a classic twin design to investigate genetic and environmental factors influencing neonatal resting-state functional connectivity including connectivity within and between networks involved in executive function and social cognition. In Aim 2 (Chapter 3), I compared functional and structural connectivity between typically developing male and female infants and infants with Turner syndrome. In Aim 3 (Chapter 4), I developed an online testing platform targeting cognitive domains that are often disrupted in TS, including executive function and social cognition, and validated the online testing platform by administering it to neurotypical males and females and to adult women with Turner syndrome. In the following paragraphs I summarize the key findings for each aim and discuss their statistical and biological significance. In Aim 1, I demonstrated that narrow-sense heritability estimates for resting-state phenotypes were relatively low with only 6 of the 36 phenotypes examined showing narrow-sense heritability estimates greater than 0.10, indicating a small effect size. This suggests that environmental factors play a greater role than genetic factors in shaping individual differences during this critical period in the establishment of brain circuitry. In addition, statistically significant associations between neonate resting-state functional connectivity phenotypes and specific demographic and medical history variables were observed. One key finding is that maternal psychiatric history was positively associated with a between-network pair that included the frontoparietal network (p=0.0012; explains 3.66% of variance). This network is involved in higher order levels of cognitive processing. 129 This result is highly significant biologically speaking as it implies a possible root to executive functioning development in the offspring. Overall, I concluded that the study was well powered to detect heritability of 0.5 or greater at 80% power, but had low power to detect heritability less than 0.25. A larger sample size has the potential to show that the low heritability estimates observed are significant, but at present I determined that these specific resting-state connectivity phenotypes are not highly heritable at this stage in development. A similar study in a larger sample size is suggested for future works along with the possibility of a study focusing specifically on the environmental determinants of these phenotypes in which I did find statistical significance. A more extensive discussion of future research directions can be found later in this chapter. For Aim 2, my initial hypotheses that infants with TS would show altered integrity of the superior longitudinal fasciculus and reduced connectivity between the right precentral gyrus and occipital and parietal regions involved in executive function, social cognition, and visuospatial processing was not supported. The FDR adjusted p-values ranged from 0.1 to 0.59 with effect sizes as calculated by eta squared of 0.07 (medium; right calcarine cortex), 0.08 (medium; left calcarine cortex), 0.081 (medium; right lingual cortex), 0.063 (medium; left lingual cortex), and low at 0.033 for the right supramarginal gyrus. This null result is important as it suggests that altered fronto-parietal connectivity in TS emerges after one year of age and could potentially be prevented by early intervention. For the right calcarine and lingual cortices a statistically significant value was found before FDR correction. In conjunction with the medium effect sizes, the statistical significance may suggest a biologically important effect. Additionally, an exploratory analysis revealed several fasciculi that differed between the groups. Some of these 130 fasciculi are involved in cognitive domains disrupted in TS, such as the inferior fronto- occipital fasciculus (XY-X0; corrected p = 0.007). I also observed unexpected results such as differences in the optic tract (XY-X0; corrected p = 0.035), which is a difference never before reported. In terms of the effect size of the findings, each tract is unique. The effect sizes of the statistically significant exploratory tracts, represented in percent change are as follows: corpus callosum tapetum FA and AD (0.01%, 8%), left inferior longitudinal fasciculus RD, FA (30%, 0.02%), left IFOF AD, FA (12%, 0.15%), corpus callosum motor AD (15%), left motor corticofugal tract FA, AD and RD (0.03%, 300%, 150%), left corticothalamic tract AD (6%), right cingulum adjoining the hippocampus AD, RD (7%, 15%), right arcuate fasciculus frontotemporal region AD (6%), and the right optic tract FA (0.08%). Though all were statistically significant, those percent changes lower than one percent may not be of clinical relevance whereas the left motor corticofugal tracts change of 300% and 150% may show biological significance. Those percent changes falling in the middle (12%-30%) may be of biological significance and a larger sample size would be able to hone in on a more accurate estimation. Overall, the key findings from my exploratory analysis are that X0 and XY individuals show greater maturity for several white matter tracts compared to XX females as indexed by AD. A logical next step would be to determine how long these differences persist by conducting a longitudinal study. Additional future directions are discussed further in this chapter. For Aim 3, Women with TS performed more poorly on tests measuring visuospatial reasoning and executive functioning with adjusted p-values ranging from 2x10-16 at the smallest to 0.013 as the largest. Using Cohen’s D to calculate effect sizes on the battery items that showed a statistically significant difference, reaction time on the mental 131 rotation task showed a medium effect size for the X0-XX comparison (0.6) and X0-XY comparison (0.58). A small to medium effect size of 0.49 was observed for the Flanker task on congruent questions for response time for the X0-XY comparison while the X0- XX comparison had a medium effect size of 0.62. As previously discussed TS can be thought of as having a masculinizing effect. This may be why the observed effect size is greater between XX and X0 females. This pattern also holds true for the response time on incongruent stimuli in the Flanker, though the effect sizes are much larger (X0-XY = 0.71; X0-XX = 0.92). This is most likely due to the increased difficulty in choosing a target when the flanker and center arrow are not in the same direction, creating a larger gap in scores. Cohen’s D was positive in the last three examples because the reaction time was greater in the X0 group than the controls. The following two examples have negative Cohen’s D scores because the controls scored larger on repeating both blocks and number strings in the Corsi Block test and Digit Span test. For the Corsi blocks, as mentioned, the TS group scored lower in repeated block patterns than the neurotypical controls with large effect sizes (X0-XY = -0.97; X0-XX = -0.94). The Digit Span additionally showed lower scoring in the repetition of numbers in TS than in the control groups though this effect was only medium to high or medium (X0-XY = -0.77; X0-XX = -0.63). Further in this chapter future directions are discussed more fully for Chapter 4, but overall the online platform successfully revealed robust differences between the three groups and a continuation of the research would be to investigate if factors contributing to individual differences in TS women can be distinguished. 132 Rigor and reproducibility Currently, science is plagued with a crisis of reproducibility500,501. Scientists are unable to replicate the studies of others, or even their own studies. Contributing factors to this crisis include inappropriate correction for multiple testing and lack of methodological details in published studies. To uphold scientific rigor and construct this manuscript in a way that is reproducible, this dissertation applied false discovery rate corrections to all applicable situations, making sure that the statistical significance was not artificially inflated due to improper accounting of multiple testing. This dissertation additionally includes the packages and versions of each R package utilized along with the version of R it was built under. All methods are detailed for ease in reproducing and active contact information is available for researchers to engage with if questions arise. Ultimately, all data from Chapter 4 will be deposited in the Turner Syndrome Research Registry for future use by scientists. Future work Longitudinal imaging of infants As previously discussed, the heritability of resting-state functional connectivity is under mild to moderate influence in adults and adolescents. At the two-week postnatal timepoint, we do not observe this established heritability. A continuation of our work in Aim 1 would be to look longitudinally at a cohort of DZ and MZ twins and watch the development of the resting-state networks over time. This has the potential to show where exactly genetics begins to play a larger role in the brain’s functional connectivity and helps to define a critical period in development where the environment may play a larger role in 133 the shaping of networks. A larger cohort of infant twin-pairs would need to be used as well to overcome the barrier of insufficient power. In a study by Gao and colleagues324 intersubject variability in the brain’s functional connectivity was shown across the age span of infancy at timepoints of 1 month, 1 year, and 2 years. The study demonstrated an overall decrease in variability over time in the connectivity in a back to front manner, with higher-order cognitive networks demonstrating the highest variability at the end of the age range. This study combined with the results from my Aim 1 speak to the process of canalization and now gives evidence to canalization of resting-state functional connectivity. As age increases intersubject variability decreases, as previously stated, and I hypothesize that with this decrease, heritability increases in the same back to front pattern that is consistently observed throughout the literature, i.e., higher order cognitive processes will be the last networks to achieve a heritability value consistent with adult and adolescent levels. Utilization of sample-derived networks in the imaging of neonates A potential path for future analysis for Aim 1 includes not using an adult-defined atlas to map the neonates’ networks onto, but instead defining the networks based on the neonatal data. An examination of heritability and environmental effects would then follow. This would be considered an exploratory study but has the potential to be more infant- specific and perhaps give different heritabilities than by using the adult atlas. For Aim 1 I chose to use the adult-defined atlas because we aimed to examine how canonical networks that are commonly identified in adults are impacted by genetic and environmental factors. Thus, adult-defined networks were used to facilitate interpretations of our findings. This chosen approach has the limitation that neonates do not have these 134 fully developed networks during this period, and the adult atlas is an estimation. An alternative approach to address this limitation is the use of independent component analysis (ICA) which is data-driven and model free, allowing for an exploratory analysis. This approach was not chosen for the current study as there were specific hypotheses made and an overall research question. However, it would be an interesting direction for future research. The temporal relationships between volume, tract integrity, and resting state connectivity in TS Aim 2’s objectives and hypotheses were largely based on previous work showing differences in brain volumes in the same cohort, as explained in Chapter 3. We did not see connectivity between these regions at this age which indicates that these characteristic markers of TS develop later in life. Pinpointing the developmental time in which the SLF loses white matter integrity as evidenced by a reduction in FA sets a period in which interventions could be given to help ameliorate the difficulties that occur later in life. This same idea holds true for the connectivity between the frontal and visual cortices. Finding the intersection between volumetric and anatomical or functional changes could also have implications for other neurodevelopmental disorders in which connectivity could still be developing even though a volume difference is observed. The reverse may also be true and would require a longitudinal study of the one-year-old infants into childhood and adolescence to look at if the strengthening of connectivity in the SLF and frontal- visual connectivity via therapeutic intervention reduces the volumetric differences seen between the three cohorts. This would help to establish a stronger connection between brain volumes and its implications for brain connectivity. Research investigating the 135 coupling of structural and functional connectivity have shown regional patterns in age, sex, and cognitive phenotypes502,503, with wide ranging patterns. The strength of these couplings has additionally been shown to follow the back to front development pattern of other brain characteristics, with the regions involved in higher order cognitive processes such as executive functioning growing in strength with age503. Implementation of the cognitive battery on a sample of TS women with sequenced exomes As discussed in previous chapters, TS is characterized by deficits in the visuospatial, executive function, and social cognitive domains. Chapter 4 in this dissertation successfully established a cognitive battery designed specifically for use in TS to measure and assess these cognitive domains. This battery would be very simple to utilize in other studies related to probing cognition in TS. Currently, there exists a cohort of 188 TS women that also have had their exomes sequenced 504. This cohort previously participated in an exome sequencing study involving investigation of the genetics of aortopathy in TS. By utilizing this cohort and having them take our validated battery, we have the potential to be the first to show genes and pathways that influence these cognitive domains in TS with implementation of an exome-wide association study. As previously stated, this cohort of women were involved in an aortopathy exome sequencing study. The result of that study yielded identification of a genetic variant associated with the specific aortopathy, indicating that the statistical methods used were powerful enough and had high enough sensitivity to discern variants, despite a sample size smaller than what is traditionally used for association studies. A recent trend in TS research is to not consider the disorder as a single disease, but as a condition that causes a predisposition 136 to a wide range of diseases caused by the loss of the sex chromosome genes sensitizing the genetic background505,506. This sensitization for developing other diseases is a contributing factor as to why a small sample size is sufficient in these types of studies. As laid out in previous chapters, TS has a larger prevalence of ASD and ADHD than the XX female population, looking more like the XY male population. These developmental disorders share similar deficits in cognition as observed in TS. The exome-wide association study then has the potential to uncover genetic variants that also play a role in these male-biased disorders. Selection of appropriate data analysis approach for Aim 1 Throughout the course of the research process, changes were made to the data analysis plan in Chapter 2 in order to provide the best statistical analysis possible relating to the estimation of heritability. Originally, a DeFries-Fulker regression analysis was run on the neonatal resting-state connectivity data using a Rogers-Kohler method that involved double entering, and then bootstrapping was used to estimate confidence intervals that we then used to gauge statistical significance in broad-sense heritability estimations. This method was abandoned in favor of the mixed effects modeling approach outlined in Chapter 2 that concentrates on narrow-sense heritability. We did this for multiple reasons, the first being that the DeFries-Fulker method produced negative heritability estimates with confidence intervals having both negative values that did not include zero. With the mixed modeling approach, the lower threshold was zero, which not only made more sense biologically but also simplified the results. Second, the mixed modeling approach allowed us to estimate p-values via permutation testing, not just the confidence intervals. This became important because multiple narrow-sense heritability 137 estimates produced confidence intervals that did not contain zero but the p-value was not statistically significant. Switching the methods from a broad-sense to narrow-sense heritability estimation has limitations in the sense that narrow-sense heritability considers only additive genetic effect and does not account for factors such as epistasis, dominance, and gene x environment interactions. Future studies in larger sample sizes are needed to probe these potential effects. Conclusion Taken together, the results presented in this dissertation add to the growing body of literature that explores cognition in both normal and pathological populations. Genetics play a large role in shaping brain connectivity which directly influences cognition. Though resting-state connectivity may not be under a substantial amount of additive genetic effect in neonates, connectivity and cognition are shaped by the presence or absence of an X chromosome. This dissertation provides groundwork for new avenues of discovery building off of the results that have been presented which provide new perspectives on the brain. 