Development of functional topography of the default mode subnetworks revealed by precision mapping

13 The default mode network (DMN) is a functionally and anatomically heterogeneous 14 network comprising distinct subsystem s that support internally directed processes 15 such as memory and self -related thought. Specialization and segregation of 16 functional brain networks are critical for cognitive maturation, yet how DMN 17 subnetworks develop remains unclear . Group-average approaches obscure 18 individual variability and blur spatial details, limiting precise characterization of 19 network topography. Using precision mapping in 547 participants aged 5 –21 years, 20 we delineated individualized DMN subnetworks and examined their development 21 across multiple aspects, including functional segregation and spatial topograp hy. 22 We found that DMN subnetworks become increasingly functionally segregated 23 from each other and from non -DMN networks with age, and exhibit greater 24 topographic distinctiveness through boundary sharpening and reduced spatial 25 overlap. Furthermore, we obser ved a selective spatial contraction of the memory -26 related subnetwork, which correlated with better episodic memory . Our findings 27 underscore the importance of characterizing multidimensional functional and 28 topographical features at the individual level to b etter understand typical 29 neurodevelopment and neurodevelopmental and psychiatric conditions. 30 31 32 33 34 35 36 37 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint

38 The default mode network (DMN) comprises interconnected and distributed brain 39 regions and is thought to be involved in internal mental processes, such as 40 remembering, thinking about future, and self/other -related thoughts1, 2. Disruptions 41 of the DMN have been shown in many neuropsychiatric and developmental 42 disorders3, 4, 5. Emerging evidence suggests that the DMN is a complex network that 43 consists of distinct subnetworks, each supporting specific cognitive processes 6, 7, 8. 44 During child development, the broader functional brain networks become 45 increasingly segregated —characterized by stronger within -network and weaker 46 between-network connectivity)9, 10, 11. The DMN as a whole becomes more internally 47 integrated while becoming increasingly segregated from other functional networks 48 during development 12, 13, 14 . This growing segregation supports cognitive 49 development by enabling more specialized network functioning13, 15. However, how 50 functional segregation between distinct DMN subnetworks develop in children 51 remains unknown. Delineating th e development of DMN subnetworks in a 52 neurotypical population is essential, as it provides a crucial baseline for 53 understanding alterations in related neuropsychiatric disorders. 54 To date, investigations of functional networks in children have largely relie d on 55 group-average (one -size-fits-all) parcellations. These group -average approaches 56 introduce spatial blurring and fail to capture individual -specific features of brain 57 organization. This limitation has motivated the growing use of precision mapping 58 approaches, which identify functional networks at the individual level. By capturing 59 individual variability, these methods offer improved reliability in linking brain 60 organization to behavior and symptoms 16, 17 . Furthermore, inter -individual 61 variability in functional brain networks may change across development 18. In 62 addition, studies using within -person precision mapping approaches have shown 63 that the individual topography of functional networks — including their size, shape, 64 and spatial location — differs markedly from the group average 19, 20 . These 65 topographical variations become more refined during youth development and are 66 associated with individual differences in cognition21, 22, with comparable alterations 67 reported in clinical populations 23, 24 . Thus, in addition to assessing within - and 68 between-network connectivity strength and other graph -based measures, this 69 approach also enables the examination of individual topographic features and spatial 70 extent of DMN subnetworks. 71 72 In the present study, we performed individual-level network parcellation in a cohort 73 of 547 participants aged 5–21 years, each with more than 21 minutes of resting-state 74 fMRI data. Using the Yeo 17 -network template as a reference, we identified three 75 distinct DMN subnetworks (DMNA, DMNB, DMNC) based on individual 76 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint functional connectivity profiles at the vertex level. A previous meta -analysis of the 77 functional pro files of the three DMN subnetworks found that DMNA was more 78 strongly associated with self -related processes, emotion and evaluation, as well as 79 social and mnemonic processes shared with the other two subnetworks 25. DMNB 80 was linked to mentalizing and social cognition, as well as story comprehension and 81 semantic/conceptual processing. In contrast, DMNC showed strong associations 82 with past and future autobiographical thought, episodic memory, and contextual 83 retrieval. To understand the development of these three DMN subnetworks, we first 84 examined functional segregation of the three DMN subnetworks, quantified by the 85 segregation index derived from the average within -network and between -network 86 functional connectivity. This allowed us to assess whether the subnetworks became 87 increasingly segregated from one another and from other non -DMN networks with 88 age. Second, we mapped the topography of the three subnetworks individually and 89 examined the topographical segregation, me asured by boundary refinement and 90 spatial differentiation between the subnetworks. Third, we examined age -related 91 changes in the surface area of each subnetwork to determine whether their cortical 92 representations expanded or contracted during development. After examining age 93 effects on these segregation indices and spatial extent, we tested whether these 94 developmental changes in brain network organization could predict corresponding 95 cognitive development. Given that DMNC is relatively specific to episodic memory, 96 we examined correlations between memory performance and DMNC metrics. 97 98

99 DMN subnetworks show increasing functional segregation with age 100 Here we employed a template -matching method to obtain individual parcellations 101 of three DMN subnetworks (Figure 1). After obtaining the individually mapped 102 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint Figure 1. Overview of the precision mapping method. A. First, we constructed the functional 103 connectivity profile for each vertex by calculating the Pearson correlation between that vertex and 104 all other vertices. Next, we computed the Dice similarity metric to quantify the spatial similarity 105 between each vertex’s connectivity profile and each of Yeo’s 17 networks. Finally, each verte x 106 was assigned to the network with which it showed the highest similarity. B. The left panel shows 107 the template of the DMN subnetworks based on the Yeo 17-network parcellation. The right panel 108 displays three representative participants selected from different age groups, illustrating individual 109 differences in the topographical distribution of the DMN subnetworks. 110 111 functional networks, we first calculated the within -network homogeneity and 112 compared it to that obtained using a group -level template. The homogeneity of 113 individual networks was significantly higher than that of the group template ( t546 = 114 15.42, p < 0.001), indicating the effectiveness of our precision mapping method. In 115 addition, we computed the Adjusted Rand Index between each subject’s parcellation 116 under a given threshold and the parcellations of the same and other subjects under 117 different thresholds. We observed that parcellations from different thresholds within 118 the same subject showed higher similarity than those between different subjec ts 119 (Figure S1 ), suggesting that our precision mapping method produces consistent 120

within individuals and is robust to parameter variations. 121 122 We quantified functional segregation by computing a segregation index based on 123 within-network and between -network functional connectivity. Within the DMN, 124 each subnetwork showed significantly increased functional segregation from the 125 other two subnetworks with age (Figure 2A, DMNA: β = 0.005, t541 = 6.37, pcorrected 126 < 0.001; DMNB: β = 0.008, t541 = 7.65, pcorrected < 0.001; DMNC: β = 0.008, t541 = 127 6.76, pcorrected < 0.001). These findings demonstrate progressive functional 128 segregation among all three DMN subnetworks across development. 129 In addition, we examined the age-related changes in functional segregation between 130 each DMN subnetwork and the other 16 large-scale functional networks. Similar to 131 segregation relative to other subnetworks of DMN, all DMN subnetworks showed 132 significant increases in segregation from the other 16 networks with age (Figure 2B, 133 DMNA: β = 0.010, t541 = 8.13, pcorrected < 0.001; DMNB: β = 0.007, t541 = 6.42, 134 pcorrected < 0.001; DMNC: β = 0.006, t541 = 6.61, pcorrected < 0.001). We also tested 135 whether increasing functional segregation of DMNC from the other two DMN 136 subnetworks and from all other 16 networks correlated with PSM scores, but no 137 significant correlations were found (all ps ≥ 0.42). These findings further highlight 138 the developmental emergence of functional differentiation of DMN subnetworks. 