138 APPENDICES 139 APPENDIX A: Supplemental data for Chapter 2 140 Table A.1 The 90 regions from the neonate specific AAL atlas assigned to the eight intrinsic functional networks. Hemispher Abbreviatio NO. Region/Node Network e n Somatosenso 1 Precentral gyrus left PreCG-L ry Somatosenso 2 Precentral gyrus right PreCG-R ry Superior frontal 3 (dorsal) left SFGdor-L Default Mode gyrus Superior frontal 4 (dorsal) right SFGdor-R Default Mode gyrus 5 Orbitofrontal cortex (superior) left ORBsup-L Limbic 6 Orbitofrontal cortex (superior) right ORBsup-R Limbic 7 Middle frontal gyrus left MFG-L Frontoparietal 8 Middle frontal gyrus right MFG-R Frontoparietal 9 Orbitofrontal cortex (middle) left ORBmid-L Frontoparietal 10 Orbitofrontal cortex (middle) right ORBmid-R Frontoparietal (opercula 11 Inferior frontal gyrus left IFGoperc-L Frontoparietal r) (opercula 12 Inferior frontal gyrus right IFGoperc-R Frontoparietal r) (triangula 13 Inferior frontal gyrus left IFGtriang-L Frontoparietal r) (triangula 14 Inferior frontal gyrus right IFGtriang-R Frontoparietal r) 15 Orbitofrontal cortex (inferior) left ORBinf-L Default Mode 16 Orbitofrontal cortex (inferior) right ORBinf-R Default Mode Somatosenso 17 Rolandic operculum left ROL-L ry Somatosenso 18 Rolandic operculum right ROL-R ry Supplementary Somatosenso 19 left SMA-L motor area ry Supplementary Somatosenso 20 right SMA-R motor area ry 21 Olfactory left OLF-L Limbic 22 Olfactory right OLF-R Limbic Superior frontal 23 (medial) left SFGmed-L Default Mode gyrus Superior frontal 24 (medial) right SFGmed-R Default Mode gyrus 25 Orbitofrontal cortex (medial) left ORBmed-L Default Mode 26 Orbitofrontal cortex (medial) right ORBmed-R Default Mode 27 Rectus gyrus left REC-L Limbic 141 Table A.1 (cont’d) 28 Rectus gyrus right REC-R Limbic Ventral 29 Insula left INS-L Attention Ventral 30 Insula right INS-R Attention Anterior cingulate 31 left ACG-L Default Mode gyrus Anterior cingulate 32 right ACG-R Default Mode gyrus Middle cingulate Ventral 33 left MCG-L gyrus Attention Middle cingulate Ventral 34 right MCG-R gyrus Attention Posterior cingulate 35 left PCG-L Default Mode gyrus Posterior cingulate 36 right PCG-R Default Mode gyrus 37 Hippocampus left HIP-L Subcortical 38 Hippocampus right HIP-R Subcortical ParaHippocampal 39 left PHG-L Limbic gyrus ParaHippocampal 40 right PHG-R Limbic gyrus 41 Am ygdal a left AMYG-L Subcortical 42 Am ygdal a right AMYG-R Subcortical 43 Calcarine cortex left CAL-L Visual 44 Calcarine cortex right CAL-R Visual 45 Cuneus left CUN-L Visual 46 Cuneus right CUN-R Visual 47 Lingual gyrus left LING-L Visual 48 Lingual gyrus right LING-R Visual Superior occipital 49 left SOG-L Visual gyrus Superior occipital 50 right SOG-R Visual gyrus Middle occipital 51 left MOG-L Visual gyrus Middle occipital 52 right MOG-R Visual gyrus Inferior occipital 53 left IOG-L Visual gyrus Inferior occipital 54 right IOG-R Visual gyrus 55 Fusiform gyrus left FFG-L Visual 142 Table A.1 (cont’d) 56 Fusiform gyrus right FFG-R Visual Somatosenso 57 Postcentral gyrus left PoCG-L ry Somatosenso 58 Postcentral gyrus right PoCG-R ry Superior parietal Dorsal 59 left SPG-L gyrus Attention Superior parieta Dorsal 60 right SPG-R lgyrus Attention Inferior parietal 61 left IPL-L Frontoparietal lobule Inferior parietal 62 right IPL-R Frontoparietal lobule Supramarginal Ventral 63 left SMG-L gyrus Attention Supramarginal Ventral 64 right SMG-R gyrus Attention 65 Angular gyrus left ANG-L Default Mode 66 Angular gyrus right ANG-R Default Mode 67 Precuneus left PCUN-L Default Mode 68 Precuneus right PCUN-R Default Mode Somatosenso 69 Paracentral lobule left PCL-L ry Somatosenso 70 Paracentral lobule right PCL-R ry 71 Caudate left CAU-L Subcortical 72 Caudate right CAU-R Subcortical 73 Putamen left PUT-L Subcortical 74 Putamen right PUT-R Subcortical 75 Pallidum left PAL-L Subcortical 76 Pallidum right PAL-R Subcortical 77 Thalamus left THA-L Subcortical 78 Thalamus right THA-R Subcortical Somatosenso 79 Heschl gyrus left HES-L ry Somatosenso 80 Heschl gyrus right HES-R ry Superior temporal Somatosenso 81 left STG-L gyrus ry Superior temporal Somatosenso 82 right STG-R gyrus ry 83 Temporal pole (superior) left TPOsup-L Limbic 84 Temporal pole (superior) right TPOsup-R Limbic 143 Table A.1 (cont’d) Middle temporal 85 left MTG-L Default Mode gyrus Middle temporal 86 right MTG-R Default Mode gyrus 87 Temporal pole (middle) left TPOmid-L Limbic 88 Temporal pole (middle) right TPOmid-R Limbic Inferior temporal Dorsal 89 left ITG-L gyrus Attention Inferior temporal Dorsal 90 right ITG-R gyrus Attention 144 Table A.2 Differences in demographics and medical history variables between the two cohorts scanned on either the Allegra or Trio MRI for Objective 1. Allegra Trio p Aver Ma Aver Ma Continuous Variables SD Min Continuous Variables SD Min age x age x 2443 44 32 2597 430. 36 Birth weight (g) 1470 Birth weight (g) 1930 0.29 026 3 63 .92 62 50 Gestational age at birth 12. 27 Gestational age at birth 257. 26 253.2 224 7.91 242 0.11 (days) 8 3 (days) 88 8 Gestational age at MRI 293.8 16. 34 Gestational age at MRI 289. 30 248 7.82 278 0.18 (days) 8 31 8 (days) 53 8 0.7 0.61 5 min APGAR score 8.63 4 10 5 min APGAR score 8.84 7 10 0.211 8 2 Maternal education Maternal education 15.56 3.6 6 24 15.3 3.18 9 21 0.86 (years) (years) 3.7 15.9 Paternal education (years) 15.28 6 24 Paternal education (years) 3.26 12 21 0.29 8 2 5.3 33.1 0.000 Maternal age (years) 29.04 16 42 Maternal age (years) 3.93 25 39 6 5 221 6.6 32.6 Paternal age (years) 32.28 20 49 Paternal age (years) 6 22 42 0.538 7 9 Residual framewise 0.0 0.06 0.1 Residual framewise 0.11 0.1 0.112 0.02 0.083 0.402 displacement 23 8 69 displacement 6 54 Categorical variables N % Categorical variables N % 145 Table A.2 (cont’d) 48.0 46.15 Sex Male 48 Sex Male 12 1 0% % Femal 52.0 Femal 53.85 52 14 e 0% e % Vagin 26.0 Vagin 53.85 Delivery Method 26 Delivery Method 14 0.009 al 0% al % C- C- 74.0 46.15 sectio 74 sectio 12 0% % n n Household 30.0 Household 61.54 High 30 High 16 income 0% income % 0.013 22.0 15.38 Mid 22 Mid 4 0% % 48.0 23.08 Low 48 Low 6 0% % 76.0 76.92 Maternal ethnicity White 76 Maternal ethnicity White 20 0% % 1 22.0 23.08 Black 22 Black 6 0% % 2.00 0.00 Asian 2 Asian 0 % % Nativ 0.00 Nativ 0.00 0 0 e % e % 146 Table A.2 (cont’d) Ameri Ameri can can 68.0 53.85 Paternal Ethnicity White 68 Paternal Ethnicity White 14 0% % 1 30.0 30.77 Black 30 Black 8 0% % 2.00 15.38 Asian 2 Asian 4 % % Nativ Nativ e 0.00 e 0.00 0 0 Ameri % Ameri % can can Maternal 80.0 Maternal 69.23 No 80 No 18 psychiatric history 0% psychiatric history % 0.29 20.0 30.77 Yes 20 Yes 8 0% % Paternal 94.0 Paternal 100.0 No 94 No 26 psychiatric history 0% psychiatric history 0% 0.34 6.00 0.00 Yes 6 Yes 0 % % 96.0 100.0 Maternal smoking No 96 Maternal smoking No 26 0% 0% 0.34 147 Table A.2 (cont’d) 4.00 0.00 Yes 4 Yes 0 % % 98.0 100.0 NICU Stay No 98 NICU Stay No 26 0% 0% 1 2.00 0.00 Yes 2 Yes 0 % % 148 Table A.3 Differences in demographics and medical history variables between the two cohorts scanned on either the Allegra or Trio MRI for Objective 2. Allegra Trio p Continuous Averag Continuous Averag SD Min Max SD Min Max Variables e Variables e 650. 482 2761.9 656.5 423 Birth weight (g) 2803.6 1289 Birth weight (g) 840 0.668 9 0 3 4 4 Gestational age at Gestational age at 261.8 17.9 210 295 259.66 13.09 227 285 0.243 birth (days) birth (days) Gestational age at 14.4 Gestational age at 295.56 248 348 291.93 10.16 276 320 0.047 MRI (days) 9 MRI (days) 5 min APGAR score 8.66 384 4 10 5 min APGAR score 8.8 0.59 7 10 0.414 Maternal education Maternal education 15.42 3.47 6 24 15.95 3.36 9 22 0.283 (years) (years) Paternal education Paternal education 14.93 3.47 6 24 16.38 3.22 12 22 0.007 (years) (years) Maternal age Maternal age 1.73E- 28.94 5.72 16 42 32.9 4.59 21 44 (years) (years) 05 Paternal age (years) 32.08 6.94 18 64 Paternal age (years) 33.38 5.63 22 45 0.09 Duration in NICU Duration in NICU 0.093 0.29 0 1 0.1 0.3 0 1 0.962 (days) (days) Residual framewise Residual framewise 0.11 0.02 0.06 0.17 0.12 0.02 0.06 0.19 0.216 displacement displacement Categorical Categorical N % N % variables variables 50.44 Sex Male 114 Sex Male 20 47.62% % 0.866 49.56 Female 112 Female 22 52.38% % Delivery 44.69 Delivery Vaginal 101 Vaginal 19 45.24% Method % Method 1 149 Table A.3 (cont’d) C- 55.31 C- 125 23 54.76% section % section Househol 26.55 Househol 0.0002 High 60 High 8 19.05% d income % d income 6 26.99 Mid 61 Mid 10 23.81% % 46.46 Low 105 Low 24 57.14% % Maternal 76.99 Maternal White 174 White 33 78.57% ethnicity % ethnicity 0.7524 20.80 Black 47 Black 8 19.05% % Asian 2 0.88% Asian 1 2.38% Native Native America 3 1.33% America 0 0.00% n n Paternal 69.47 Paternal White 157 White 27 64.29% Ethnicity % Ethnicity 0.1293 27.88 Black 63 Black 11 26.19% % Asian 5 2.21% Asian 0 0.00% Native Native America 1 0.44% America 4 9.52% n n Maternal Maternal 69.91 psychiatri No 158 psychiatri No 31 73.81% % c history c history 0.713 30.09 Yes 68 Yes 11 26.19% % 150 Table A.3 (cont’d) Paternal Paternal 84.96 psychiatri No 192 psychiatri No 38 90.48% % c history c history 0.713 15.04 Yes 34 Yes 4 9.52% % Maternal 89.38 Maternal 100.00 No 202 No 42 smoking % smoking % 0.019 10.62 Yes 24 Yes 0 0.00% % Gestation 44.25 Gestation Twin 100 Twin 26 61.90% Number % Number 0.011 Singleto 55.75 Singleto 126 16 38.10% n % n NICU 90.71 NICU No 205 No 38 90.48% Stay % Stay 1 Yes 21 9.29% Yes 4 9.52% 151 Table A.4 Narrow-sense heritability estimates for between-network connectivity phenotypes. Network p- h2 SE CI Pair value SS.Vis 0.059315 0.105614 0, 0.165 0.343 DA.Vis 1.50E-08 0.144986 0, 0 0.611 DA.SS 0.104354 0.105552 0, 0.21 0.263 VA.Vis 1.41E-08 0.099494 0, 0 0.627 VA.SS 6.17E-09 0.140848 0, 0 0.776 VA.DA 9.35E-09 0.098091 0, 0 0.689 Lim.Vis 1.79E-08 0.105051 0, 0 0.625 Lim.SS 0.00713 0.110165 0, 0.117 0.467 Lim.DA 2.59E-07 0.110272 0, 0 0.494 Lim.VA 0.106135 0.118269 0, 0.224 0.26 FP.Vis 1.69E-09 0.084142 0, 0 0.994 FP.SS 1.09E-08 0.072328 0, 0 0.67 FP.DA 0.102387 0.113716 0, 0.216 0.241 FP.VA 4.15E-09 0.092033 0, 0 0.847 FP.Lim 7.55E-09 0.102465 0, 0 0.728 DM.Vis 2.53E-09 0.090047 0, 0 0.951 DM.SS 0.095551 0.102864 0, 0.198 0.23 DM.DA 0.06769 0.116867 0, 0.184 0.3 DM.VA 5.81E-07 0.126204 0, 0 0.461 DM.Lim 4.75E-09 0.118395 0, 0 0.833 0.042, DM.FP 0.147621 0.105236 0.162 0.252 0.008, SC.Vis 0.126146 0.117892 0.188 0.244 SC.SS 1.71E-08 0.09408 0, 0 0.604 SC.DA 0.016909 0.117945 0, 0.134 0.445 SC.VA 7.21E-08 0.117867 0, 0 0.484 152 Table A.5. Narrow-sense heritability estimates for within-network connectivity phenotypes. p- Network h2 SE CI value Vis 2.73E-09 0.107542 0, 0 0.94 SS 2.06E-08 0.129447 0, 0 0.566 DA 1.81E-08 0.11841 0, 0 0.565 0, VA 0.081704 0.150855 0.265 0.232 Lim 4.18E-09 0.135226 0, 0 0.867 FP 9.19E-09 0.138587 0, 0 0.678 DM 9.96E-09 0.134392 0, 0 0.675 SC 1.73E-09 0.106999 0, 0 0.994 153 Table A.6 Mixed linear modeling results after backwards elimination for between- network phenotypes. Between-network R- p- pair squared Predictor value Beta r^2 Somatosensory-Visual 0.075 Income 0.1230 Income 6.53E- (Middle) 0.00846 04 1.72E- Income (High) 0.041322 02 Gestational 1.41E- Number 0.1320 0.03515 02 Gestational 3.47E- Age at Birth 0.0647 -0.00158 02 4.46E- Birthweight 0.0358 4.75E-05 02 Gestational 5.20E- Age at MRI 0.0004 0.00239 02 Dorsal Attention - 1.54E- Visual 0.063 Scanner 0.0372 0.0456179 02 1.46E- Gender 0.0470 0.032319 02 Gestational 3.47E- Age at Birth 0.0021 0.001449 02 Maternal Psychiatric 8.20E- History 0.1300 -0.026034 03 Dorsal Attention- Method of 1.19E- Somatosensory 0.049 Delivery 0.0847 -0.02859 02 Gestational 6.63E- Age at Birth 0.0106 -0.00195 02 2.89E- Birthweight 0.0834 3.41E-05 02 Maternal Education at 4.31E- Enrollment 0.0245 -0.007836 02 Paternal Education at 3.33E- Enrollment 0.0493 0.006866 02 Ventral Attention- 3.46E- Visual 0.083 Scanner 0.0009 0.067067 02 1.54E- Gender 0.0292 -0.03257 02 Gestational 8.60E- Age at Birth 0.0026 -0.002236 02 3.18E- Birthweight 0.0528 3.60E-05 02 154 Table A.6 (cont’d) Gestational 2.12E- Age at MRI 0.0156 0.00137 02 Ventral Attention - 1.08E- Somatosensory 0.011 Scanner 0.0818 0.6579 02 Ventral Attention - 1.24E- Dorsal Attention 0.066 Gender 0.0607 -0.02818 02 Gestational 1.04E- Age at Birth 0.0009 -0.002366 01 2.42E- Birthweight 0.1000 3.03E-05 02 Residual Framewise 1.48E- Displacement 0.0368 0.64719 02 Paternal Limbic - Visual 0.045 Ethnicity 0.3030 Paternal Ethnicity 1.00E- (Black) 0.0283 02 Paternal Ethnicity (Native 3.59E- American) 0.12467 03 Paternal Ethnicity 3.29E- (Asian) 0.040412 03 Maternal Education at 2.31E- Enrollment 0.1030 0.005592 02 Paternal Education at 1.61E- Enrollment 0.1670 -0.0046566 02 Paternal Psychiatric 1.39E- History 0.0520 0.04249 02 Residual Framewise 1.40E- Displacement 0.0498 -0.63119 02 Limbic - Gestational 2.66E- Somatosensory 0.04 Age at MRI 0.0083 -0.00147 02 Maternal Psychiatric 1.84E- History 0.0272 0.03674 02 Limbic - Dorsal Gestational 8.57E- Attention 0.052 Age at Birth 0.0026 -0.00204 02 155 Table A.6 (cont’d) 2.29E- Birthweight 0.1190 2.79E-05 02 Paternal Ethnicity 0.1690 Paternal Ethnicity 1.78E- (Black) 0.0358 02 Paternal Ethnicity (Native 2.59E- American) 0.0003167 08 Paternal Ethnicity 1.58E- (Asian) 0.02645 03 Limbic - Ventral 9.90E- Attention 0.048 Scanner 0.1040 -0.03387 03 2.44E- Gender 0.0112 -0.03866 02 Gestational 8.53E- Age at MRI 0.1320 -0.0008229 03 Maternal Psychiatric 1.24E- History 0.0665 0.02956 02 Frontoparietal - Visual 0.043 Income 0.1820 Income 3.46E- (Middle) 0.0005633 06 8.96E- Income (High) -0.02727 03 Maternal Psychiatric 2.36E- History 0.0059 -0.0441 02 Residual Framewise 1.31E- Displacement 0.0428 -0.6435 02 Frontoparietal - 9.64E- Somatosensory 0.067 Gender 0.0816 0.02694 03 Gestational 4.01E- Age at Birth 0.0389 0.001597 02 2.83E- Birthweight 0.0700 -3.56E-05 02 Gestational 2.38E- Age at MRI 0.0131 -0.001519 02 Maternal Psychiatric 3.23E- History 0.0018 0.05302 02 156 Table A.6 (cont’d) Frontoparietal - Dorsal 9.04E- Attention 0.04 Scanner 0.1080 -0.03138 03 Gestational 3.18E- Age at Birth 0.0874 -0.00124 02 2.85E- Birthweight 0.0785 3.12E-05 02 1.18E- Stay in NICU 0.1250 0.044723 02 Paternal Psychiatric 8.19E- History 0.1480 -0.0308 03 Frontoparietal - Ventral 1.80E- Attention 0.059 Scanner 0.0151 -0.04911 02 Maternal Ethnicity 0.4590 Maternal Ethnicity 7.25E- (Black) 0.028 03 Maternal Ethnicity (Native 1.60E- American) 0.0505 03 Maternal Ethnicity 1.92E- (Asian) 0.01755 04 Paternal Education at 1.48E- Enrollment 0.0409 0.0004675 02 Maternal Psychiatric 3.66E- History 0.0012 0.0547388 02 Frontoparietal - Limbic 0.051 Income 0.0778 Income 2.56E- (Middle) 0.01653 03 1.98E- Income (High) 0.04373 02 1.32E- Scanner 0.0337 -0.0455 02 2.76E- Gender 0.0028 -0.04787 02 Maternal Default Mode - Visual 0.037 Ethnicity 0.1290 Maternal Ethnicity 1.69E- (Black) 0.030902 02 157 Table A.6 (cont’d) Maternal Ethnicity (Native 1.51E- American) 0.03547 03 Maternal Ethnicity 1.47E- (Asian) -0.003501 05 Paternal Age 1.33E- at Enrollment 0.0421 0.00164 02 Paternal Psychiatric 8.40E- History 0.1170 -0.0249 03 Default Mode - Method of 1.19E- Somatosensory 0.02 Delivery 0.5770 0.007276 03 Maternal Psychiatric 1.80E- History 0.0279 0.03033 02 Default Mode - Dorsal 3.86E- Attention 0.056 Stay in NICU 0.0016 0.062188 02 Maternal Age 3.46E- at Enrollment 0.0177 0.002989 02 Paternal Age 2.27E- at Enrollment 0.0520 -0.0020569 02 Paternal Psychiatric 7.37E- History 0.1600 -0.0224075 03 Default Mode - Ventral Method of 1.01E- Attention 0.014 Delivery 0.1160 -0.0187952 02 Gestational - 1.02E- Age at Birth 0.1070 0.00054467 02 Gestational 4.64E- Default Mode - Limbic 0.04 Number 0.2710 0.01468 03 1.82E- Scanner 0.0240 -0.03949 02 APGAR 5- 1.22E- minute 0.0875 0.01449 02 7.67E- Stay in NICU 0.1680 0.032023 03 Maternal Ethnicty 0.4440 Maternal Ethnicity 1.18E- (Black) 0.00904 03 Maternal 1.11E- Ethnicity 0.03367 03 158 Table A.6 (cont’d) (Native American) Maternal Ethnicity 9.52E- (Asian) -0.09862 03 Paternal Ethnicity 0.7810 Paternal Ethnicity 4.85E- (Black) -0.01655 03 Paternal Ethnicity (Native 5.52E- American) 0.001295 07 Paternal Ethnicity 1.84 e- (Asian) 0.0253288 3 Gestational 8.22E- Age at MRI 1.0000 -2.19E-07 10 Default Mode - 4.28E- Frontoparietal 0.109 Scanner 0.2510 -0.0195 02 Gestational 4.90E- Age at Birth 0.0004 -0.001395 02 Maternal Ethnicity 0.0063 Maternal Ethnicity 1.43E- (Black) -0.032082 02 Maternal Ethnicity (Native 4.10E- American) -0.20876 02 Maternal Ethnicity 3.49E- (Asian) -0.060884 03 Paternal Ethnicity 0.3710 Paternal Ethnicity 3.77E- (Black) 0.0148856 03 Paternal Ethnicity (Native 5.97E- American) 0.13739 03 159 Table A.6 (cont’d) Paternal Ethnicity 5.00E- (Asian) 0.04255 03 Gestational 2.89E- Age at MRI 0.0062 0.001324 02 Subcortical - Visual None Selected - - - Subcortical - 7.81E- Somatosensory 0.114 Smoking 0.1430 -0.04617 03 Gestational 1.65E- Number 0.0313 -0.03891 02 APGAR 5- 2.48E- minute 0.0096 -0.029024 02 1.50E- Stay in NICU 0.0539 -0.06275 02 Maternal Age 2.23E- at Enrollment 0.0585 0.00389 02 Paternal Age 3.04E- at Enrollment 0.0161 -0.00385 02 Paternal Education at 8.95E- Enrollment 0.1130 -0.004077 03 Gestational 3.15E- Age at MRI 0.0015 -0.0019 02 Paternal Psychiatric 1.65E- History 0.0296 0.054333 02 Residual Framewise 9.75E- Displacement 0.0774 -0.6194 03 Subcortical - Dorsal 9.95E- Attention 0.0354 Stay in NICU 0.0978 0.0451012 03 Maternal Education at 2.84E- Enrollment 0.0776 -0.006418 02 Paternal Education at 2.97E- Enrollment 0.0716 0.0065465 02 Motion 8.15E- Correction 0.1380 0.49936 03 Subcortical - Ventral 1.41E- Attention 0.038 Scanner 0.0281 -0.05022 02 4.07E- Gender 0.7280 -0.00621 04 Gestational 1.73E- Age at Birth 0.4710 0.00037198 03 160 Table A.6 (cont’d) Paternal Psychiatric 1.08E- History 0.0820 0.045381 02 Residual Framewise 8.15E- Displacement 