139 140 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint 141 142 Figure 2. DMN subnetworks show increasing functional segregation across development. A. The 143 segregation of each DMN subnetwork relative to other DMN subnetworks significantly increased 144 with age. B. The segregation of each DMN subnetwork from all other large -scale functi onal 145 networks also showed a significant age -related increase. Shaded bars reflect 95% confidence 146 intervals of the regression line. 147 148 Topographical Boundaries of DMN Subnetworks Become Sharper with Age 149 Here, we examined the spatial distribution of the three DMN subnetworks among 150 the participants using the density map (Figure 3A). We found that the highest density 151 regions of each subnetwork were primarily located within their respective canonical 152 cortical zones, but also extended into regions typically associa ted with other 153 subnetworks, albeit at lower densities. For instance, the posterior cingulate cortex 154 and precuneus regions were typically assigned to DMNA, although some individuals 155 showed assignments to DMNB. The anterior temporal regions were generally 156 associated with DMNB, but a substantial proportion of participants had anterior 157 temporal regions that were assigned to DMNA. Similarly, the ventral medial 158 prefrontal cortex was most often assigned to DMNA, yet assignments to DMNB and 159 DMNC were also observed in some individuals. This pattern reflects both individual 160 variability and their shared characteristics, with the three networks being distributed 161 yet adjacent. 162 Furthermore, we aimed to investigate whether the topographical boundaries between 163 DMN subnetworks become sharper with age. To do so, we first identified boundary 164 vertices based on their proximity (within 5 mm) to the borders between adjacent 165 subnetwork clusters (Figure 3B). Boundary sharpness was defined as the extent to 166 which these vertices showed stronger functional connectivity with their own 167 subnetwork compared to the neighboring one, indicating a clearer distinction 168 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint between adjacent subnetworks. Here, we present two examples illustrating a blurred 169 versus a sharp boundary (Figure 3C). We examined age-related changes in mean 170 boundary sharpness for each DMN subsystem, calculated by averaging the 171 sharpness values of the boundaries between the given subnetwork and the other two 172 subnetworks. DMNA showed a significant increase in mean boundary sharpness 173 with age ( β = 0.01, t541 = 4.91, pcorrected < 0.001, Figure 3D), driven by sharpness 174 increases at both its interfaces with DMNB ( β = 0.01, t541 = 3.65, pcorrected < 0.001) 175 and DMNC ( β = 0.01, t541 = 2.78, pcorrected = 0.010, Figure S2). We didn’t find 176 significant age-related changes in mean boundary sharpness in DMNB ( β = 0.006, 177 t541 = 1.77, pcorrected = 0.114) or DMNC (β = 0.002, t541 = 0.34, pcorrected = 0.731) after 178 FDR correction. This pattern of results suggests that DMNA boundaries undergo 179 progressive sharpening during development, whereas DMNB and DMNC 180 boundaries remain relatively stable. 181 182 Figure 3 Topographical boundaries of DMN subnetworks became sharper with development. A. 183 The density maps indicated the proportion of subjects for whom a giv en vertex was assigned to 184 each of the three DMN subnetworks. The spatial locations of the three DMN subnetworks were 185 primarily centered around their respective centroids within each cortical zone, but they also 186 extended into regions typically associated with other subnetworks. B. Two clusters from different 187 DMN subnetworks were selected as examples to illustrate how boundary sharpness was quantified. 188 For each border vertex between the two clusters, we identified all vertices located within 5 mm 189 and used them for the analysis. C. Two example subjects illustrating boundary sharpness between 190 DMNA and DMNB. Red outline indicates the DMNB cluster, and the yellow outline indicates the 191 DMNA cluster. The color bar represents the functional connectivity strength of each vertex to the 192 mean time series of the DMNA cluster. The top panel shows a subject with a blurred boundary, 193 where vertices along the border exhibit highly similar connectivity to both subnetworks. The 194 bottom panel shows a subject with a sharper boundary , where border vertices display distinct 195 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint connectivity profiles between the two subnetworks. Note that while most cluster vertices are 196 shown for visualization, sharpness was calculated only for border vertices within 5 mm (panel B). 