0.1130 -0.597 03 Gestational 2.96E- Subcortical - Limbic 0.036 Number 0.0259 -0.05495 02 1.04E- Gender 0.0874 -0.03218 02 2.28E- Birthweight 0.0512 -3.65E-05 02 Paternal Age 2.58E- at Enrollment 0.3970 0.0011855 03 Residual Framewise 5.36E- Displacement 0.2190 0.48468 03 Subcortical - Frontoparietal 0.027 Income 0.3290 Income 1.68E- (Middle) 0.004623 04 1.00E- Income (High) 0.034023 02 Gestational 9.43E- Number 0.1340 -0.031004 03 Maternal Psychiatric 8.37E- History 0.1490 0.03095 03 Subcortical - Default Mode 0.09 Income 0.1800 Income 9.73E- (Middle) 0.02719 03 1.05E- Income (High) 0.026916 02 1.76E- Smoking 0.0263 -2.84E-05 02 2.88E- Scanner 0.0048 -0.05677 02 9.93E- Gender 0.0889 -0.02424 03 2.29E- Birthweight 0.0120 -2.84E-05 02 APGAR 5- 1.26E- minute 0.0528 0.01749 02 161 Table A.6 (cont’d) Paternal Psychiatric 1.03E- History 0.0833 0.03503 02 162 Table A.7 Mixed linear modeling results after backwards elimination for between-network phenotype. R- p- Network squared Predictor value Beta r^2 7.09E- 2.69E- Visual 0.078 Gestational Number 03 0.05935 02 8.93E- 8.12E- Scanner 02 -0.0443 03 1.21E- 7.54E- Method of Delivery 01 -0.0312 03 1.02E- 2.24E- NICU Stay 02 0.09195 02 Gestational Age at 4.14E- 1.27E- MRI 02 0.00144 02 Residual Framewise 2.44E- 3.05E- Displacement 03 1.312 02 6.14E- 1.26E- Somatosensory 0.052 Smoking 02 0.0765 02 Paternal Age at 3.39E- 1.57E- Enrollment 02 0.003609 02 2.75E- Paternal Ethnicity 01 Paternal Ethnicity 1.34E- (Black) 0.05033 02 Paternal Ethnicity (Native 2.80E- American) 0.00533 06 Paternal Ethnicity 1.12E- (Asian) 0.0361 03 Paternal Psychiatric 2.21E- 1.88E- History 02 0.0756 02 Dorsal 1.59E- 8.47E- Attention 0.1 Gestational Number 04 0.118 02 Gestational Age at 1.96E- 3.10E- Birth 02 0.00204 02 Maternal Education at 2.01E- 4.56E- Enrollment 02 0.0123 02 Paternal Education at 4.92E- 1.08E- Enrollment 04 -0.0188 01 163 Table A.7 (cont’d) Residual Framewise 1.33E- 2.13E- Displacement 02 1.22 02 Ventral 1.44E- 7.73E- Attention 0.052 Scanner 01 0.0426 03 1.11E- 9.37E- Gender 02 0.0341 01 1.19E- 2.30E- Method of Delivery 02 -0.0537 02 2.98E- Paternal Ethnicity 01 Paternal Ethnicity 5.04E- (Black) 0.0279 03 Paternal Ethnicity (Native 9.88E- American) 0.0287 05 Paternal Ethnicity 1.00E- (Asian) 0.0977 02 7.10E- 1.22E- Limbic 0.056 Gender 02 -0.0364 02 1.44E- Maternal Ethnicity 01 Maternal Ethnicity 9.36E- (Black) -0.0394 03 Maternal Ethnicity (Native 3.51E- American) -0.0029 06 Maternal Ethnicity 1.14E- (Asian) -0.0167 02 Paternal Age at 8.06E- 2.67E- Enrollment 03 -0.00399 02 2.80E- 7.09E- Frontoparietal 0.017 Gestational Number 01 -0.02859 03 6.34E- 1.41E- Gender 02 -0.0398 02 Gestational Age at 2.82E- - 7.20E- Birth 01 0.000826 03 1.31E- 8.20E- Default Mode 0.008 Scanner 01 0.025 03 1.33E- 3.74E- Subcortical 0.028 Gestational Number 02 -0.0766 02 1.89E- 6.18E- Gender 01 0.03075 03 9.96E- 1.69E- Birthweight 02 -3.91 02 164 APPENDIX B: Supplemental data for Chapter 3 165 Table B.1 The 46 tracts that passed quality control with global FDR p-values given, along with local p-values in the estimation of axial diffusivity (AD), radial diffusivity (RD), and fractional anisotropy (FA). Global_Omnibu Global_FD s R AD RD FA Arcuate, left, frontoparietal Female control, Male control 0.086 0.16483 0.163 0.23 0.064 Female control, Turner syndrome 0.489 0.5442097 0.751 0.475 0.129 Male control, Turner syndrome 0.138 0.226714 0.238 0.229 0.192 Arcuate, left, frontotemporal Female control, Male control 0.004 0.02208 0 0.557 0.505 Female control, Turner syndrome 0.536 0.5870476 0.39 0.499 0.939 Male control, Turner syndrome 0.328 0.4066726 0.398 0.363 0.169 Arcuate, left, temporoparietal Female control, Male control 0.0001 0.00092 0.098 0 0.051 Female control, Turner syndrome 0.02 0.069 0.25 0.113 0.056 Male control, Turner syndrome 0.046 0.1075932 0.015 0.15 0.3 Arcuate, right, fronotparietal Female control, Male control 0.076 0.1518 0.137 0.029 0.358 166 Table B.1 (cont’d) Female control, Turner syndrome 0.057 0.1191818 0.103 0.418 0.145 Male control, Turner syndrome 0.123 0.2139 0.03 0.063 0.056 Arcuate right, frontotemporal Female control, Male control 0.026 0.0763404 0.062 0.013 0.255 Female control, Turner syndrome 0.007 0.0333103 0.009 0.224 0.078 Male control, Turner syndrome 0.387 0.4557479 0.329 0.725 0.319 Cingulate gyrus, left Female control, Male control 0.137 0.2267143 0.103 0.193 0.51 Female control, Turner syndrome 0.39 0.4557479 0.287 0.513 0.715 Male control, Turner syndrome 0.387 0.4557479 0.78 0.441 0.569 Cingulate gyrus, right Female control, Male control 0.023 0.0721364 0.031 0.046 0.332 Female control, Turner syndrome 0.047 0.1081 0.052 0.25 0.539 Male control, Turner syndrome 0.708 0.7291343 0.501 0.879 0.977 Cingulum adjoining hippocampus, left Female control, Male control 0.002 0.0131429 0.154 0.007 0.078 Female control, Turner syndrome 0.018 0.0671351 0.24 0.009 0.035 167 Table B.1 (cont’d) Male control, Turner syndrome 0.22 0.31625 0.43 0.399 0.058 Cingulum adjoining hippocampus, right Female control, Male control 0.0001 0.00092 0.004 0 0.526 Female control, Turner syndrome 0.0001 0.00092 0.001 0.002 0.197 Male control, Turner syndrome 0.466 0.5228293 0.608 0.87 0.899 Corpus callosum, body Female control, Male control 0.001 0.0072632 0.004 0.144 0.495 Female control, Turner syndrome 0.159 0.2493409 0.217 0.631 0.466 Male control, Turner syndrome 0.386 0.4557479 0.168 0.91 0.723 Corpus callosum, genu Female control, Male control 0.005 0.0265385 0.004 0.106 0.074 Female control, Turner syndrome 0.188 0.2882667 0.627 0.378 0.249 Male control, Turner syndrome 0.333 0.4066726 0.318 0.639 0.893 Corpus callosum, motor Female control, Male control 0.013 0.0543636 0.099 0.863 0.641 Female control, Turner syndrome 0.007 0.0333103 0.027 0.426 0.261 168 Table B.1 (cont’d) Male control, Turner syndrome 0.607 0.6443538 0.419 0.584 0.477 Corpus callosum, parietal Female control, Male control 0.0001 0.00092 0 0.007 0.112 Female control, Turner syndrome 0.052 0.1104 0.173 0.088 0.06 Male control, Turner syndrome 0.656 0.6858182 0.59 0.381 0.145 Corpus callosum, premotor Female control, Male control 0.148 0.2402824 0.334 0.384 0.176 Female control, Turner syndrome 0.199 0.2952903 0.067 0.861 0.361 Male control, Turner syndrome 0.157 0.2490345 0.074 0.83 0.614 Corpus callosum, rostrum Female control, Male control 0.023 0.0721364 0.148 0.386 0.023 Female control, Turner syndrome 0.108 0.1910769 0.08 0.33 0.121 Male control, Turner syndrome 0.445 0.51175 0.037 0.685 0.339 Corpus callosum, splenium Female control, Male control 0.009 0.0388125 0.033 0.042 0.087 Female control, Turner syndrome 0.021 0.0706829 0.016 0.013 0.022 169 Table B.1 (cont’d) Male control, Turner syndrome 0.328 0.4066726 0.766 0.367 0.107 Corpus callosum, tapetum Female control, Male control 0.312 0.4038716 0.84 0.941 0.104 Female control, Turner syndrome 0.004 0.02208 0.042 0.241 0.011 Male control, Turner syndrome 0.051 0.1104 0.178 0.505 0.09 Corticofugal, left, motor Female control, Male control 0.0001 0.00092 0.0001 0.0001 0.654 Female control, Turner syndrome 0.0001 0.00092 0.0001 0.0001 0.021 Male control, Turner syndrome 0.0001 0.00092 0.004 0.087 0.0001 Corticofugal, left, parietal Female control, Male control 0.007 0.0333103 0.055 0.012 0.233 Female control, Turner syndrome 0.099 0.1774286 0.128 0.383 0.105 Male control, Turner syndrome 0.002 0.0131429 0.097 0.029 0.02 Corticofugal, right, motor Female control, Male control 0.033 0.0843333 0.015 0.231 0.517 Female control, Turner syndrome 0.045 0.107069 0.089 0.811 0.176 170 Table B.1 (cont’d) Male control, Turner syndrome 0.008 0.0356129 0.013 0.35 0.002 Corticofugal, right, parietal Female control, Male control 0.023 0.0721364 0.048 0.058 0.472 Female control, Turner syndrome 0.263 0.3537087 0.205 0.42 0.833 Male control, Turner syndrome 0.02 0.069 0.103 0.861 0.034 Corticoreticular, left Female control, Male control 0.0001 0.00092 0.001 0 0.351 Female control, Turner syndrome 0.062 0.1277015 0.374 0.086 0.016 Male control, Turner syndrome 0.463 0.5228293 0.188 0.946 0.285 Corticoreticular, right Female control, Male control 0.052 0.1104 0.059 0.057 0.352 Female control, Turner syndrome 0.051 0.1104 0.072 0.164 0.19 Male control, Turner syndrome 0.314 0.4038716 0.432 0.509 0.132 Corticospinal, right Female control, Male control 0.003 0.018 0.03 0.054 0.094 Female control, Turner syndrome 0.352 0.4261053 0.163 0.849 0.567 Male control, Turner syndrome 0.019 0.069 0.006 0.212 0.387 171 Table B.1 (cont’d) Corticothalamic, left, motor Female control, Male control 0.319 0.4038716 0.29 0.241 0.204 Female control, Turner syndrome 0.319 0.4038716 0.29 0.241 0.204 Male control, Turner syndrome 0.853 0.8592263 0.892 0.962 0.882 Corticothalamic, left, parietal Female control, Male control 0.0001 0.00092 0.009 0.005 0.192 Female control, Turner syndrome 0.688 0.7138647 0.506 0.745 0.367 Male control, Turner syndrome 0.218 0.31625 0.308 0.774 0.141 Corticothalamic, left, prefrontal Female control, Male control 0.042 0.1016842 0.097 0.133 0.16 Female control, Turner syndrome 0.083 0.1613239 0.067 0.137 0.1 Male control, Turner syndrome 0.241 0.3359394 0.056 0.211 0.506 Corticothalamic, left, premotor Female control, Male control 0.0001 0.00092 0 0.07 0.227 Female control, Turner syndrome 0.036 0.0887143 0.081 0.047 0.037 Male control, Turner syndrome 0.638 0.6720916 0.325 0.971 0.761 172 Table B.1 (cont’d) Corticothalamic, left, superior Female control, Male control 0.0001 0.00092 0 0.019 0.206 Female control, Turner syndrome 0.193 0.2926813 0.599 0.196 0.133 Male control, Turner syndrome 0.845 0.8574265 0.539 0.8 0.901 Corticothalamic, right, motor Female control, Male control 0.027 0.077625 0.01 0.619 0.039 Female control, Turner syndrome 0.031 0.0822692 0.008 0.405 0.167 Male control, Turner syndrome 0.871 0.871 0.892 0.865 0.911 Corticothalamic, right, parietal Female control, Male control 0.0001 0.00092 0.003 0.001 0.176 Female control, Turner syndrome 0.393 0.4557479 0.343 0.224 0.592 Male control, Turner syndrome 0.136 0.2267143 0.102 0.275 0.054 Corticothalamic, right, prefrontal Female control, Male control 0.241 0.3359394 0.263 0.497 0.594 Female control, Turner syndrome 0.281 0.3728654 0.131 0.618 0.7 Male control, Turner syndrome 0.29 0.3811429 0.454 0.255 0.963 173 Table B.1 (cont’d) Corticothalamic, right, premotor Female control, Male control 0.0001 0.00092 0 0.338 0.008 Female control, Turner syndrome 0.003 0.018 0.005 0.311 0.129 Male control, Turner syndrome 0.452 0.5155041 0.319 0.574 0.626 Corticothalamic, right, superior Female control, Male control 0.0001 0.00092 0.001 0.507 0.042 Female control, Turner syndrome 0.025 0.075 0.019 0.435 0.251 Male control, Turner syndrome 0.507 0.559728 0.318 0.573 0.957 Fornix, left Female control, Male control 0.015 0.0591429 0.101 0.037 0.011 Female control, Turner syndrome 0.014 0.0568235 0.055 0.054 0.064 Male control, Turner syndrome 0.229 0.3257938 0.161 0.13 0.363 Fornix, right Female control, Male control 0.172 0.2666966 0.435 0.17 0.252 Female control, Turner syndrome 0.57 0.6145313 0.403 0.767 0.941 Male control, Turner syndrome 0.554 0.6019843 0.588 0.637 0.726 Inferior fronto-occiptal fasciculus, left 174 Table B.1 (cont’d) Female control, Male control 0.029 0.08004 0.075 0.378 0.077 Female control, Turner syndrome 0.032 0.0833208 0.088 0.14 0.046 Male control, Turner syndrome 0.001 0.0072632 0.003 0.066 0.012 Inferior fronto-occipital fasciculus, right Female control, Male control 0.089 0.1682466 0.127 0.11 0.126 Female control, Turner syndrome 0.153 0.2455116 0.184 0.772 0.275 Male control, Turner syndrome 0.264 0.3537087 0.181 0.229 0.251 Inferior longitudinal fasciculus, left Female control, Male control 0.017 0.0651667 0.045 0.801 0.313 Female control, Turner syndrome 0.001 0.0072632 0.201 0.044 0.039 Male control, Turner syndrome 0.028 0.0788571 0.339 0.247 0.073 Inferior longitudinal fasciculus, right Female control, Male control 0.031 0.0822692 0.106 0.031 0.139 Female control, Turner syndrome 0.05 0.1104 0.279 0.132 0.072 Male control, Turner syndrome 0.124 0.2139 0.589 0.118 0.273 Optic tract, right 175 Table B.1 (cont’d) Female control, Male control 0.024 0.0736 0.098 0.084 0.018 Female control, Turner syndrome 0.035 0.0878182 0.406 0.211 0.388 Male control, Turner syndrome 0.008 0.0356129 0.145 0.06 0.007 Optic radiation, left Female control, Male control 0.001 0.0072632 0.1 0 0.001 Female control, Turner syndrome 0.095 0.1748 0.935 0.033 0.031 Male control, Turner syndrome 0.246 0.33948 0.502 0.169 0.352 Optic radiation, right Female control, Male control 0.0001 0.00092 0.001 0.005 0.048 Female control, Turner syndrome 0.072 0.1461176 0.065 0.013 0.188 Male control, Turner syndrome 0.806 0.8239111 0.55 0.928 0.528 Superior longitudinal fasciculus, left Female control, Male control 0.099 0.1774286 0.068 0.31 0.654 Female control, Turner syndrome 0.585 0.625814 0.427 0.515 0.296 Male control, Turner syndrome 0.092 0.1715676 0.212 0.077 0.205 Uncinate, left Female control, Male control 0.197 0.2952903 0.14 0.45 0.136 176 Table B.1 (cont’d) Female control, Turner syndrome 0.25 0.3415842 0.177 0.763 0.154 Male control, Turner syndrome 0.077 0.1518 0.057 0.067 0.182 Uncinate, right Female control, Male control 0.13 0.2214815 0.027 0.669 0.039 Female control, Turner syndrome 0.21 0.3082979 0.094 0.471 0.202 Male control, Turner syndrome 0.332 0.4066726 0.231 0.443 0.467 177 Figure B.1 Model of DTI results for the tapetum portion of the corpus callosum for measures of axial diffusivity. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 178 Figure B.2 Model of DTI results for the left corticofugal tract for measures of fractional anisotropy. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 179 Figure B.3 Model of DTI results for the left motor corticofugal tract for measures of axial diffusivity. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 180 Figure B.4 Model of DTI results for the left motor corticofugal tract for measures of radial diffusivity. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 181 Figure B.5 Model of DTI results for the right optic tract for measures of fractional anisotropy. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 182 Figure B.6 Model of DTI results for right frontotemporal region of the arcuate fasciculus for measures of axial diffusivity. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 183 Figure B.7 Model of DTI results for the cingulum adjoining the hippocampus for measures of axial diffusivity. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 184 Figure B.8 Model of DTI results for the cingulum adjoining the hippocampus for measures of radial diffusivity. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 185 Figure B.9 Model of DTI results for the motor bundle of the corpus callosum for measures of axial diffusivity. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 186 Figure B.10 Model of DTI results for the left inferior fronto-occipital fasciculus for measures of fractional anisotropy. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 187 Figure B.11 Model of DTI results for the inferior longitudinal fasciculus for measures of fractional anisotropy. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 188 Figure B.12 Model of DTI results for the tapetum region of the corpus callosum for measures of fractional anisotropy. In each panel, the leftmost graph shows the beta value over the arclength of the fasciculus, with areas of local statistical significance highlighted in grey. Below that is the percent change in the beta value over time for that tract in the specified comparison for the specific diffusivity measure. The middlemost image in each panel shows an overlay of the p-values on the fasciculus, with regions of statistical significance showing a color other than yellow. Finally, the rightmost panel shows the beta values from the specific comparison with the model of the tract. 189 APPENDIX C: Supplemental data for Chapter 4 190 Figure C.1 Density plot of reaction time in milliseconds for the Mental Rotation Task between the three groups. Figure C.2 Density plot of reaction time in milliseconds after log transformation for the Mental Rotation Task between the three groups. 191 Figure C.3 Density plot of accuracy for the Mental Rotation Task between the three groups. Figure C.4 Density plot of reaction time (ms) for the Flanker Task for congruent responses between the three groups. 192 Figure C.5 Density plot of reaction time (ms) after log transformation for congruent responses for the Flanker Task between the three groups. Figure C.6 Density plot of reaction time (ms) for the Flanker Task for incongruent responses between the three groups. 193 Figure C.7 Density plot of reaction time (ms) after log transformation for incongruent responses for the Flanker Task between the three groups. Figure C.8 Density plot of accuracy for Reading the Mind in the Eyes between the three groups. 194 Figure C.9 Density plot of accuracy for the Continuous Performance Task between the three groups before outlier removal. Figure C.10 Density plot of accuracy for the Continuous Performance Task between the three groups after outlier removal. 195 Figure C.11 Density plot of reaction time (ms) for the Continuous Performance Task between the three groups before outlier removal. Figure C.12 Density plot of reaction time (ms) for the Continuous Performance Task between the three groups after outlier removal. 196 Figure C.13 Density plot of reaction time (ms) for the Continuous Performance Task for errors of commission between the three groups before outlier removal. Figure C.14 Density plot of reaction time (ms) for the Continuous Performance Task for errors of commission between the three groups after outlier removal. 