197 D. The scatter plot show s that DMNA boundaries with both other subnetworks became 198 significantly sharper with age. 199 200 Spatial overlap of DMN subnetworks decreases with age 201 After demonstrating the refinement of boundaries with age, we next sought to 202 investigate the spatial differentiation among DMN subnetworks. We defined spatial 203 overlap between DMN subnetworks as vertices that simultaneously showed high 204 functional connectivity with the mean time series of two subnetworks, where "high" 205 was defined as connectivity exceeding t he subject -specific mean by at least one 206 standard deviation. We then examined how the surface area of these overlapping 207 regions changed with age. Among the subnetworks, the spatial overlap between 208 DMNA and DMNB exhibited the highest surface area proportion , and this overlap 209 was significantly negatively correlated with age ( β = -0.003, t541 = -2.53, pcorrected = 210 0.017, Figure 4A). Similarly, the overlap between DMNB and DMNC also showed 211 a significant negative correlation with age (β = -0.002, t541 = -2.71, pcorrected = 0.017, 212 Figure 4B). The spatial overlap between DMNA and DMNC did not show a 213 significant relationship with age ( β = 6.96E -05, t541 = 0.07, pcorrected = 0.947). 214 Additional analyses across different correlation thresholds for defining high spatial 215 overlap (from the subject -specific mean, to two standard deviations above) 216 confirmed the robustness of the age-related decrease in overlap between DMNA and 217 DMNB (Figure S3). These findings indicate that the spatial distribution of the DMN 218 subnetworks becomes more differentiated during development, especially between 219 DMNA and DMNB. 220 221 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint 222 Figure 4. Overlap of DMN subnetworks decreased with age. A. Two examples illustrate the spatial 223 overlap between DMNA and DMNB from a younger and an older participant, respectively. Only 224 the left hemisphere is displayed for visual clarity. Analyses were conducted on both hemispheres. 225 The accompanying scatter plot demonstrates a significant decrease in overlap with age. B. Two 226 examples show the spatial overlap between DMNB and DMNC from a younger and an older 227 participant. The scatter plot again indicates a significant age-related decline in overlap. 228 Surface area of DMNC decreases with age 229 We first examined age -related changes in the surface area of DMN subnetworks 230 using a GLM, controlling for sex and scanning sites. After correction for multiple 231 comparisons, we observed a significant age -related decrease in the surface area of 232 the DMNC subn etwork ( β = –0.04, t541 = –4.00, pcorrected < 0.001). No significant 233 associations with age were found for the DMNA ( β = 0.01, t541 = 0.58, pcorrected = 234 0.63) or DMNB (β = –0.01, t541 = –0.94, pcorrected = 0.45) subnetworks (Figure 5A). 235 When treating the DMN as a whole network, its overall surface area does not exhibit 236 significant age-related changes ( β = -0.02, t541 = -1.53, pcorrected = 0.19). To further 237 illustrate age -related changes in DMNC surface ar ea, subjects were divided into 238 three age groups based on age: children (≤12 years, n=147), adolescents (12 -18 239 years, n=262), and adults (≥18 years, n=138). The corresponding density maps show 240 that as age increases, the surface area of the DMNC gradually de creases and 241 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint becomes more spatially constrained (Figure 5B). We also observed a significant age-242 related reduction of DMNC in the ventromedial prefrontal cortex (vmPFC), a region 243 typically associated with the DMNA (Figure 5B). This pattern suggests a spatial 244 reorganization of DMN subnetworks during development, with DMNC territory 245 potentially redistributing to other subnetworks in adulthood. 246 To further characterize the spatial organization of age -related changes in 247 individualized DMNC, we compared each participant's DMNC map with the group-248 level template as a reference. This approach enabled us to identify which networks 249 occupy the regions where the DMNC contracts during development, relative to the 250 canonical network spatial distribution . We first quantified t he spatial overlap 251 between each participant's DMNC and the group -level DMNA. As expected, the 252 proportion of individual DMNC overlapping with group-level DMNA (overlap area 253 / individual DMNC area) significantly decreased with age (β = –0.001, t403 = –4.92, 254 p < 0.001, Figure 5C) . This indicates that regions initially occupied by DMNC in 255 younger children are progressively reassigned to canonical DMNA territory. In 256 contrast, when examining the overlap between youth DMNC and adult group -level 257 DMNB, nearly all subjects showed minimal to no overlap (Figure S4), confirming 258 that the developmental changes in DMNC are relatively spatially specific to DMNA-259 associated regi ons. Collectively, these findings demonstrate age -related spatial 260 reorganization among DMN subnetworks, whereby cortical territories undergo 261 systematic reassignment during development. 