197 Figure C.15 Density plot of reaction time (ms) for the Continuous Performance Task for errors of omission between the three groups before outlier removal. Figure C.16 Density plot of reaction time (ms) for the Continuous Performance Task for errors of omission between the three groups after outlier removal. 198 Figure C.17 Density plot of the maximum blocks repeated for the Corsi Block Task between the three groups before outlier removal. Figure C.18 Density plot of the maximum blocks repeated for the Corsi Block Task between the three groups after outlier removal. 199 Figure C.19 Density plot of the maximum number length repeated for the Corsi Block Task between the three groups. Figure C. 20 Density plot of the reaction time in milliseconds on the Simple Response Time Task between the three groups before log transformation. 200 Figure C.21 Density plot of the reaction time in milliseconds on the Simple Response Time Task between the three groups after log transformation. Figure C.22 Density plot of the accuracy on the Autism Spectrum Quotient between the three groups. 201 Figure C.23 Density plot of the score on the ADHD report scale between the three groups before outlier removal. Figure C.24 Density plot of the score on the ADHD report scale between the three groups after outlier removal. 202 REFERENCES 203 REFERENCES 1. Logothetis, N. K. The neural basis of the blood-oxygen-level-dependent functional magnetic resonance imaging signal. Philos Trans R Soc Lond B Biol Sci 357, 1003–1037 (2002). 2. Biswal, B., Yetkin, F. Z., Haughton, V. M. & Hyde, J. S. Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magn Reson Med 34, 537–541 (1995). 3. Biswal, B. B., Kylen, J. V. & Hyde, J. S. Simultaneous assessment of flow and BOLD signals in resting-state functional connectivity maps. NMR in Biomedicine 10, 165–170 (1997). 4. Raichle, M. E. et al. A default mode of brain function. Proc Natl Acad Sci U S A 98, 676–682 (2001). 5. Fox, M. D. et al. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proceedings of the National Academy of Sciences 102, 9673–9678 (2005). 6. Bzdok, D. et al. Characterization of the temporo-parietal junction by combining data-driven parcellation, complementary connectivity analyses, and functional decoding. NeuroImage 81, 381–392 (2013). 7. Spreng, R. N., Mar, R. A. & Kim, A. S. N. The Common Neural Basis of Autobiographical Memory, Prospection, Navigation, Theory of Mind, and the Default Mode: A Quantitative Meta-analysis. Journal of Cognitive Neuroscience 21, 489–510 (2009). 8. Herlin, B., Navarro, V. & Dupont, S. The temporal pole: From anatomy to function—A literature appraisal. Journal of Chemical Neuroanatomy 113, 101925 (2021). 9. Kringelbach, M. L. The human orbitofrontal cortex: linking reward to hedonic experience. Nat Rev Neurosci 6, 691–702 (2005). 10. Marek, S. & Dosenbach, N. U. F. The frontoparietal network: function, electrophysiology, and importance of individual precision mapping. Dialogues Clin Neurosci 20, 133–140 (2018). 11. Vossel, S., Geng, J. J. & Fink, G. R. Dorsal and Ventral Attention Systems. Neuroscientist 20, 150–159 (2014). 12. Le Bihan, D. et al. Diffusion tensor imaging: concepts and applications. J Magn Reson Imaging 13, 534–546 (2001). 204 13. Jütten, K. et al. Diffusion Tensor Imaging Reveals Microstructural Heterogeneity of Normal-Appearing White Matter and Related Cognitive Dysfunction in Glioma Patients. Front. Oncol. 9, (2019). 14. Winklewski, P. J. et al. Understanding the Physiopathology Behind Axial and Radial Diffusivity Changes—What Do We Know? Front. Neurol. 9, (2018). 15. Ivanova, M. V., Zhong, A., Turken, A., Baldo, J. V. & Dronkers, N. F. Functional Contributions of the Arcuate Fasciculus to Language Processing. Frontiers in Human Neuroscience 15, (2021). 16. Barbey, A. K. et al. An integrative architecture for general intelligence and executive function revealed by lesion mapping. Brain 135, 1154–1164 (2012). 17. De Benedictis, A., Efisio Marras, C., Petit, L. & Sarubbo, S. The inferior fronto- occipital fascicle: a century of controversies from anatomy theaters to operative neurosurgery. Journal of Neurosurgical Sciences (2021). 18. Koshiyama, D. et al. Association between the superior longitudinal fasciculus and perceptual organization and working memory: A diffusion tensor imaging study. Neuroscience Letters 738, 135349 (2020). 19. Urger, S. E. et al. The Superior Longitudinal Fasciculus in Typically Developing Children and Adolescents: Diffusion Tensor Imaging and Neuropsychological Correlates. J Child Neurol 30, 9–20 (2015). 20. Herbet, G., Zemmoura, I. & Duffau, H. Functional Anatomy of the Inferior Longitudinal Fasciculus: From Historical Reports to Current Hypotheses. Front. Neuroanat. 0, (2018). 21. Bubb, E. J., Metzler-Baddeley, C. & Aggleton, J. P. The cingulum bundle: Anatomy, function, and dysfunction. Neurosci Biobehav Rev 92, 104–127 (2018). 22. Benear, S. L., Ngo, C. T. & Olson, I. R. Dissecting the Fornix in Basic Memory Processes and Neuropsychiatric Disease: A Review. Brain Connect 10, 331–354 (2020). 23. Goldstein, A., Covington, B. P., Mahabadi, N. & Mesfin, F. B. Neuroanatomy, Corpus Callosum. in StatPearls (StatPearls Publishing, 2022). 24. Natali, A. L., Reddy, V. & Bordoni, B. Neuroanatomy, Corticospinal Cord Tract. in StatPearls (StatPearls Publishing, 2022). 25. Nuñez, A. & Malmierca, E. Corticofugal modulation of sensory information. Adv Anat Embryol Cell Biol 187, 1 p following table of contents, 1–74 (2007). 26. Jang, S. H. & Lee, S. J. Corticoreticular Tract in the Human Brain: A Mini Review. Frontiers in Neurology 10, (2019). 205 27. Diamond, A. Executive Functions. Annu Rev Psychol 64, 135–168 (2013). 28. Lehto, J. E., Juujarvi, P., Kooistra, L. & Pulkkinen, L. Dimensions of executive functioning: Evidence from children. The British Journal of Developmental Psychology 21, 59 (2003). 29. Eriksen, B. A. & Eriksen, C. W. Effects of noise letters upon the identification of a target letter in a nonsearch task. Perception & Psychophysics 16, 143–149 (1974). 30. Stroop, J. R. Studies of interference in serial verbal reactions. Journal of Experimental Psychology 18, 643–662 (1935). 31. Kessels, R. P., van Zandvoort, M. J., Postma, A., Kappelle, L. J. & de Haan, E. H. The Corsi Block-Tapping Task: standardization and normative data. Appl Neuropsychol 7, 252–258 (2000). 32. Wechsler, D. The measurement of adult intelligence. ix, 226 (Williams & Wilkins Co, 1939). doi:10.1037/10020-000. 33. Insel, T. et al. Research domain criteria (RDoC): Toward a new classification framework for research on mental disorders. The American Journal of Psychiatry 167, 748–751 (2010). 34. Martel, M., Nikolas, M. & Nigg, J. T. Executive Function in Adolescents With ADHD. Journal of the American Academy of Child & Adolescent Psychiatry 46, 1437–1444 (2007). 35. Holmes, J. et al. The Diagnostic Utility of Executive Function Assessments in the Identification of ADHD in Children. Child and Adolescent Mental Health 15, 37–43 (2010). 36. Johnson, J. & Reid, R. Overcoming Executive Function Deficits With Students With ADHD. Theory Into Practice 50, 61–67 (2011). 37. Channon, S. & Green, P. S. S. Executive function in depression: the role of performance strategies in aiding depressed and non-depressed participants. Journal of Neurology, Neurosurgery & Psychiatry 66, 162–171 (1999). 38. Cotrena, C., Branco, L. D., Ponsoni, A., Shansis, F. M. & Fonseca, R. P. Neuropsychological Clustering in Bipolar and Major Depressive Disorder. Journal of the International Neuropsychological Society 23, 584–593 (2017). 39. Velligan, D. I. & Bow-Thomas, C. C. Executive function in schizophrenia. Semin Clin Neuropsychiatry 4, 24–33 (1999). 40. Eisenberg, D. P. & Berman, K. F. Executive Function, Neural Circuitry, and Genetic Mechanisms in Schizophrenia. Neuropsychopharmacol 35, 258–277 (2010). 206 41. Wobrock, T. et al. Cognitive impairment of executive function as a core symptom of schizophrenia. The World Journal of Biological Psychiatry 10, 442–451 (2009). 42. Demetriou, E. A. et al. Autism spectrum disorders: a meta-analysis of executive function. Mol Psychiatry 23, 1198–1204 (2018). 43. Demetriou, E. A., DeMayo, M. M. & Guastella, A. J. Executive Function in Autism Spectrum Disorder: History, Theoretical Models, Empirical Findings, and Potential as an Endophenotype. Front. Psychiatry 0, (2019). 44. Johnston, K., Murray, K., Spain, D., Walker, I. & Russell, A. Executive Function: Cognition and Behaviour in Adults with Autism Spectrum Disorders (ASD). Journal of Autism and Developmental Disorders 49, 4181–4192 (2019). 45. Danielson, M. L. et al. Prevalence of Parent-Reported ADHD Diagnosis and Associated Treatment Among U.S. Children and Adolescents, 2016. J Clin Child Adolesc Psychol 47, 199–212 (2018). 46. Diagnostic and statistical manual of mental disorders: DSM-5TM, 5th ed. xliv, 947 (American Psychiatric Publishing, Inc., 2013). doi:10.1176/appi.books.9780890425596. 47. Brown, T. E. ADD/ADHD and impaired executive function in clinical practice. Curr Atten Disord Rep 1, 37–41 (2009). 48. Yuan, P. & Raz, N. Prefrontal Cortex and Executive Functions in Healthy Adults: A Meta-Analysis of Structural Neuroimaging Studies. Neurosci Biobehav Rev 0, 180– 192 (2014). 49. Corbetta, M. & Shulman, G. L. Spatial Neglect and Attention Networks. Annual Review of Neuroscience 34, 569–599 (2011). 50. Menon, V. & D’Esposito, M. The role of PFC networks in cognitive control and executive function. Neuropsychopharmacol. 47, 90–103 (2022). 51. Aron, A. R., Robbins, T. W. & Poldrack, R. A. Inhibition and the right inferior frontal cortex: one decade on. Trends in Cognitive Sciences 18, 177–185 (2014). 52. Wallis, G., Stokes, M., Cousijn, H., Woolrich, M. & Nobre, A. C. Frontoparietal and Cingulo-opercular Networks Play Dissociable Roles in Control of Working Memory. Journal of Cognitive Neuroscience 27, 2019–2034 (2015). 53. Todd, J. J., Fougnie, D. & Marois, R. Visual Short-Term Memory Load Suppresses Temporo-Parietal Junction Activity and Induces Inattentional Blindness. Psychol Sci 16, 965–972 (2005). 207 54. Shulman, G. L. et al. Interaction of Stimulus-Driven Reorienting and Expectation in Ventral and Dorsal Frontoparietal and Basal Ganglia-Cortical Networks. J. Neurosci. 29, 4392–4407 (2009). 55. Rapado-Castro, M. et al. Fronto-Parietal Gray Matter Volume Loss Is Associated with Decreased Working Memory Performance in Adolescents with a First Episode of Psychosis. Journal of Clinical Medicine 10, 3929 (2021). 56. Haldane, M., Cunningham, G., Androutsos, C. & Frangou, S. Structural brain correlates of response inhibition in Bipolar Disorder I. J Psychopharmacol 22, 138– 143 (2008). 57. Collette, F. et al. Involvement of both prefrontal and inferior parietal cortex in dual- task performance. Cognitive Brain Research 24, 237–251 (2005). 58. Barch, D. M. & Csernansky, J. G. Abnormal Parietal Cortex Activation During Working Memory in Schizophrenia: Verbal Phonological Coding Disturbances Versus Domain-General Executive Dysfunction. AJP 164, 1090–1098 (2007). 59. Olesen, P. J., Westerberg, H. & Klingberg, T. Increased prefrontal and parietal activity after training of working memory. Nat Neurosci 7, 75–79 (2004). 60. Jubault, T., Ody, C. & Koechlin, E. Serial Organization of Human Behavior in the Inferior Parietal Cortex. J. Neurosci. 27, 11028–11036 (2007). 61. Hahn, B., Robinson, B. M., Leonard, C. J., Luck, S. J. & Gold, J. M. Posterior Parietal Cortex Dysfunction Is Central to Working Memory Storage and Broad Cognitive Deficits in Schizophrenia. J. Neurosci. 38, 8378–8387 (2018). 62. Shomstein, S. Cognitive functions of the posterior parietal cortex: top-down and bottom-up attentional control. Frontiers in Integrative Neuroscience 6, (2012). 63. Esterman, M., Chiu, Y.-C., Tamber-Rosenau, B. J. & Yantis, S. Decoding cognitive control in human parietal cortex. PNAS 106, 17974–17979 (2009). 64. Cole, M. W., Pathak, S. & Schneider, W. Identifying the brain’s most globally connected regions. Neuroimage 49, 3132–3148 (2010). 65. Cole, M. W., Yarkoni, T., Repovš, G., Anticevic, A. & Braver, T. S. Global Connectivity of Prefrontal Cortex Predicts Cognitive Control and Intelligence. J Neurosci 32, 8988–8999 (2012). 66. Reineberg, A. E., Andrews-Hanna, J. R., Depue, B. E., Friedman, N. P. & Banich, M. T. Resting-state networks predict individual differences in common and specific aspects of executive function. NeuroImage 104, 69–78 (2015). 67. Cole, M. W. et al. Multi-task connectivity reveals flexible hubs for adaptive task control. Nat. Neurosci. 16, 1348–1355 (2013). 208 68. Anticevic, A. et al. The Role of Default Network Deactivation in Cognition and Disease. Trends Cogn Sci 16, 584–592 (2012). 69. Jia, W. et al. Disruptions of frontoparietal control network and default mode network linking the metacognitive deficits with clinical symptoms in schizophrenia. Human Brain Mapping 41, 1445–1458 (2020). 70. Schneider, F. C. et al. Modulation of the default mode network is task-dependant in chronic schizophrenia patients. Schizophr Res 125, 110–117 (2011). 71. Dreher, J.-C. et al. Common and differential pathophysiological features accompany comparable cognitive impairments in medication-free patients with schizophrenia and in healthy aging subjects. Biol Psychiatry 71, 890–897 (2012). 72. Whitfield-Gabrieli, S. et al. Hyperactivity and hyperconnectivity of the default network in schizophrenia and in first-degree relatives of persons with schizophrenia. Proc Natl Acad Sci U S A 106, 1279–1284 (2009). 73. Anticevic, A., Repovs, G. & Barch, D. M. Working memory encoding and maintenance deficits in schizophrenia: neural evidence for activation and deactivation abnormalities. Schizophr Bull 39, 168–178 (2013). 74. Frye, R. E. et al. Superior longitudinal fasciculus and cognitive dysfunction in adolescents born preterm and at term. Developmental Medicine & Child Neurology 52, 760–766 (2010). 75. Sasson, N. et al. Controlling for Response Biases Clarifies Sex and Age Differences in Facial Affect Recognition. Journal of Nonverbal Behavior 34, 207– 221 (2010). 76. Vestergaard, M. et al. White matter microstructure in superior longitudinal fasciculus associated with spatial working memory performance in children. J Cogn Neurosci 23, 2135–2146 (2011). 77. Sasson, E., Doniger, G. M., Pasternak, O., Tarrasch, R. & Assaf, Y. White matter correlates of cognitive domains in normal aging with diffusion tensor imaging. Front. Neurosci. 0, (2013). 78. Karlsgodt, K. H. et al. Diffusion Tensor Imaging of the Superior Longitudinal Fasciculus and Working Memory in Recent-Onset Schizophrenia. Biological Psychiatry 63, 512–518 (2008). 79. Fitzgerald, J. et al. Abnormal fronto-parietal white matter organisation in the superior longitudinal fasciculus branches in autism spectrum disorders. European Journal of Neuroscience 47, 652–661 (2018). 80. Kinoshita, M. et al. Chronic spatial working memory deficit associated with the superior longitudinal fasciculus: a study using voxel-based lesion-symptom 209 mapping and intraoperative direct stimulation in right prefrontal glioma surgery. Journal of Neurosurgery 125, 1024–1032 (2016). 81. Cristofori, I. et al. White and gray matter contributions to executive function recovery after traumatic brain injury. Neurology 84, 1394–1401 (2015). 82. Arain, M. et al. Maturation of the adolescent brain. Neuropsychiatr Dis Treat 9, 449–461 (2013). 83. Gogtay, N. et al. Dynamic mapping of human cortical development during childhood through early adulthood. PNAS 101, 8174–8179 (2004). 84. Hansen, P. E., Ballesteros, M. C., Soila, K., Garcia, L. & Howard, J. M. MR imaging of the developing human brain. Part 1. Prenatal development. RadioGraphics 13, 21–36 (1993). 85. Cholfin, J. A. & Rubenstein, J. L. R. Patterning of frontal cortex subdivisions by Fgf17. Proc Natl Acad Sci U S A 104, 7652–7657 (2007). 86. Cholfin, J. A. & Rubenstein, J. L. R. Frontal Cortex Subdivision Patterning Is Coordinately Regulated by Fgf8, Fgf17, and Emx2. J Comp Neurol 509, 144–155 (2008). 87. Fuster, J. M. Frontal lobe and cognitive development. J Neurocytol 31, 373–385 (2002). 88. Kostović, I. Chapter 10 Structural and histochemical reorganization of the human prefrontal cortex during perinatal and postnatal life. in Progress in Brain Research (eds. Uylings, H. B. M., Van Eden, C. G., De Bruin, J. P. C., Corner, M. A. & Feenstra, M. G. P.) vol. 85 223–240 (Elsevier, 1991). 89. Ruff, H. A. & Rothbart, M. K. Attention in Early Development: Themes and Variations. (Oxford University Press, 2001). 90. Hendry, A., Jones, E. J. H. & Charman, T. Executive function in the first three years of life: Precursors, predictors and patterns. Developmental Review 42, 1–33 (2016). 91. Reznick, J. S., Morrow, J. D., Goldman, B. D. & Snyder, J. The Onset of Working Memory in Infants. Infancy 6, 145–154 (2004). 92. van Houdt, C. A., Oosterlaan, J., van Wassenaer-Leemhuis, A. G., van Kaam, A. H. & Aarnoudse-Moens, C. S. H. Executive function deficits in children born preterm or at low birthweight: a meta-analysis. Developmental Medicine & Child Neurology 61, 1015–1024 (2019). 93. Taylor, H. G. & Clark, C. A. C. Executive function in children born preterm: Risk factors and implications for outcome. Seminars in perinatology 40, 520 (2016). 210 94. Power, J., van IJzendoorn, M., Lewis, A. J., Chen, W. & Galbally, M. Maternal perinatal depression and child executive function: A systematic review and meta- analysis. Journal of Affective Disorders 291, 218–234 (2021). 95. DiPietro, J. A. Maternal Stress in Pregnancy: Considerations for Fetal Development. Journal of Adolescent Health 51, S3–S8 (2012). 96. Teffer, K. & Semendeferi, K. Chapter 9 - Human prefrontal cortex: Evolution, development, and pathology. in Progress in Brain Research (eds. Hofman, M. A. & Falk, D.) vol. 195 191–218 (Elsevier, 2012). 97. Giedd, J. N. et al. Brain development during childhood and adolescence: a longitudinal MRI study. Nat Neurosci 2, 861–863 (1999). 98. Giedd, J. N. et al. Puberty-related influences on brain development. Molecular and Cellular Endocrinology 254–255, 154–162 (2006). 99. Fair, D. A. et al. Development of distinct control networks through segregation and integration. PNAS 104, 13507–13512 (2007). 100. Gao, W., Alcauter, S., Smith, J. K., Gilmore, J. H. & Lin, W. Development of human brain cortical network architecture during infancy. Brain Struct Funct 220, 1173– 1186 (2015). 101. Gao, W. et al. Temporal and Spatial Evolution of Brain Network Topology during the First Two Years of Life. PLOS ONE 6, e25278 (2011). 102. Huang, H. et al. Anatomical Characterization of Human Fetal Brain Development with Diffusion Tensor Magnetic Resonance Imaging. J. Neurosci. 29, 4263–4273 (2009). 103. Horgos, B. et al. White Matter Dissection of the Fetal Brain. Frontiers in Neuroanatomy 14, 65 (2020). 104. Hermoye, L. et al. Pediatric diffusion tensor imaging: Normal database and observation of the white matter maturation in early childhood. NeuroImage 29, 493–504 (2006). 105. Zhang, J. et al. Evidence of Slow Maturation of the Superior Longitudinal Fasciculus in Early Childhood by Diffusion Tensor Imaging. Neuroimage 38, 239– 247 (2007). 106. Tau, G. Z. & Peterson, B. S. Normal Development of Brain Circuits. Neuropsychopharmacology 35, 147–168 (2010). 107. Friedman, N. P. et al. Individual Differences in Executive Functions Are Almost Entirely Genetic in Origin. J Exp Psychol Gen 137, 201–225 (2008). 211 108. Engelhardt, L. E., Briley, D. A., Mann, F. D., Harden, K. P. & Tucker-Drob, E. M. Genes Unite Executive Functions in Childhood. Psychol Sci 26, 1151–1163 (2015). 109. Harden, K. P. et al. Genetic Associations Between Executive Functions and a General Factor of Psychopathology. Journal of the American Academy of Child & Adolescent Psychiatry 59, 749–758 (2020). 110. Ando, J., Ono, Y. & Wright, M. J. Genetic structure of spatial and verbal working memory. Behav Genet 31, 615–624 (2001). 111. Anokhin, A. P., Heath, A. C. & Myers, E. Genetics, prefrontal cortex, and cognitive control: a twin study of event-related brain potentials in a response inhibition task. Neurosci Lett 368, 314–318 (2004). 112. Chang, S., Yang, L., Wang, Y. & Faraone, S. V. Shared polygenic risk for ADHD, executive dysfunction and other psychiatric disorders. Transl Psychiatry 10, 1–9 (2020). 113. Owens, S. F. et al. Genetic overlap between schizophrenia and selective components of executive function. Schizophrenia Research 127, 181–187 (2011). 114. Chen, L.-S., Rice, T. K., Thompson, P. A., Barch, D. M. & Csernansky, J. G. Familial aggregation of clinical and neurocognitive features in sibling pairs with and without schizophrenia. Schizophrenia Research 111, 159–166 (2009). 115. Toulopoulou, T. et al. Substantial Genetic Overlap Between Neurocognition and Schizophrenia: Genetic Modeling in Twin Samples. Archives of General Psychiatry 64, 1348–1355 (2007). 116. Berlin, L. & Bohlin, G. Response Inhibition, Hyperactivity, and Conduct Problems Among Preschool Children. Journal of Clinical Child & Adolescent Psychology 31, 242–251 (2002). 117. Carlson, S. M. & Moses, L. J. Individual Differences in Inhibitory Control and Children’s Theory of Mind. Child Development 72, 1032–1053 (2001). 118. Vuontela, V. et al. Audiospatial and Visuospatial Working Memory in 6–13 Year Old School Children. Learn. Mem. 10, 74–81 (2003). 119. Gaillard, A., Fehring, D. J. & Rossell, S. L. A systematic review and meta-analysis of behavioural sex differences in executive control. European Journal of Neuroscience 53, 519–542 (2021). 120. Polanczyk, G., de Lima, M. S., Horta, B. L., Biederman, J. & Rohde, L. A. The Worldwide Prevalence of ADHD: A Systematic Review and Metaregression Analysis. AJP 164, 942–948 (2007). 212 121. Rowland, A. S. et al. The Prevalence of ADHD in a Population-Based Sample. J Atten Disord 19, 741–754 (2015). 122. Danielson, M. L. et al. Prevalence of Parent-Reported ADHD Diagnosis and Associated Treatment Among U.S. Children and Adolescents, 2016. J Clin Child Adolesc Psychol 47, 199–212 (2018). 123. Cuffe, S. P., Moore, C. G. & McKeown, R. E. Prevalence and Correlates of ADHD Symptoms in the National Health Interview Survey. J Atten Disord 9, 392–401 (2005). 124. Gaub, M. & Carlson, C. L. Gender differences in ADHD: a meta-analysis and critical review. J Am Acad Child Adolesc Psychiatry 36, 1036–1045 (1997). 125. Willcutt, E. G. The Prevalence of DSM-IV Attention-Deficit/Hyperactivity Disorder: A Meta-Analytic Review. Neurotherapeutics 9, 490–499 (2012). 126. Biederman, J. et al. Clinical Correlates of ADHD in Females: Findings From a Large Group of Girls Ascertained From Pediatric and Psychiatric Referral Sources. Journal of the American Academy of Child & Adolescent Psychiatry 38, 966–975 (1999). 127. Quinn, P. O. Attention-deficit/hyperactivity disorder and its comorbidities in women and girls: An evolving picture. Curr Psychiatry Rep 10, 419–423 (2008). 128. Loyer Carbonneau, M., Demers, M., Bigras, M. & Guay, M.-C. Meta-Analysis of Sex Differences in ADHD Symptoms and Associated Cognitive Deficits. J Atten Disord 1087054720923736 (2020) doi:10.1177/1087054720923736. 129. Martin, J. et al. Sex-specific manifestation of genetic risk for attention deficit hyperactivity disorder in the general population. Journal of Child Psychology and Psychiatry 59, 908–916 (2018). 130. Ottosen, C. et al. Sex Differences in Comorbidity Patterns of Attention- Deficit/Hyperactivity Disorder. Journal of the American Academy of Child & Adolescent Psychiatry 58, 412-422.e3 (2019). 131. Mowlem, F. D. et al. Sex differences in predicting ADHD clinical diagnosis and pharmacological treatment. Eur Child Adolesc Psychiatry 28, 481–489 (2019). 132. Mowlem, F., Agnew-Blais, J., Taylor, E. & Asherson, P. Do different factors influence whether girls versus boys meet ADHD diagnostic criteria? Sex differences among children with high ADHD symptoms. Psychiatry Research 272, 765–773 (2019). 133. Higgins, E. T. & Bargh, J. A. Social Cognition and Social Perception. Annual Review of Psychology 38, 369–425 (1987). 213 134. Renfrew, C., Frith, C., Malafouris, L. & Frith, C. D. Social cognition. Philosophical Transactions of the Royal Society B: Biological Sciences 363, 2033–2039 (2008). 135. Senju, A. Atypical development of spontaneous social cognition in autism spectrum disorders. Brain and Development 35, 96–101 (2013). 136. Korkmaz, B. Theory of Mind and Neurodevelopmental Disorders of Childhood. Pediatr Res 69, 101–108 (2011). 137. Baron-Cohen, S., Wheelwright, S., Hill, J., Raste, Y. & Plumb, I. The “Reading the Mind in the Eyes” Test Revised Version: A Study with Normal Adults, and Adults with Asperger Syndrome or High-functioning Autism. The Journal of Child Psychology and Psychiatry and Allied Disciplines 42, 241–251 (2001). 138. Gur, R. C. et al. A method for obtaining 3-dimensional facial expressions and its standardization for use in neurocognitive studies. J Neurosci Methods 115, 137– 143 (2002). 139. Pinkham, A. E., Harvey, P. D. & Penn, D. L. Social Cognition Psychometric Evaluation: Results of the Final Validation Study. Schizophr Bull 44, 737–748 (2018). 140. Baksh, R. A., Abrahams, S., Auyeung, B. & MacPherson, S. E. The Edinburgh Social Cognition Test (ESCoT): Examining the effects of age on a new measure of theory of mind and social norm understanding. PLOS ONE 13, e0195818 (2018). 141. McDonald, S., Flanagan, S., Rollins, J. & Kinch, J. TASIT: A new clinical tool for assessing social perception after traumatic brain injury. J Head Trauma Rehabil 18, 219–238 (2003). 142. Isaksson, J. et al. Social Cognition in Autism and Other Neurodevelopmental Disorders: A Co-twin Control Study. J Autism Dev Disord 49, 2838–2848 (2019). 143. Bishop-Fitzpatrick, L., Mazefsky, C. A., Eack, S. M. & Minshew, N. J. Correlates of social functioning in autism spectrum disorder: The role of social cognition. Research in Autism Spectrum Disorders 35, 25–34 (2017). 144. Shattuck, P. T. et al. Change in Autism Symptoms and Maladaptive Behaviors in Adolescents and Adults with an Autism Spectrum Disorder. J Autism Dev Disord 37, 1735–1747 (2007). 145. Addington, J., Girard, T. A., Christensen, B. K. & Addington, D. Social cognition mediates illness-related and cognitive influences on social function in patients with schizophrenia-spectrum disorders. J Psychiatry Neurosci 35, 49–54 (2010). 146. Middelboe, T. et al. The Nordic Study on schizophrenic patients living in the community. Subjective needs and perceived help. European Psychiatry 16, 207– 214 (2001). 214 147. Boada, L. et al. Social Cognition in Autism and Schizophrenia Spectrum Disorders: The Same but Different? J Autism Dev Disord 50, 3046–3059 (2020). 148. Crespi, B. & Badcock, C. Psychosis and autism as diametrical disorders of the social brain. Behavioral and Brain Sciences 31, 241–261 (2008). 149. Roberts, D. L. et al. A randomized, controlled trial of Social Cognition and Interaction Training (SCIT) for outpatients with schizophrenia spectrum disorders. British Journal of Clinical Psychology 53, 281–298 (2014). 150. Coursey, R. D., Keller, A. B. & Farrell, E. W. Individual Psychotherapy and Persons With Serious Mental Illness: The Clients’ Perspective. Schizophrenia Bulletin 21, 283–301 (1995). 151. Bora, E. & Pantelis, C. Meta-analysis of social cognition in attention- deficit/hyperactivity disorder (ADHD): comparison with healthy controls and autistic spectrum disorder. Psychological Medicine 46, 699–716 (2016). 152. Uekermann, J. et al. Social cognition in attention-deficit hyperactivity disorder (ADHD). Neuroscience & Biobehavioral Reviews 34, 734–743 (2010). 153. Jusyte, A., Gulewitsch, M. D. & Schönenberg, M. Recognition of peer emotions in children with ADHD: Evidence from an animated facial expressions task. Psychiatry Research 258, 351–357 (2017). 154. Sibley, M. H., Evans, S. W. & Serpell, Z. N. Social Cognition and Interpersonal Impairment in Young Adolescents with ADHD. J Psychopathol Behav Assess 32, 193–202 (2010). 155. Parke, E. M. et al. Social Cognition in Children With ADHD. J Atten Disord 25, 519–529 (2021). 156. Gallagher, H. L. et al. Reading the mind in cartoons and stories: an fMRI study of ‘theory of mind’ in verbal and nonverbal tasks. Neuropsychologia 38, 11–21 (2000). 157. Happé, F. et al. ‘Theory of mind’ in the brain. Evidence from a PET scan study of Asperger syndrome. NeuroReport 8, 197–201 (1996). 158. Fletcher, P. C. et al. Other minds in the brain: a functional imaging study of “theory of mind” in story comprehension. Cognition 57, 109–128 (1995). 159. Birbaumer, N. et al. fMRI reveals amygdala activation to human faces in social phobics. NeuroReport 9, 1223–1226 (1998). 160. Hart, A. J. et al. Differential response in the human amygdala to racial outgroup vs ingroup face stimuli. NeuroReport 11, 2351–2354 (2000). 215 161. Baron-Cohen, S. et al. Social intelligence in the normal and autistic brain: an fMRI study. European Journal of Neuroscience 11, 1891–1898 (1999). 162. Beer, J. S., John, O. P., Scabini, D. & Knight, R. T. Orbitofrontal cortex and social behavior: integrating self-monitoring and emotion-cognition interactions. J Cogn Neurosci 18, 871–879 (2006). 163. Jonker, F. A., Jonker, C., Scheltens, P. & Scherder, E. J. A. The role of the orbitofrontal cortex in cognition and behavior. Rev Neurosci 26, 1–11 (2015). 164. Cicerone, K. D. & Tanenbaum, L. N. Disturbance of social cognition after traumatic orbitofrontal brain injury. Archives of Clinical Neuropsychology 12, 173–188 (1997). 165. Amodio, D. M. & Frith, C. D. Meeting of minds: the medial frontal cortex and social cognition. Nat Rev Neurosci 7, 268–277 (2006). 166. Santiesteban, I., Banissy, M. J., Catmur, C. & Bird, G. Enhancing Social Ability by Stimulating Right Temporoparietal Junction. Current Biology 22, 2274–2277 (2012). 167. McAlonan, G. M. et al. Mapping the brain in autism. A voxel-based MRI study of volumetric differences and intercorrelations in autism. Brain 128, 268–276 (2005). 168. Bzdok, D. et al. Left inferior parietal lobe engagement in social cognition and language. Neuroscience & Biobehavioral Reviews 68, 319–334 (2016). 169. Cheng, Y., Chou, K.-H., Fan, Y.-T. & Lin, C.-P. ANS: Aberrant Neurodevelopment of the Social Cognition Network in Adolescents with Autism Spectrum Disorders. PLOS ONE 6, e18905 (2011). 170. Li, W., Mai, X. & Liu, C. The default mode network and social understanding of others: what do brain connectivity studies tell us. Front Hum Neurosci 8, 74 (2014). 171. Wang, K. et al. Altered social cognition and connectivity of default mode networks in the co-occurrence of autistic spectrum disorder and attention deficit hyperactivity disorder. Aust N Z J Psychiatry 53, 760–771 (2019). 172. Sutcubasi, B. et al. Resting-state network dysconnectivity in ADHD: A system- neuroscience-based meta-analysis. The World Journal of Biological Psychiatry 21, 662–672 (2020). 173. Samson, D., Apperly, I. A., Chiavarino, C. & Humphreys, G. W. Left temporoparietal junction is necessary for representing someone else’s belief. Nat Neurosci 7, 499–500 (2004). 174. Jackson, P. L., Brunet, E., Meltzoff, A. N. & Decety, J. Empathy examined through the neural mechanisms involved in imagining how I feel versus how you feel pain. Neuropsychologia 44, 752–761 (2006). 216 175. Young, L., Camprodon, J. A., Hauser, M., Pascual-Leone, A. & Saxe, R. Disruption of the right temporoparietal junction with transcranial magnetic stimulation reduces the role of beliefs in moral judgments. Proc Natl Acad Sci U S A 107, 6753–6758 (2010). 176. Mars, R. B., Sallet, J., Neubert, F.-X. & Rushworth, M. F. S. Connectivity profiles reveal the relationship between brain areas for social cognition in human and monkey temporoparietal cortex. PNAS 110, 10806–10811 (2013). 177. Penner, J. et al. Temporoparietal Junction Functional Connectivity in Early Schizophrenia and Major Depressive Disorder. Chronic Stress 2, 2470547018815232 (2018). 178. Poeppl, T. B. et al. Imbalance in subregional connectivity of the right temporoparietal junction in major depression. Human Brain Mapping 37, 2931– 2942 (2016). 179. Wang, Y., Metoki, A., Alm, K. H. & Olson, I. R. White matter pathways and social cognition. Neuroscience & Biobehavioral Reviews 90, 350–370 (2018). 180. Fletcher, P. T. et al. Microstructural connectivity of the arcuate fasciculus in adolescents with high-functioning autism. Neuroimage 51, 1117–1125 (2010). 181. Barnea-Goraly, N., Lotspeich, L. J. & Reiss, A. L. Similar white matter aberrations in children with autism and their unaffected siblings: a diffusion tensor imaging study using tract-based spatial statistics. Arch Gen Psychiatry 67, 1052–1060 (2010). 182. Fishman, I., Datko, M., Cabrera, Y., Carper, R. A. & Müller, R.-A. Reduced integration and differentiation of the imitation network in autism: A combined functional connectivity magnetic resonance imaging and diffusion-weighted imaging study. Annals of Neurology 78, 958–969 (2015). 183. Im, W. Y. et al. Impaired White Matter Integrity and Social Cognition in High- Function Autism: Diffusion Tensor Imaging Study. Psychiatry Investig 15, 292–299 (2018). 184. Lo, Y.-C., Chen, Y.-J., Hsu, Y.-C., Tseng, W.-Y. I. & Gau, S. S.-F. Reduced tract integrity of the model for social communication is a neural substrate of social communication deficits in autism spectrum disorder. Journal of Child Psychology and Psychiatry 58, 576–585 (2017). 185. Noriuchi, M. et al. Altered white matter fractional anisotropy and social impairment in children with autism spectrum disorder. Brain Research 1362, 141–149 (2010). 186. Miyata, J. et al. Reduced white matter integrity as a neural correlate of social cognition deficits in schizophrenia. Schizophrenia Research 119, 232–239 (2010). 217 187. Kana, R. K., Libero, L. E., Hu, C. P., Deshpande, H. D. & Colburn, J. S. Functional Brain Networks and White Matter Underlying Theory-of-Mind in Autism. Social Cognitive and Affective Neuroscience 9, 98–105 (2014). 188. Stiles, J. & Jernigan, T. L. The Basics of Brain Development. Neuropsychol Rev 20, 327–348 (2010). 189. Prayer, D. et al. MRI of normal fetal brain development. European Journal of Radiology 57, 199–216 (2006). 190. Grossmann, T. & Johnson, M. H. The development of the social brain in human infancy. European Journal of Neuroscience 25, 909–919 (2007). 191. Li, G. et al. Mapping Region-Specific Longitudinal Cortical Surface Expansion from Birth to 2 Years of Age. Cereb Cortex 23, 2724–2733 (2013). 192. Nelson, C. A. The Recognition of Facial Expressions in the First Two Years of Life: Mechanisms of Development. Child Development 58, 889–909 (1987). 193. Liszkowski, U., Carpenter, M., Striano, T. & Tomasello, M. 12- and 18-Month-Olds Point to Provide Information for Others. Journal of Cognition and Development 7, 173–187 (2006). 194. Slaughter, V. Theory of Mind in Infants and Young Children: A Review. Australian Psychologist 50, 169–172 (2015). 195. Rodger, H., Vizioli, L., Ouyang, X. & Caldara, R. Mapping the development of facial expression recognition. Developmental Science 18, 926–939 (2015). 196. Wellman, H. M., Cross, D. & Watson, J. Meta-Analysis of Theory-of-Mind Development: The Truth about False Belief. Child Development 72, 655–684 (2001). 197. Wellman, H. M. & Liu, D. Scaling of Theory-of-Mind Tasks. Child Development 75, 523–541 (2004). 198. Bowman, L. C., Thorpe, S. G., Cannon, E. N. & Fox, N. A. Action mechanisms for social cognition: behavioral and neural correlates of developing Theory of Mind. Developmental Science 20, e12447 (2017). 199. Mitter, C., Prayer, D., Brugger, P. C., Weber, M. & Kasprian, G. In Vivo Tractography of Fetal Association Fibers. PLOS ONE 10, e0119536 (2015). 200. Huang, H. et al. White and gray matter development in human fetal, newborn and pediatric brains. NeuroImage 33, 27–38 (2006). 