262 263 To further examine whether the DMNC subnetwork contracting is associat ed with 264 the development of episodic memory, we also examined the relationship between 265 PSM scores and the surface area of DMN subnetworks, controlling for age, sex, and 266 scanning site. A significant positive association was observed between age and PSM 267 score (β = 0.71, t396 = 8.05, p < 0.001), indicating that episodic memory performance 268 improves with development. We found that the surface area of the DMNC 269 subnetwork was significantly negatively correlated with PSM scores ( β = -1.17, t395 270 = -2.27, p = 0.02), indicating that a reduced surface area was associated with better 271 episodic memory performance. These findings suggest that during development, 272 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint DMNC exhibits a more compact cortical organization compared to other 273 subnetworks, which may underlie its link to memory performance (Figure 5D). 274 275 Figure 5. Surface area of DMNC decreased with age and memory performance. A. This bar plot 276 illustrates the effect of age on the surface area of DMN subnetworks, based on a general linear 277 model in which age was treated as th e independent variable and surface area as the dependent 278 variable, while controlling for confounding factors (sex, site, etc.). The width of each bar reflects 279 the regression coefficient for age. Asterisks indicate levels of statistical significance after F DR 280 correction: **p < 0.001; *p < 0.01; p < 0.05. B. Panel B shows the density maps for different age 281 groups. Density values represent the proportion of subjects within each group in which a given 282 vertex was assigned to the DMNC. C. The left panel shows two example subjects, illustrating the 283 spatial overlap between individual DMNC (blue) and the group -level DMNA from template 284 (yellow). The scatter plot demonstrates an age-related decrease in the overlap proportion (overlap 285 area / individual's total DMNC area), indicating that as DMNC shrinks during development, it is 286 primarily the regions that correspond to canonical DMNA that are lost. D. A negative association 287 was observed between the DMNC surface area proportion (individual’s DMNC area / individual’s 288 total cortical area) and Picture Sequence Memory performances. 289 290

291 In this study, we systematically examined the devel opmental segregation and 292 specialization of DMN subnetworks in 547 participants aged 5 -21 years, using 293 precision mapping to obtain individualized parcellations of DMN subnetworks. Our 294 findings reveal developmental segregation occurring across multiple dimen sions. 295 First, the three DMN subnetworks exhibited increased functional segregation from 296 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint each other and from all other networks with age. Second, topographical segregation 297 was observed: boundaries between DMNA and the other two subnetworks became 298 significantly sharper with age, accompanied by reduced spatial overlap between 299 DMNA and DMNB. Third, we observed a selective contraction of DMNC that 300 correlated with better episodic memory, with the contracted regions predominantly 301 overlapping with canonical DMNA te rritory. Together, these findings provide a 302 comprehensive characterization of DMN subnetwork development, encompassing 303 progressive functional and topographical segregation, and selective spatial 304 focalization. 305 306 How to identify the DMN subnetworks remains a n area of active research 26, 27. A 307 commonly used approach is to divide it into three subnetworks, as proposed by 308 Andrews-Hanna et al. (2010 , 2014)7, 25, with similar parcellations also re ported by 309 Yeo et al. (2011) 28. DMNA is associated with self -related processes, emotion, and 310 evaluation, while also sharing social and mnemonic functions with DMNB and 311 DMNC. DMNB is linked to social processing, whereas DMNC shows relative 312 specificity to episodic memory 25. Based on these prior findings, we decided to 313 parcellate the DMN into three subsystems, following Yeo et al. Previous studies 314 have examined the functional segregation of the DMN by treating it as a single 315 network. For example, one study reported increased within-network connectivity