218 201. Cohen, A. H. et al. Development of human white matter fiber pathways: From newborn to adult ages. International Journal of Developmental Neuroscience 50, 26–38 (2016). 202. Turk, E. et al. Functional Connectome of the Fetal Brain. J. Neurosci. 39, 9716– 9724 (2019). 203. Hoff, G. E. A.-J., Van Den Heuvel, M., Benders, M. J. N. L., Kersbergen, K. J. & de Vries, L. S. On development of functional brain connectivity in the young brain. Frontiers in Human Neuroscience 7, 650 (2013). 204. Collin, G. & van den Heuvel, M. P. The Ontogeny of the Human Connectome: Development and Dynamic Changes of Brain Connectivity Across the Life Span. Neuroscientist 19, 616–628 (2013). 205. Thomason, M. E. et al. Intrinsic functional brain architecture derived from graph theoretical analysis in the human fetus. PLoS ONE 9, e94423 (2014). 206. Thomason, M. E. et al. Age-related increases in long-range connectivity in fetal functional neural connectivity networks in utero. Dev Cogn Neurosci 11, 96–104 (2015). 207. van den Heuvel, M. I. & Thomason, M. E. Functional Connectivity of the Human Brain in Utero. Trends in Cognitive Sciences 20, 931–939 (2016). 208. Fransson, P. et al. Spontaneous Brain Activity in the Newborn Brain During Natural Sleep—An fMRI Study in Infants Born at Full Term. Pediatric Research 66, 301– 305 (2009). 209. Gao, W. et al. Evidence on the emergence of the brain’s default network from 2- week-old to 2-year-old healthy pediatric subjects. PNAS (2009) doi:10.1073/pnas.0811221106. 210. Fair, D. A. et al. The maturing architecture of the brain’s default network. PNAS 105, 4028–4032 (2008). 211. Davis, M. C. et al. Associations between oxytocin receptor genotypes and social cognitive performance in individuals with schizophrenia. Schizophrenia Research 159, 353–357 (2014). 212. Barlati, S. et al. Social Cognition in a Research Domain Criteria Perspective: A Bridge Between Schizophrenia and Autism Spectra Disorders. Frontiers in Psychiatry 11, (2020). 213. Wilczyński, K. M., Siwiec, A. & Janas-Kozik, M. Systematic Review of Literature on Single-Nucleotide Polymorphisms Within the Oxytocin and Vasopressin Receptor Genes in the Development of Social Cognition Dysfunctions in Individuals Suffering From Autism Spectrum Disorder. Front Psychiatry 10, 380 (2019). 219 214. Reuter, M. et al. Functional characterization of an oxytocin receptor gene variant (rs2268498) previously associated with social cognition by expression analysis in vitro and in human brain biopsy. Social Neuroscience 12, 604–611 (2017). 215. Robinson, E. B. et al. Genetic risk for autism spectrum disorders and neuropsychiatric variation in the general population. Nat Genet 48, 552–555 (2016). 216. Micalizzi, L. et al. Single nucleotide polymorphism heritability and differential patterns of genetic overlap between inattention and four neurocognitive factors in youth. Dev Psychopathol 33, 76–86 (2021). 217. Coleman, J. R. I. et al. Genome-wide association study of facial emotion recognition in children and association with polygenic risk for mental health disorders. Am J Med Genet B Neuropsychiatr Genet 174, 701–711 (2017). 218. St Pourcain, B. et al. Common variation contributes to the genetic architecture of social communication traits. Mol Autism 4, 34 (2013). 219. St Pourcain, B. et al. Variability in the common genetic architecture of social- communication spectrum phenotypes during childhood and adolescence. Mol Autism 5, 18 (2014). 220. Scourfield, J., Martin, N., Lewis, G. & McGuffin, P. Heritability of social cognitive skills in children and adolescents. The British Journal of Psychiatry 175, 559–564 (1999). 221. Robinson, E. B. et al. The genetic architecture of pediatric cognitive abilities in the Philadelphia Neurodevelopmental Cohort. Mol Psychiatry 20, 454–458 (2015). 222. Germine, L. et al. Association between polygenic risk for schizophrenia, neurocognition and social cognition across development. Transl Psychiatry 6, e924–e924 (2016). 223. Russell, T. A., Tchanturia, K., Rahman, Q. & Schmidt, U. Sex differences in theory of mind: A male advantage on Happé’s “cartoon” task. Cognition and Emotion 21, 1554–1564 (2007). 224. Proverbio, A. M., Matarazzo, S., Brignone, V., Zotto, M. D. & Zani, A. Processing valence and intensity of infant expressions: The roles of expertise and gender. Scandinavian Journal of Psychology 48, 477–485 (2007). 225. Proverbio, A. M. Sex differences in social cognition: The case of face processing. Journal of Neuroscience Research 95, 222–234 (2017). 226. Hall, J. K., Hutton, S. B. & Morgan, M. J. Sex differences in scanning faces: Does attention to the eyes explain female superiority in facial expression recognition? Cognition and Emotion 24, 629–637 (2010). 220 227. Hall, J. A. Gender effects in decoding nonverbal cues. Psychological Bulletin 85, 845–857 (1978). 228. Thompson, A. E. & Voyer, D. Sex differences in the ability to recognise non-verbal displays of emotion: A meta-analysis. Cognition and Emotion 28, 1164–1195 (2014). 229. Dimberg, U. & Lundquist, L.-O. Gender differences in facial reactions to facial expressions. Biological Psychology 30, 151–159 (1990). 230. Di Tella, M., Miti, F., Ardito, R. B. & Adenzato, M. Social cognition and sex: Are men and women really different? Personality and Individual Differences 162, 110045 (2020). 231. Messinger, D. S. et al. Early sex differences are not autism-specific: A Baby Siblings Research Consortium (BSRC) study. Molecular Autism 6, 32 (2015). 232. Loomes, R., Hull, L. & Mandy, W. P. L. What Is the Male-to-Female Ratio in Autism Spectrum Disorder? A Systematic Review and Meta-Analysis. Journal of the American Academy of Child & Adolescent Psychiatry 56, 466–474 (2017). 233. Özerk, K. & Cardinal, D. Prevalence of Autism/ASD Among Preschool and School- age Children in Norway. Contemp School Psychol 24, 419–428 (2020). 234. Maenner, M. J. Prevalence of Autism Spectrum Disorder Among Children Aged 8 Years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2016. MMWR Surveill Summ 69, (2020). 235. Baird, G. et al. Prevalence of disorders of the autism spectrum in a population cohort of children in South Thames: the Special Needs and Autism Project (SNAP). The Lancet 368, 210–215 (2006). 236. Russell, G., Rodgers, L. R., Ukoumunne, O. C. & Ford, T. Prevalence of Parent- Reported ASD and ADHD in the UK: Findings from the Millennium Cohort Study. J Autism Dev Disord 44, 31–40 (2014). 237. Petrou, A. M., Parr, J. R. & McConachie, H. Gender differences in parent-reported age at diagnosis of children with autism spectrum disorder. Research in Autism Spectrum Disorders 50, 32–42 (2018). 238. Evans, S. C., Boan, A. D., Bradley, C. & Carpenter, L. A. Sex/Gender Differences in Screening for Autism Spectrum Disorder: Implications for Evidence-Based Assessment. Journal of Clinical Child & Adolescent Psychology 48, 840–854 (2019). 239. Lawson, L. P., Joshi, R., Barbaro, J. & Dissanayake, C. Gender Differences During Toddlerhood in Autism Spectrum Disorder: A Prospective Community-Based Longitudinal Follow-Up Study. J Autism Dev Disord 48, 2619–2628 (2018). 221 240. Dean, M., Harwood, R. & Kasari, C. The art of camouflage: Gender differences in the social behaviors of girls and boys with autism spectrum disorder. Autism 21, 678–689 (2017). 241. Halsall, J., Clarke, C. & Crane, L. “Camouflaging” by adolescent autistic girls who attend both mainstream and specialist resource classes: Perspectives of girls, their mothers and their educators. Autism 25, 2074–2086 (2021). 242. Schuck, R. K., Flores, R. E. & Fung, L. K. Brief Report: Sex/Gender Differences in Symptomology and Camouflaging in Adults with Autism Spectrum Disorder. J Autism Dev Disord 49, 2597–2604 (2019). 243. Dworzynski, K., Ronald, A., Bolton, P. & Happé, F. How Different Are Girls and Boys Above and Below the Diagnostic Threshold for Autism Spectrum Disorders? Journal of the American Academy of Child & Adolescent Psychiatry 51, 788–797 (2012). 244. de Giambattista, C., Ventura, P., Trerotoli, P., Margari, F. & Margari, L. Sex Differences in Autism Spectrum Disorder: Focus on High Functioning Children and Adolescents. Frontiers in Psychiatry 12, (2021). 245. Rutherford, M. et al. Gender ratio in a clinical population sample, age of diagnosis and duration of assessment in children and adults with autism spectrum disorder. Autism 20, 628–634 (2016). 246. TURNER, H. H. A SYNDROME OF INFANTILISM, CONGENITAL WEBBED NECK, AND CUBITUS VALGUS1. Endocrinology 23, 566–574 (1938). 247. Nielsen, J. & Wohlert, M. Chromosome abnormalities found among 34,910 newborn children: results from a 13-year incidence study in Arhus, Denmark. Hum. Genet. 87, 81–83 (1991). 248. McCarthy, K. & Bondy, C. A. Turner syndrome in childhood and adolescence. Expert Review of Endocrinology & Metabolism 3, 771–775 (2008). 249. Ford, C. E., Jones, K. W., Polani, P. E., De Almeida, J. C. & Briggs, J. H. A sex- chromosome anomaly in a case of gonadal dysgenesis (Turner’s syndrome). Lancet 1, 711–713 (1959). 250. Pievsky, M. A. & McGrath, R. E. The Neurocognitive Profile of Attention- Deficit/Hyperactivity Disorder: A Review of Meta-Analyses. Arch Clin Neuropsychol 33, 143–157 (2018). 251. Bedford, R. et al. Neurocognitive and observational markers: prediction of autism spectrum disorder from infancy to mid-childhood. Molecular Autism 8, 49 (2017). 222 252. Creswell, C. S. & Skuse, D. H. Autism in association with Turner syndrome: Genetic implications for male vulnerability to pervasive developmental disorders. Neurocase 5, 511–518 (1999). 253. Avdic, H. B. et al. Neurodevelopmental and psychiatric disorders in females with Turner syndrome: a population-based study. Journal of Neurodevelopmental Disorders 13, (2021). 254. Russell, H. F. et al. Increased Prevalence of ADHD in Turner Syndrome with No Evidence of Imprinting Effects. J Pediatr Psychol 31, 945–955 (2006). 255. Green, T. et al. Elucidating X chromosome influences on Attention Deficit Hyperactivity Disorder and executive function. Journal of Psychiatric Research 68, 217–225 (2015). 256. Carrel, L. & Willard, H. F. X-inactivation profile reveals extensive variability in X- linked gene expression in females. Nature 434, 400–404 (2005). 257. Cutter, W. J. et al. Influence of X chromosome and hormones on human brain development: a magnetic resonance imaging and proton magnetic resonance spectroscopy study of Turner syndrome. Biol. Psychiatry 59, 273–283 (2006). 258. Holzapfel, M., Barnea-Goraly, N., Eckert, M. A., Kesler, S. R. & Reiss, A. L. Selective Alterations of White Matter Associated with Visuospatial and Sensorimotor Dysfunction in Turner Syndrome. J. Neurosci. 26, 7007–7013 (2006). 259. Molko, N. et al. Brain Anatomy in Turner Syndrome: Evidence for Impaired Social and Spatial–Numerical Networks. Cereb Cortex 14, 840–850 (2004). 260. Brown, W. E. et al. Brain development in Turner syndrome: a magnetic resonance imaging study. Psychiatry Res 116, 187–196 (2002). 261. Marzelli, M. J., Hoeft, F., Hong, D. S. & Reiss, A. L. Neuroanatomical spatial patterns in Turner syndrome. Neuroimage 55, 439–447 (2011). 262. Kesler, S. R. et al. Amygdala and hippocampal volumes in Turner syndrome: a high-resolution MRI study of X-monosomy. Neuropsychologia 42, 1971 (2004). 263. Knickmeyer, R. C. & Hooper, S. R. The deep biology of cognition: Moving toward a comprehensive neurodevelopmental model of Turner syndrome. American Journal of Medical Genetics Part C: Seminars in Medical Genetics 181, 51–59 (2019). 264. Villalon, J. et al. White matter microstructural abnormalities in girls with chromosome 22q11.2 deletion syndrome, Fragile X or Turner syndrome as evidenced by diffusion tensor imaging. Neuroimage 81, 441–454 (2013). 265. Yamagata, B. et al. White matter aberrations in prepubertal estrogen-naive girls with monosomic Turner syndrome. Cereb. Cortex 22, 2761–2768 (2012). 223 266. van Eimeren, L. et al. Structure-function relationships underlying calculation: a combined diffusion tensor imaging and fMRI study. Neuroimage 52, 358–363 (2010). 267. Thiebaut de Schotten, M. et al. A lateralized brain network for visuospatial attention. Nat. Neurosci. 14, 1245–1246 (2011). 268. Green, T., Saggar, M., Ishak, A., Hong, D. S. & Reiss, A. L. X-Chromosome Effects on Attention Networks: Insights from Imaging Resting-State Networks in Turner Syndrome. Cereb Cortex 28, 3176–3183 (2018). 269. Xie, S. et al. The Effects of the X Chromosome on Intrinsic Functional Connectivity in the Human Brain: Evidence from Turner Syndrome Patients. Cereb. Cortex 27, 474–484 (2017). 270. Bray, S., Dunkin, B., Hong, D. S. & Reiss, A. L. Reduced Functional Connectivity during Working Memory in Turner Syndrome. Cereb Cortex 21, 2471–2481 (2011). 271. Bray, S., Hoeft, F., Hong, D. S. & Reiss, A. L. Aberrant functional network recruitment of posterior parietal cortex in turner syndrome. Human Brain Mapping 34, 3117–3128 (2013). 272. Davenport, M. L. et al. Altered Brain Structure in Infants with Turner Syndrome. Cerebral Cortex 30, 587–596 (2020). 273. Auerbach, J. G., Atzaba-Poria, N., Berger, A. & Landau, R. Emerging developmental pathways to ADHD: possible path markers in early infancy. Neural Plast. 11, 29–43 (2004). 274. Miller, M., Iosif, A.-M., Young, G. S., Hill, M. M. & Ozonoff, S. Brief Report: Early detection of ADHD: Insights from infant siblings of children with autism. J Clin Child Adolesc Psychol 47, 737–744 (2018). 275. Stephens, R. L., Elsayed, H. E., Reznick, J. S., Crais, E. R. & Watson, L. R. Infant Attentional Behaviors Are Associated with ADHD Symptomatology and Executive Function in Early Childhood. J Atten Disord 25, 1908–1918 (2021). 276. McKinnon, C. J. et al. Restricted and Repetitive Behavior and Brain Functional Connectivity in Infants at Risk for Developing Autism Spectrum Disorder. Biol Psychiatry Cogn Neurosci Neuroimaging 4, 50–61 (2019). 277. Piven, J., Elison, J. T. & Zylka, M. J. Toward a conceptual framework for early brain and behavior development in autism. Mol Psychiatry 22, 1385–1394 (2017). 278. Courchesne, E., Gazestani, V. H. & Lewis, N. E. Prenatal Origins of ASD: The When, What, and How of ASD Development. Trends in Neurosciences 43, 326– 342 (2020). 224 279. Birnbaum, R., Jaffe, A. E., Hyde, T. M., Kleinman, J. E. & Weinberger, D. R. Prenatal Expression Patterns of Genes Associated With Neuropsychiatric Disorders. AJP 171, 758–767 (2014). 280. Hall, L. S. et al. Cis-effects on gene expression in the human prenatal brain associated with genetic risk for neuropsychiatric disorders. Mol Psychiatry 26, 2082–2088 (2021). 281. Walton, E. et al. Epigenetic profiling of ADHD symptoms trajectories: a prospective, methylome-wide study. Mol Psychiatry 22, 250–256 (2017). 282. St Clair, D. et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959-1961. JAMA 294, 557–562 (2005). 283. Susser, E. et al. Schizophrenia after prenatal famine. Further evidence. Arch Gen Psychiatry 53, 25–31 (1996). 284. Susser, E., Hoek, H. W. & Brown, A. Neurodevelopmental disorders after prenatal famine: The story of the Dutch Famine Study. Am J Epidemiol 147, 213–216 (1998). 285. Pearson, R. M. et al. Maternal Depression During Pregnancy and the Postnatal Period: Risks and Possible Mechanisms for Offspring Depression at Age 18 Years. JAMA Psychiatry 70, 1312–1319 (2013). 286. Plant, D. T., Pariante, C. M., Sharp, D. & Pawlby, S. Maternal depression during pregnancy and offspring depression in adulthood: Role of child maltreatment. The British Journal of Psychiatry 207, 213–220 (2015). 287. Bergman, K., Sarkar, P., O’connor, T. G., Modi, N. & Glover, V. Maternal Stress During Pregnancy Predicts Cognitive Ability and Fearfulness in Infancy. Journal of the American Academy of Child & Adolescent Psychiatry 46, 1454–1463 (2007). 288. Lyoo, I. K. et al. Lithium-Induced Gray Matter Volume Increase As a Neural Correlate of Treatment Response in Bipolar Disorder: A Longitudinal Brain Imaging Study. Neuropsychopharmacology 35, 1743–1750 (2010). 289. Ho, B.-C., Andreasen, N. C., Ziebell, S., Pierson, R. & Magnotta, V. Long-term Antipsychotic Treatment and Brain Volumes: A Longitudinal Study of First-Episode Schizophrenia. Arch Gen Psychiatry 68, 128–137 (2011). 290. Klein, K. O. et al. Estrogen Replacement in Turner Syndrome: Literature Review and Practical Considerations. J. Clin. Endocrinol. Metab. 103, 1790–1803 (2018). 291. Bondy, C. A. Care of Girls and Women with Turner Syndrome: A Guideline of the Turner Syndrome Study Group. J Clin Endocrinol Metab 92, 10–25 (2007). 225 292. Emerson, R. W. et al. Functional neuroimaging of high-risk 6-month-old infants predicts a diagnosis of autism at 24 months of age. Science Translational Medicine 9, eaag2882 (2017). 293. Dawson, G. et al. Randomized, Controlled Trial of an Intervention for Toddlers With Autism: The Early Start Denver Model. Pediatrics 125, e17–e23 (2010). 294. Biswal, B., Yetkin, F. Z., Haughton, V. M. & Hyde, J. S. Functional connectivity in the motor cortex of resting human brain using echo-planar mri. Magnetic Resonance in Medicine 34, 537–541 (1995). 295. Barkhof, F., Haller, S. & Rombouts, S. A. R. B. Resting-State Functional MR Imaging: A New Window to the Brain. Radiology 272, 29–49 (2014). 296. Wang, D. et al. Altered resting-state network connectivity in congenital blind. Human Brain Mapping 35, 2573–2581 (2014). 297. Tedeschi, G. et al. Increased interictal visual network connectivity in patients with migraine with aura. Cephalalgia 36, 139–147 (2016). 298. Maudoux, A. et al. Auditory Resting-State Network Connectivity in Tinnitus: A Functional MRI Study. PLOS ONE 7, e36222 (2012). 299. Gavrilescu, M. et al. Reduced connectivity of the auditory cortex in patients with auditory hallucinations: a resting state functional magnetic resonance imaging study. Psychol. Med. 40, 1149–1158 (2010). 