of 316 the entire DMN during rest from childhood to adolescence 12. Similar effects have 317 been observed in children during task states 13. More recently, a large -scale 318 population study found that functional segregation of the canonical DMN, measured 319 using the same index, increases rapidly with age and peaks at the end of the third 320 decade14. Only one previous study has investigated the segregation of DMN 321 subnetworks in a developmental sample, comparing the modularity of DMN 322 subsystems between individuals with autism and neurotypical controls29. However, 323 this study employed an ROI-to-ROI approach with relatively limited ROI coverage 324 and the same ROI definitions across all participants, which could have overlooked 325 inter-individual variations and thereby contributed to the finding of no marked age -326 related changes in the neurotypical group29. Our approach overcomes this limitation 327 and reveals developmental effects showing that the DMN, a heterogeneous network 328 as established by functional –anatomical evidence 7, 25, 28, 30, 31, 32 , becomes 329 increasingly heterogeneous and functionally segregated over the course of 330 development. Segregation of ass ociative networks is consistently associated with 331 enhanced cognitive performance, while the loss of such segregation in later life has 332 been implicated in cognitive decline 9, 33, 34, 35, 36, 37, 38 . DMN subnetworks share a 333 common organizational motif that supports internally generated representations, 334 with each network specialized for distinct processing domains 27, and their 335 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint segregation with development may reflect the enhancement of specific cognitive 336 functions. 337 338 We also observed an age-related increase in segregation in the topography of DMN 339 subnetworks. Our precision mapping results reveal individual parcellations with 340 clear and specific spatial boundaries. Given that the three DMN subnetworks are 341 spatially adjacent to one another we examined the age -related effects on boundary 342 sharpness between subnet works and found that the boundaries around DMNA 343 become progressively sharper across development. At the same time, we found that 344 the spatial overlap between DMNA and DMNB decreases with age, suggesting that 345 DMNA gradually emerges as a more spatially distin ct and clearly delineated 346 subnetwork. Notably, DMNA showed the most pronounced age -related changes in 347 topographical segregation among all three DMN subnetworks. DMNA encompasses 348 the so -called midline core regions of the DMN, including the anterior medial 349 prefrontal cortex, angular gyrus, and posterior cingulate cortex, which are among 350 the most consistently engaged regions within the network 7, 25 . Previous group -351 averaged studies examining functional connectivity between pairs of DMN nodes 352 have reported age-related increases primarily in midline core regions, which is 353 consistent with our findings 12. Convergent structural and functional connectivity 354 analyses in prior work also suggest that PCC –mPFC connectivity (largely 355 overlapping with DMNA) represents the most immature link within the DMN in 356 children, further highlighting the pro nounced age -related effects 39. Importantly, 357 previous studies have demonstrated that behavioral phenotypes across cognition, 358 personality, and emotion can be predicted by individual -specific network 359 topography with modest accuracy, comparable to predictions based on connectivity 360 strength40. Our study advances this line of research by introducing two novel 361 measures of topographical segregation —boundary sharpness and spatial overlap —362 that uncover new insights into the developmental segregation of DMN subnetworks. 363 These topographical measures, enabled by within -person precision mapping, may 364 provide new biomarkers for predicting behavioral outcomes and developmental 365 trajectories at the individual level, complementing traditional connectivity -based 366 approaches. 367 Finally, we examined age-related effects on the surface area of DMN subnetworks. 