300. Chadick, J. Z. & Gazzaley, A. Differential coupling of visual cortex with default or frontal-parietal network based on goals. Nat. Neurosci. 14, 830–832 (2011). 301. Ekman, M., Derrfuss, J., Tittgemeyer, M. & Fiebach, C. J. Predicting errors from reconfiguration patterns in human brain networks. Proc. Natl. Acad. Sci. U.S.A. 109, 16714–16719 (2012). 302. Multani, N. et al. Association Between Social Cognition Changes and Resting State Functional Connectivity in Frontotemporal Dementia, Alzheimer’s Disease, Parkinson’s Disease, and Healthy Controls. Front. Neurosci. 13, (2019). 303. Bisecco, A. et al. Resting-State Functional Correlates of Social Cognition in Multiple Sclerosis: An Explorative Study. Front. Behav. Neurosci. 13, (2020). 304. Jimenez, A. M., Riedel, P., Lee, J., Reavis, E. A. & Green, M. F. Linking resting- state networks and social cognition in schizophrenia and bipolar disorder. Hum Brain Mapp 40, 4703–4715 (2019). 305. Reineberg, A. E., Gustavson, D. E., Benca, C., Banich, M. T. & Friedman, N. P. The Relationship Between Resting State Network Connectivity and Individual Differences in Executive Functions. Front. Psychol. 9, (2018). 226 306. Markett, S. et al. Assessing the function of the fronto-parietal attention network: Insights from resting-state fMRI and the attentional network test. Human Brain Mapping 35, 1700–1709 (2014). 307. Rosenberg, M. D. et al. A neuromarker of sustained attention from whole-brain functional connectivity. Nature Neuroscience 19, 165–171 (2016). 308. Li, P. et al. Altered Brain Network Connectivity as a Potential Endophenotype of Schizophrenia. Scientific Reports 7, 5483 (2017). 309. Tu, P.-C., Lee, Y.-C., Chen, Y.-S., Li, C.-T. & Su, T.-P. Schizophrenia and the brain’s control network: Aberrant within- and between-network connectivity of the frontoparietal network in schizophrenia. Schizophrenia Research 147, 339–347 (2013). 310. Repovs, G., Csernansky, J. G. & Barch, D. M. Brain Network Connectivity in Individuals with Schizophrenia and Their Siblings. Biological Psychiatry 69, 967– 973 (2011). 311. Fan, J. et al. Resting-State Default Mode Network Related Functional Connectivity Is Associated With Sustained Attention Deficits in Schizophrenia and Obsessive- Compulsive Disorder. Front. Behav. Neurosci. 12, (2018). 312. Nomi, J. S. & Uddin, L. Q. Developmental changes in large-scale network connectivity in autism. NeuroImage: Clinical 7, 732–741 (2015). 313. Jakab, A. et al. Fetal functional imaging portrays heterogeneous development of emerging human brain networks. Front Hum Neurosci 8, 852 (2014). 314. Lin, W. et al. Functional Connectivity MR Imaging Reveals Cortical Functional Connectivity in the Developing Brain. American Journal of Neuroradiology 29, 1883–1889 (2008). 315. Doria, V. et al. Emergence of resting state networks in the preterm human brain. PNAS (2010) doi:10.1073/pnas.1007921107. 316. Smyser, C. D. et al. Longitudinal Analysis of Neural Network Development in Preterm Infants. Cereb Cortex 20, 2852–2862 (2010). 317. Smyser, C. D. et al. Effects of White Matter Injury on Resting State fMRI Measures in Prematurely Born Infants. PLOS ONE 8, e68098 (2013). 318. Gao, W. et al. Functional Network Development During the First Year: Relative Sequence and Socioeconomic Correlations. Cereb. Cortex 25, 2919–2928 (2015). 319. Satterthwaite, T. D. & Baker, J. T. How Can Studies of Resting-state Functional Connectivity Help Us Understand Psychosis as a Disorder of Brain Development? Curr Opin Neurobiol 0, 85–91 (2015). 227 320. Glahn, D. C. et al. Genetic control over the resting brain. PNAS 107, 1223–1228 (2010). 321. Colclough, G. L. et al. The heritability of multi-modal connectivity in human brain activity. eLife 6, e20178 (2017). 322. Reineberg, A. E., Hatoum, A. S., Hewitt, J. K., Banich, M. T. & Friedman, N. P. Genetic and Environmental Influence on the Human Functional Connectome. Cereb Cortex 30, 2099–2113 (2020). 323. Teeuw, J. et al. Genetic and environmental influences on functional connectivity within and between canonical cortical resting-state networks throughout adolescent development in boys and girls. Neuroimage 202, 116073 (2019). 324. Gao, W. et al. Intersubject Variability of and Genetic Effects on the Brain’s Functional Connectivity during Infancy. J. Neurosci. 34, 11288–11296 (2014). 325. Qiu, A. et al. Prenatal maternal depression alters amygdala functional connectivity in 6-month-old infants. Translational Psychiatry 5, e508–e508 (2015). 326. Malykhin, N. V., Carter, R., Hegadoren, K. M., Seres, P. & Coupland, N. J. Fronto- limbic volumetric changes in major depressive disorder. Journal of Affective Disorders 136, 1104–1113 (2012). 327. Holmes, A. J. et al. Individual Differences in Amygdala-Medial Prefrontal Anatomy Link Negative Affect, Impaired Social Functioning, and Polygenic Depression Risk. J. Neurosci. 32, 18087–18100 (2012). 328. Eijndhoven, P. van et al. Amygdala Volume Marks the Acute State in the Early Course of Depression. Biological Psychiatry 65, 812–818 (2009). 329. Salzwedel, A. P. et al. Prenatal Drug Exposure Affects Neonatal Brain Functional Connectivity. J Neurosci 35, 5860–5869 (2015). 330. Salzwedel, A., Chen, G., Chen, Y., Grewen, K. & Gao, W. Functional dissection of prenatal drug effects on baby brain and behavioral development. Hum Brain Mapp 41, 4789–4803 (2020). 331. Li, Z., Lei, K., Coles, C. D., Lynch, M. E. & Hu, X. Longitudinal changes of amygdala functional connectivity in adolescents prenatally exposed to cocaine. Drug and Alcohol Dependence 200, 50–58 (2019). 332. Radhakrishnan, R. et al. Resting state functional MRI in infants with prenatal opioid exposure—a pilot study. Neuroradiology (2020) doi:10.1007/s00234-020-02552-3. 333. Salzwedel, A. P., Grewen, K. M., Goldman, B. D. & Gao, W. Thalamocortical functional connectivity and behavioral disruptions in neonates with prenatal cocaine exposure. Neurotoxicology and Teratology 56, 16–25 (2016). 228 334. Jha, S. C. et al. Environmental Influences on Infant Cortical Thickness and Surface Area. Cereb Cortex 29, 1139–1149 (2019). 335. He, L. et al. Early prediction of cognitive deficits in very preterm infants using functional connectome data in an artificial neural network framework. NeuroImage: Clinical 18, 290–297 (2018). 336. Chen, M. et al. Early Prediction of Cognitive Deficit in Very Preterm Infants Using Brain Structural Connectome With Transfer Learning Enhanced Deep Convolutional Neural Networks. Front. Neurosci. 14, (2020). 337. Knickmeyer, R. C. et al. Impact of Demographic and Obstetric Factors on Infant Brain Volumes: A Population Neuroscience Study. Cereb. Cortex 27, 5616–5625 (2017). 338. Gilmore, J. H. et al. Genetic and environmental contributions to neonatal brain structure: A twin study. Hum Brain Mapp 31, 1174–1182 (2010). 339. Knickmeyer, R. C. et al. A structural MRI study of human brain development from birth to 2 years. J. Neurosci. 28, 12176–12182 (2008). 340. Chen, Y. et al. The Subgrouping Structure of Newborns with Heterogenous Brain– Behavior Relationships. Cerebral Cortex 31, 301–311 (2021). 341. Smith, S. M. et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 23 Suppl 1, S208-219 (2004). 342. Cox, R. W. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput. Biomed. Res. 29, 162–173 (1996). 343. Power, J. D. et al. Functional network organization of the human brain. Neuron 72, 665–678 (2011). 344. Satterthwaite, T. D. et al. Impact of in-scanner head motion on multiple measures of functional connectivity: relevance for studies of neurodevelopment in youth. Neuroimage 60, 623–632 (2012). 345. Satterthwaite, T. D. et al. An improved framework for confound regression and filtering for control of motion artifact in the preprocessing of resting-state functional connectivity data. Neuroimage 64, 240–256 (2013). 346. Gao, W., Lin, W., Grewen, K. & Gilmore, J. H. Functional Connectivity of the Infant Human Brain: Plastic and Modifiable. Neuroscientist 23, 169–184 (2017). 347. Gao, W. et al. The synchronization within and interaction between the default and dorsal attention networks in early infancy. Cerebral Cortex 23, 594–603 (2013). 229 348. Alcauter, S. et al. Development of Thalamocortical Connectivity during Infancy and Its Cognitive Correlations. J. Neurosci. 34, 9067–9075 (2014). 349. Braun, U. et al. Test–retest reliability of resting-state connectivity network characteristics using fMRI and graph theoretical measures. NeuroImage 59, 1404– 1412 (2012). 350. Li, Z., Kadivar, A., Pluta, J., Dunlop, J. & Wang, Z. Test-retest stability analysis of resting brain activity revealed by blood oxygen level-dependent functional MRI. J Magn Reson Imaging 36, 344–354 (2012). 351. Birn, R. M. et al. The effect of scan length on the reliability of resting-state fMRI connectivity estimates. NeuroImage 83, 550–558 (2013). 352. Shi, J. et al. Genome-wide association study of recurrent early-onset major depressive disorder. Mol Psychiatry 16, 193–201 (2011). 353. Shi, F., Salzwedel, A. P., Lin, W., Gilmore, J. H. & Gao, W. Functional Brain Parcellations of the Infant Brain and the Associated Developmental Trends. Cereb Cortex 28, 1358–1368 (2018). 354. Tzourio-Mazoyer, N. et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 15, 273–289 (2002). 355. Yeo, B. T. T. et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J Neurophysiol 106, 1125–1165 (2011). 356. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria (2020). 357. Gamer, M., Lemon, J., Fellows, I. & Puspendra, S. irr: Various Coefficients of Interrater Reliability and Agreement. (2019). 358. Xia, K. et al. Environmental and Genetic Contributors to Salivary Testosterone Levels in Infants. Front Endocrinol (Lausanne) 5, (2014). 359. Achterberg, M. et al. Distinctive heritability patterns of subcortical-prefrontal cortex resting state connectivity in childhood: A twin study. NEUROIMAGE 175, 138–149 (2018). 360. Hodel, A. S. Rapid infant prefrontal cortex development and sensitivity to early environmental experience. Developmental Review 48, 113–144 (2018). 361. Saudino, K. J., Cherny, S. S. & Plomin, R. Parent ratings of temperament in twins: explaining the ‘too low’ DZ correlations. Twin Res 3, 224–233 (2000). 230 362. Smyser, C. D. et al. Prediction of brain maturity in infants using machine-learning algorithms. NeuroImage 136, 1–9 (2016). 363. Galéra, C. et al. Early risk factors for hyperactivity-impulsivity and inattention trajectories from age 17 months to 8 years. Arch Gen Psychiatry 68, 1267–1275 (2011). 364. Rasic, D., Hajek, T., Alda, M. & Uher, R. Risk of mental illness in offspring of parents with schizophrenia, bipolar disorder, and major depressive disorder: a meta-analysis of family high-risk studies. Schizophr Bull 40, 28–38 (2014). 365. Wolford, E. et al. Maternal depressive symptoms during and after pregnancy are associated with attention-deficit/hyperactivity disorder symptoms in their 3- to 6- year-old children. PLoS One 12, e0190248 (2017). 366. Chen, M.-H. et al. Coaggregation of Major Psychiatric Disorders in First-Degree Relatives of Individuals With Attention-Deficit/Hyperactivity Disorder: A Nationwide Population-Based Study. J Clin Psychiatry 80, (2019). 367. Vizzini, L. et al. Maternal anxiety, depression and sleep disorders before and during pregnancy, and preschool ADHD symptoms in the NINFEA birth cohort study. Epidemiology and Psychiatric Sciences 28, 521–531 (2019). 368. Chen, L.-C. et al. Association of parental depression with offspring attention deficit hyperactivity disorder and autism spectrum disorder: A nationwide birth cohort study. J Affect Disord 277, 109–114 (2020). 369. Eilertsen, T. et al. Parental socioeconomic status and child intellectual functioning in a Norwegian sample. Scandinavian Journal of Psychology 57, 399–405 (2016). 370. Jeong, J., Kim, R. & Subramanian, S. V. How consistent are associations between maternal and paternal education and child growth and development outcomes across 39 low-income and middle-income countries? J Epidemiol Community Health 72, 434–441 (2018). 371. González, L. et al. The role of parental social class, education and unemployment on child cognitive development. Gaceta Sanitaria 34, 51–60 (2020). 372. Østergaard, S. D. et al. Polygenic risk score, psychosocial environment and the risk of attention-deficit/hyperactivity disorder. Translational Psychiatry 10, 1–11 (2020). 373. Rohr, C. S. et al. Functional Connectivity of the Dorsal Attention Network Predicts Selective Attention in 4–7 year-old Girls. Cerebral Cortex 27, 4350–4360 (2017). 374. Aboitiz, F., Ossandón, T., Zamorano, F., Palma, B. & Carrasco, X. Irrelevant stimulus processing in ADHD: catecholamine dynamics and attentional networks. Front Psychol 5, 183 (2014). 231 375. Luke, B. & Keith, L. G. The contribution of singletons, twins and triplets to low birth weight, infant mortality and handicap in the United States. J Reprod Med 37, 661– 666 (1992). 376. Levy, F., McLaughlin, M., Wood, C., Hay, D. & Waldman, I. Twin-Sibling Differences in Parental Reports of ADHD, Speech, Reading and Behaviour Problems. Journal of Child Psychology and Psychiatry 37, 569–578 (1996). 377. Lehn, H. et al. Attention Problems and Attention-Deficit/Hyperactivity Disorder in Discordant and Concordant Monozygotic Twins: Evidence of Environmental Mediators. Journal of the American Academy of Child & Adolescent Psychiatry 46, 83–91 (2007). 378. Groen-Blokhuis, M. M., Middeldorp, C. M., van Beijsterveldt, C. E. M. & Boomsma, D. I. Evidence for a Causal Association of Low Birth Weight and Attention Problems. Journal of the American Academy of Child & Adolescent Psychiatry 50, 1247-1254.e2 (2011). 379. Garite, T. J., Clark, R. H., Elliott, J. P., Thorp, J. A., & the Pediatrix/Obstetrix Perinatal Research Group. Twins and triplets: The effect of plurality and growth on neonatal outcome compared with singleton infants. American Journal of Obstetrics and Gynecology 191, 700–707 (2004). 380. Rao, A., Sairam, S. & Shehata, H. Obstetric complications of twin pregnancies. Best Practice & Research Clinical Obstetrics & Gynaecology 18, 557–576 (2004). 381. Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar, B. L. & Petersen, S. E. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage 59, 2142–2154 (2012). 382. Van Dijk, K. R. A., Sabuncu, M. R. & Buckner, R. L. The influence of head motion on intrinsic functional connectivity MRI. Neuroimage 59, 431–438 (2012). 383. Couvy-Duchesne, B. et al. Heritability of head motion during resting state functional MRI in 462 healthy twins. Neuroimage 102, 424–434 (2014). 384. Engelhardt, L. E. et al. Children’s head motion during fMRI tasks is heritable and stable over time. Developmental Cognitive Neuroscience 25, 58–68 (2017). 385. Mitra, A. et al. Resting-state fMRI in sleeping infants more closely resembles adult sleep than adult wakefulness. PLOS ONE 12, e0188122 (2017). 386. Picchioni, D. et al. fMRI differences between early and late stage-1 sleep. Neuroscience Letters 441, 81–85 (2008). 387. Redcay, E., Kennedy, D. P. & Courchesne, E. fMRI during natural sleep as a method to study brain function during early childhood. NeuroImage 38, 696–707 (2007). 232 388. Fransson, P., Åden, U., Blennow, M. & Lagercrantz, H. The Functional Architecture of the Infant Brain as Revealed by Resting-State fMRI. Cereb Cortex 21, 145–154 (2011). 389. Liu, W.-C., Flax, J. F., Guise, K. G., Sukul, V. & Benasich, A. A. Functional connectivity of the sensorimotor area in naturally sleeping infants. Brain Research 1223, 42–49 (2008). 390. Andropoulos, D. B. Effect of Anesthesia on the Developing Brain: Infant and Fetus. FDT 43, 1–11 (2018). 391. York, T. P. et al. Fetal and maternal genes’ influence on gestational age in a quantitative genetic analysis of 244,000 Swedish births. Am J Epidemiol 178, 543– 550 (2013). 392. Wu, W. et al. The heritability of gestational age in a two-million member cohort: implications for spontaneous preterm birth. Hum Genet 134, 803–808 (2015). 393. Mook-Kanamori, D. O. et al. Heritability Estimates of Body Size in Fetal Life and Early Childhood. PLOS ONE 7, e39901 (2012). 394. Turkheimer, E. & Waldron, M. Nonshared environment: A theoretical, methodological, and quantitative review. Psychological Bulletin 126, 78–108 (2000). 395. Zhao, C. & Gong, G. Mapping the effect of the X chromosome on the human brain: Neuroimaging evidence from Turner syndrome. Neuroscience & Biobehavioral Reviews 80, 263–275 (2017). 396. Raznahan, A. et al. Sex-chromosome dosage effects on gene expression in humans. PNAS 115, 7398–7403 (2018). 397. Balaton, B. P., Cotton, A. M. & Brown, C. J. Derivation of consensus inactivation status for X-linked genes from genome-wide studies. Biol Sex Differ 6, (2015). 398. Gravholt, C. H. et al. Clinical practice guidelines for the care of girls and women with Turner syndrome: proceedings from the 2016 Cincinnati International Turner Syndrome Meeting. Eur. J. Endocrinol. 177, G1–G70 (2017). 399. Hong, D., Kent, J. S. & Kesler, S. COGNITIVE PROFILE OF TURNER SYNDROME. Dev Disabil Res Rev 15, 270–278 (2009). 400. Hong, D. S., Dunkin, B. & Reiss, A. L. Psychosocial functioning and social cognitive processing in girls with Turner syndrome. J Dev Behav Pediatr 32, 512– 520 (2011). 401. Burnett, A. C., Reutens, D. C. & Wood, A. G. Social cognition in Turner’s Syndrome. Journal of Clinical Neuroscience 17, 283–286 (2010). 233 402. Kirk, J. W., Mazzocco, M. M. M. & Kover, S. T. Assessing Executive Dysfunction in Girls With Fragile X or Turner Syndrome Using the Contingency Naming Test (CNT). Developmental Neuropsychology 28, 755–777 (2005). 403. Mauger, C. et al. Executive Functions in Children and Adolescents with Turner Syndrome: A Systematic Review and Meta-Analysis. Neuropsychol Rev 28, 188– 215 (2018). 404. Haberecht, M. F. et al. Functional neuroanatomy of visuo-spatial working memory in turner syndrome. Human Brain Mapping 14, 96–107 (2001). 405. Hart, S. J., Davenport, M. L., Hooper, S. R. & Belger, A. Visuospatial executive function in Turner syndrome: functional MRI and neurocognitive findings. Brain 129, 1125–1136 (2006). 406. Spiliotis, B. E. Recombinant human growth hormone in the treatment of Turner syndrome. Ther Clin Risk Manag 4, 1177–1183 (2008). 407. Rubin, R. T., Reinisch, J. M. & Haskett, R. F. Postnatal Gonadal Steroid Effects on Human Behavior. Science 211, 1318–1324 (1981). 408. Berman, K. F. et al. Modulation of cognition-specific cortical activity by gonadal steroids: A positron-emission tomography study in women. Proceedings of the National Academy of Sciences 94, 8836–8841 (1997). 409. Schulz, K. M. & Sisk, C. L. The organizing actions of adolescent gonadal steroid hormones on brain and behavioral development. Neuroscience & Biobehavioral Reviews 70, 148–158 (2016). 410. Girault, J. B. et al. White matter microstructural development and cognitive ability in the first 2 years of life. Human Brain Mapping 40, 1195–1210 (2019). 411. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing. (2021). 412. Verde, A. R. et al. UNC-Utah NA-MIC framework for DTI fiber tract analysis. Front Neuroinform 7, (2014). 413. Oguz, I. et al. DTIPrep: quality control of diffusion-weighted images. Front. Neuroinform. 8, (2014). 414. Jha, S. C. et al. Genetic influences on neonatal cortical thickness and surface area. Human Brain Mapping 39, 4998–5013 (2018). 415. Zhu, H. et al. FADTTS: functional analysis of diffusion tensor tract statistics. Neuroimage 56, 1412–1425 (2011). 234 416. Kamali, A., Flanders, A. E., Brody, J., Hunter, J. V. & Hasan, K. M. Tracing superior longitudinal fasciculus connectivity in the human brain using high resolution diffusion tensor tractography. Brain Struct Funct 219, 269–281 (2014). 417. Rizio, A. A. & Diaz, M. T. Language, aging, and cognition: Frontal aslant tract and superior longitudinal fasciculus contribute to working memory performance in older adults. Neuroreport 27, 689–693 (2016). 418. Koyama, T. & Domen, K. Diffusion Tensor Fractional Anisotropy in the Superior Longitudinal Fasciculus Correlates with Functional Independence Measure Cognition Scores in Patients with Cerebral Infarction. Journal of Stroke and Cerebrovascular Diseases 26, 1704–1711 (2017). 419. Madhavan, K. M., McQueeny, T., Howe, S. R., Shear, P. & Szaflarski, J. Superior longitudinal fasciculus and language functioning in healthy aging. Brain Research 1562, 11–22 (2014). 420. Pretzel, R. E. et al. Early Development of Infants with Turner Syndrome. J Dev Behav Pediatr 41, 470–479 (2020). 421. Sävendahl, L. & Davenport, M. L. Delayed diagnoses of Turner’s syndrome: proposed guidelines for change. J Pediatr 137, 455–459 (2000). 422. Moreno-García, M., Fernández-Martínez, F. J. & Barreiro Miranda, E. Chromosomal anomalies in patients with short stature. Pediatr Int 47, 546–549 (2005). 423. Bondy, C. A. Congenital Cardiovascular Disease in Turner Syndrome. Congenital Heart Disease 3, 2–15 (2008). 424. Eckhauser, A., South, S. T., Meyers, L., Bleyl, S. B. & Botto, L. D. Turner Syndrome in Girls Presenting with Coarctation of the Aorta. The Journal of Pediatrics 167, 1062–1066 (2015). 425. McGoey, K. E., Eckert, T. L. & Dupaul, G. J. Early Intervention for Preschool-Age Children with ADHD: A Literature Review. Journal of Emotional and Behavioral Disorders 10, 14–28 (2002). 426. Jones, K., Daley, D., Hutchings, J., Bywater, T. & Eames, C. Efficacy of the Incredible Years Basic parent training programme as an early intervention for children with conduct problems and ADHD. Child: Care, Health and Development 33, 749–756 (2007). 427. Re, A. M., Capodieci, A. & Cornoldi, C. Effect of training focused on executive functions (attention, inhibition, and working memory) in preschoolers exhibiting ADHD symptoms. Front. Psychol. 6, (2015). 235 428. Eapen, V., Črnčec, R. & Walter, A. Clinical outcomes of an early intervention program for preschool children with Autism Spectrum Disorder in a community group setting. BMC Pediatrics 13, 3 (2013). 429. Zwaigenbaum, L. et al. Early Intervention for Children With Autism Spectrum Disorder Under 3 Years of Age: Recommendations for Practice and Research. Pediatrics 136, S60–S81 (2015). 430. Saaybi, S. et al. Pre- and Post-therapy Assessment of Clinical Outcomes and White Matter Integrity in Autism Spectrum Disorder: Pilot Study. Front. Neurol. 10, (2019). 431. Swanson, M. R. & Hazlett, H. C. White matter as a monitoring biomarker for neurodevelopmental disorder intervention studies. J Neurodevelop Disord 11, 33 (2019). 432. Stephens, R. L. et al. White Matter Development from Birth to 6 Years of Age: A Longitudinal Study. Cereb Cortex 30, 6152–6168 (2020). 433. Hüppi, P. S. et al. Microstructural Development of Human Newborn Cerebral White Matter Assessed in Vivo by Diffusion Tensor Magnetic Resonance Imaging. Pediatr Res 44, 584–590 (1998). 434. Forest, M. G., Cathiard, A. M. & Bertrand, J. A. EVIDENCE OF TESTICULAR ACTIVITY IN EARLY INFANCY. The Journal of Clinical Endocrinology & Metabolism 37, 148–151 (1973). 435. REYES, F. I., BORODITSKY, R. S., WINTER, J. S. D. & FAIMAN, C. Studies on Human Sexual Development. II. Fetal and Maternal Serum Gonadotropin and Sex Steroid Concentrations1. The Journal of Clinical Endocrinology & Metabolism 38, 612–617 (1974). 436. Shigeo, T. et al. Sex differences in fetal gonadotropins and androgens. Journal of Steroid Biochemistry 8, 609–620 (1977). 437. TAPANAINEN, J., KELLOKUMPU-LEHTINEN, P., PELLINIEMI, L. & HUHTANIEMI, I. Age-Related Changes in Endogenous Steroids of Human Fetal Testis during Early and Midpregnancy*. The Journal of Clinical Endocrinology & Metabolism 52, 98–102 (1981). 438. Andersson, A.-M. et al. Longitudinal Reproductive Hormone Profiles in Infants: Peak of Inhibin B Levels in Infant Boys Exceeds Levels in Adult Men1. The Journal of Clinical Endocrinology & Metabolism 83, 675–681 (1998). 439. Kuiri-Hänninen, T. et al. Increased Activity of the Hypothalamic-Pituitary-Testicular Axis in Infancy Results in Increased Androgen Action in Premature Boys. The Journal of Clinical Endocrinology & Metabolism 96, 98–105 (2011). 236 440. Gilmore, J. H., Santelli, R. K. & Gao, W. Imaging structural and functional brain development in early childhood. Nat Rev Neurosci 19, 123–137 (2018). 441. Ikuta, T. et al. Abnormal cingulum bundle development in autism: A probabilistic tractography study. Psychiatry Research: Neuroimaging 221, 63–68 (2014). 442. Bathelt, J., Johnson, A., Zhang, M. & Astle, D. E. The cingulum as a marker of individual differences in neurocognitive development. Sci Rep 9, 2281 (2019). 443. Tamnes, C. K. et al. Brain Maturation in Adolescence and Young Adulthood: Regional Age-Related Changes in Cortical Thickness and White Matter Volume and Microstructure. Cerebral Cortex 20, 534–548 (2010). 444. Lebel, C., Treit, S. & Beaulieu, C. A review of diffusion MRI of typical white matter development from early childhood to young adulthood. NMR in Biomedicine 32, e3778 (2019). 445. Wu, Y., Sun, D., Wang, Y., Wang, Y. & Ou, S. Segmentation of the Cingulum Bundle in the Human Brain: A New Perspective Based on DSI Tractography and Fiber Dissection Study. Front. Neuroanat. 0, (2016). 446. Metzler-Baddeley, C. et al. Temporal association tracts and the breakdown of episodic memory in mild cognitive impairment. Neurology 79, 2233–2240 (2012). 447. Green, T. et al. Sex differences in amygdala shape: Insights from Turner syndrome. Human Brain Mapping 37, 1593–1601 (2016). 448. Skuse, D. H., Morris, J. S. & Dolan, R. J. Functional dissociation of amygdala- modulated arousal and cognitive appraisal, in Turner syndrome. Brain 128, 2084– 2096 (2005). 449. Nijhuis-van der Sanden, M., Smits-Engelsman, B., Eling, P., Nijhuis, B. & Galen, G. Low Elementary Movement Speed Is Associated With Poor Motor Skill in Turner’s Syndrome. Developmental neuropsychology 22, 643–70 (2002). 450. Nijhuis-van der Sanden, M. W. G., Eling, P. A. T. M. & Otten, B. J. A review of neuropsychological and motor studies in Turner Syndrome. Neuroscience & Biobehavioral Reviews 27, 329–338 (2003). 451. Ross, J. L., Kushner, H. & Roeltgen, D. P. Developmental changes in motor function in girls with Turner syndrome. Pediatr Neurol 15, 317–322 (1996). 452. Sanden, R. W. G. N. der, Smits-Engelsman, B. C. & Eling, P. A. Motor performance in girls with Turner syndrome. Developmental Medicine and Child Neurology 42, 685–690 (2000). 237 453. Pizzini, F. B. et al. Review of Corpus Callosum Topography, Analysis of Diffusion Values for the Different Callosal Fibers and Sex Differences. Neuroradiol J 21, 745–754 (2008). 454. Xie, S. et al. The Effects of X Chromosome Loss on Neuroanatomical and Cognitive Phenotypes During Adolescence: a Multi-modal Structural MRI and Diffusion Tensor Imaging Study. Cereb Cortex 25, 2842–2853 (2015). 455. Tusa, R. J. & Ungerleider, L. G. The inferior longitudinal fasciculus: A reexamination in humans and monkeys. Annals of Neurology 18, 583–591 (1985). 456. Hanaie, R. et al. Abnormal Corpus Callosum Connectivity, Socio-communicative Deficits, and Motor Deficits in Children with Autism Spectrum Disorder: A Diffusion Tensor Imaging Study. J Autism Dev Disord 44, 2209–2220 (2014). 457. Takao, H., Hayashi, N. & Ohtomo, K. Sex dimorphism in the white matter: Fractional anisotropy and brain size. Journal of Magnetic Resonance Imaging 39, 917–923 (2014). 458. Shaw, P. et al. White Matter Microstructure and the Variable Adult Outcome of Childhood Attention Deficit Hyperactivity Disorder. Neuropsychopharmacol 40, 746–754 (2015). 459. Jou, R. J. et al. Structural Neural Phenotype of Autism: Preliminary Evidence from a Diffusion Tensor Imaging Study Using Tract-Based Spatial Statistics. American Journal of Neuroradiology 32, 1607–1613 (2011). 460. Koldewyn, K. et al. Differences in the right inferior longitudinal fasciculus but no general disruption of white matter tracts in children with autism spectrum disorder. PNAS 111, 1981–1986 (2014). 461. Boets, B. et al. Alterations in the inferior longitudinal fasciculus in autism and associations with visual processing: a diffusion-weighted MRI study. Molecular Autism 9, 10 (2018). 462. Krogsrud, S. et al. Development of white matter microstructure in relation to verbal and visuospatial working memory—A longitudinal study. PLOS ONE 13, e0195540 (2018). 463. Peper, J. S., van den Heuvel, M. P., Mandl, R. C. W., Pol, H. E. H. & van Honk, J. Sex steroids and connectivity in the human brain: A review of neuroimaging studies. Psychoneuroendocrinology 36, 1101–1113 (2011). 464. Green, T., Flash, S. & Reiss, A. L. Sex differences in psychiatric disorders: what we can learn from sex chromosome aneuploidies. Neuropsychopharmacology 44, 9–21 (2019). 238 465. Ross, J. et al. Persistent cognitive deficits in adult women with Turner syndrome. Neurology 58, 218–25 (2002). 466. Quintero, A. I., Beaton, E. A., Harvey, D. J., Ross, J. L. & Simon, T. J. Common and specific impairments in attention functioning in girls with chromosome 22q11.2 deletion, fragile X or Turner syndromes. J Neurodevelop Disord 6, 1–15 (2014). 467. Attout, L., Noël, M.-P., Nassogne, M.-C. & Rousselle, L. The role of short-term memory and visuo-spatial skills in numerical magnitude processing: Evidence from Turner syndrome. PLOS ONE 12, e0171454 (2017). 468. McCauley, E., Kay, T., Ito, J. & Treder, R. The Turner Syndrome: Cognitive Deficits, Affective Discrimination, and Behavior Problems. Child Development 58, 464–473 (1987). 469. Downey, J. et al. Cognitive Ability and Everyday Functioning in Women with Turner Syndrome. J Learn Disabil 24, 32–39 (1991). 470. Romans, S. M., Roeltgen, D. P., Kushner, H. & Ross, J. L. Executive function in girls with turner’s syndrome. Developmental Neuropsychology 13, 23–40 (1997). 471. Harris, P. A. et al. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics 42, 377–381 (2009). 472. Harris, P. A. et al. The REDCap consortium: Building an international community of software platform partners. Journal of Biomedical Informatics 95, 103208 (2019). 473. Ganis, G. & Kievit, R. A New Set of Three-Dimensional Shapes for Investigating Mental Rotation Processes: Validation Data and Stimulus Set. Journal of Open Psychology Data 3, e3 (2015). 474. Baron-Cohen, S., Jolliffe, T., Mortimore, C. & Robertson, M. Another Advanced Test of Theory of Mind: Evidence from Very High Functioning Adults with Autism or Asperger Syndrome. Journal of Child Psychology and Psychiatry 38, 813–822 (1997). 475. Klee, S. H. & Garfinkel, B. D. The computerized continuous performance task: A new measure of inattention. J Abnorm Child Psychol 11, 487–495 (1983). 476. Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J. & Clubley, E. The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High- Functioning Autism, Malesand Females, Scientists and Mathematicians. J Autism Dev Disord 31, 5–17 (2001). 477. Siemann, J., Herrmann, M. & Galashan, D. The effect of feature-based attention on flanker interference processing: An fMRI-constrained source analysis. Sci Rep 8, 1580 (2018). 239 478. van Veen, V., Cohen, J. D., Botvinick, M. M., Stenger, V. A. & Carter, C. S. Anterior cingulate cortex, conflict monitoring, and levels of processing. Neuroimage 14, 1302–1308 (2001). 479. Wei, P., Szameitat, A. J., Müller, H. J., Schubert, T. & Zhou, X. The neural correlates of perceptual load induced attentional selection: an fMRI study. Neuroscience 250, 372–380 (2013). 480. Hübner, R. & Töbel, L. Conflict resolution in the Eriksen flanker task: Similarities and differences to the Simon task. PLOS ONE 14, e0214203 (2019). 481. Hübner, R., Steinhauser, M. & Lehle, C. A dual-stage two-phase model of selective attention. Psychol Rev 117, 759–784 (2010). 482. Cyr, A.-A., Romero, K. & Galin-Corini, L. Web-Based Cognitive Testing of Older Adults in Person Versus at Home: Within-Subjects Comparison Study. JMIR Aging 4, e23384 (2021). 483. Bogler, C., Vowinkel, A., Zhutovsky, P. & Haynes, J.-D. Default Network Activity Is Associated with Better Performance in a Vigilance Task. Frontiers in Human Neuroscience 11, (2017). 484. Bonnelle, V. et al. Default Mode Network Connectivity Predicts Sustained Attention Deficits after Traumatic Brain Injury. J. Neurosci. 31, 13442–13451 (2011). 485. Danckert, J. & Merrifield, C. Boredom, sustained attention and the default mode network. Exp Brain Res 236, 2507–2518 (2018). 486. Tana, M. G., Montin, E., Cerutti, S. & Bianchi, A. M. Exploring Cortical Attentional System by Using fMRI during a Continuous Perfomance Test. Comput Intell Neurosci 2010, 329213 (2010). 487. Woods, D. L., Wyma, J. M., Yund, E. W., Herron, T. J. & Reed, B. Factors influencing the latency of simple reaction time. Frontiers in Human Neuroscience 9, (2015). 488. Turken, U. et al. Cognitive processing speed and the structure of white matter pathways: Convergent evidence from normal variation and lesion studies. NeuroImage 42, 1032–1044 (2008). 489. Brankaer, C., Ghesquière, P., De Wel, A., Swillen, A. & De Smedt, B. Numerical magnitude processing impairments in genetic syndromes: a cross-syndrome comparison of Turner and 22q11.2 deletion syndromes. Developmental Science 20, n/a-N.PAG (2017). 490. Ross, J. L., Roeltgen, D., Feuillan, P., Kushner, H. & Cutler, G. B. Use of estrogen in young girls with Turner syndrome: Effects on memory. Neurology 54, 164–164 (2000). 240 491. Chechlacz, M., Rotshtein, P. & Humphreys, G. W. Neuronal substrates of Corsi Block span: Lesion symptom mapping analyses in relation to attentional competition and spatial bias. Neuropsychologia 64, 240–251 (2014). 492. Wilson, B. A. et al. Egocentric Disorientation following Bilateral Parietal Lobe Damage. Cortex 41, 547–554 (2005). 493. Rovet, J. & Netley, C. The mental rotation task performance of Turner syndrome subjects. Behav Genet 10, 437–443 (1980). 494. Rovet, J. & Netley, C. Processing deficits in Turner’s syndrome. Developmental Psychology 18, 77–94 (1982). 495. Gauthier, I. et al. BOLD Activity during Mental Rotation and Viewpoint-Dependent Object Recognition. Neuron 34, 161–171 (2002). 496. Kirkland, R., Peterson, E., Baker, C., Miller, S. & Pulos, S. Meta-analysis reveals adult female superiority in ‘Reading the Mind in the Eyes Test’. North American Journal of Psychology 15, 449–458 (2013). 497. Sato, W. et al. Structural Neural Substrates of Reading the Mind in the Eyes. Frontiers in Human Neuroscience 10, (2016). 498. Sato, W. et al. Structural Correlates of Reading the Mind in the Eyes in Autism Spectrum Disorder. Front Hum Neurosci 11, 361 (2017). 499. Holt, R. J. et al. ‘Reading the Mind in the Eyes’: an fMRI study of adolescents with autism and their siblings. Psychol Med 44, 3215–3227 (2014). 500. Baker, M. 1,500 scientists lift the lid on reproducibility. Nature 533, 452–454 (2016). 501. National Academies of Sciences, E. Reproducibility and Replicability in Science. (2019). doi:10.17226/25303. 502. Gu, Z., Jamison, K. W., Sabuncu, M. R. & Kuceyeski, A. Heritability and interindividual variability of regional structure-function coupling. Nat Commun 12, 4894 (2021). 503. Baum, G. L. et al. Development of structure–function coupling in human brain networks during youth. Proc Natl Acad Sci U S A 117, 771–778 (2020). 504. Corbitt, H. et al. TIMP3 and TIMP1 are risk genes for bicuspid aortic valve and aortopathy in Turner syndrome. PLOS Genetics 14, e1007692 (2018). 505. Huang, A. C., Olson, S. B. & Maslen, C. L. A Review of Recent Developments in Turner Syndrome Research. J Cardiovasc Dev Dis 8, 138 (2021). 241 506. Kruszka, P. & Silberbach, M. The state of Turner syndrome science: Are we on the threshold of discovery? American Journal of Medical Genetics Part C: Seminars in Medical Genetics 181, 4–6 (2019). 242