368 While many studies have investigated developmental trajectories of total or regional 369 cortical surface area based on group averages 41, 42, 43 , research focusing on the 370 surface area of individual functional networks remains limited. One study on human 371 lifespan functional connectome changes, using personalized ma pping, reported a 372 slight expansion of the whole DMN size during the first month of life but no 373 significant change thereafter14. These findings are consistent with ours. Specifically, 374 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint when we first tested the proportion of the whole DMN surface area relative to the 375 total cortical surface area, we found no significant age -related effects. This result 376 highlights that treating the DMN as a single entity may obscure the heterogeneous 377 developmental trajectories of its subnetworks. When we examined each subnetwork 378 separately, we found that the surface area of DMNC progressively decreased with 379 age. This reduction was predominantly localized to regions that were more 380 frequently assigned to canonical DMNA, indicating spatial reorganization within 381 DMN subnetworks during development. Critically, this contraction was correlated 382 with episodic memory performance, suggesting that its areal specialization supports 383 ongoing cognitive development. 384 Our study has several limitations. First, we divided the DMN into three subnetworks; 385 however, the exact number of DMN subnetworks remains unclear due to the 386 complexity of DMN functio n, and our study does not address this question, which 387 warrants further investigation. Second, when examining DMN -related cognitive 388 performance—for example, the close association between the DMNA and self -389 related thoughts —we lacked corresponding behavioral measures to explore these 390 relationships, which may require a broader behavioral phenotype. Finally, our 391

of cognitive functions associated with DMN subnetworks primarily relies 392 on previous literature rather than functional validation using speci fic tasks, which 393 should be addressed in future studies. 394 In summary, our study establishes a novel, multi -dimensional framework for 395 investigating DMN subsystem segregation during development by incorporating 396 measures of functional segregation, topographical segregation, and spatial extent 397 alterations through individualized precision mapping. We demonstrate that 398 progressive functional segregation of DMN subnetworks is accompanied by 399 boundary sharpening and spatial differentiation —manifested as reduced overlap 400 between subnetworks—as well as selective spatial refinement of DMNC that relates 401 to episodic memory performance. By transcending group -averaged approaches and 402 examining individual DMN subnetworks, our findings emphasize the importance of 403 characterizing mu lti-dimensional functional and topographical features at the 404 individual level to elucidate network segregation and specialization processes 405 essential for typical neurodevelopment and relevant to neuropsychiatric and 406 developmental disorders. 407 408

409 Data 410 All participants are from the Human Connectome Project in Development (HCP-D). 411 HCP-D is a cross-sectional project including participants aged 5 –21 years. HCP-D 412 MRI data were acquired at four sites: Harvard University, University of California–413 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint Los Angeles, University of Minnesota, and Washington University in St. Louis 44. 414 Our analysis is based on Release 2.0 (N = 625). All participants provided informed 415 assent or consent. Individuals exhibiting excessive head motion (frame -wise 416 displacement > 0.25 mm for more than one -third of the scanning duration) were 417 excluded from further analysis. The final sample consisted of 547 subjects aged 5 to 418 21 years (mean age = 15 ± 3.8 years, 285 females). The sample was predominantly 419 White (63.4%), followed by individuals identifying as more than one race (15.8%), 420 Black or African American (10.8%), Asian (7.5%), and Other (2.5%). The median 421 annual household income of the sample was $119,000. 422 Image acquisition 423 MRI scans were acquired at 4 sites on 3T Siemens Prisma scanners with a 32 -424 channel Prisma head coil. T1w images were acquir ed with a 3D multi -echo 425 MPRAGE (Magnetization Prepared Rapid Gradient Echo) sequence with an in -426 plane acceleration factor of 2. The scanning parameters included: repetition time of 427 2,500 ms, echo times of 1.8, 3.6, 5.4, and 7.2 ms, inversion time of 1,000 ms, flip 428 angle of 8 degrees, 208 slices, and a voxel resolution of 0.8 mm isotropic. The fMRI 429 scans are acquired with a 2D multiband gradient -recalled echo (GRE) echo -planar 430 imaging (EPI) sequence (MB8, TR/TE = 800/37 ms, flip angle = 52) and 2.0 mm 431 isotropic voxels covering the whole brain (72 oblique-axial slices). To mitigate bias 432 associated with a specific phase-encoding direction, functional scans were acquired 433 in pairs, with one run using anterior-to-posterior (AP) and the other using posterior-434 to-anterior (PA) phase encoding. Participants aged 8 years and older underwent 26 435 minutes of resting -state scanning, consisting of four runs of 6.5 minutes each. For 436 younger participants aged 5 to 7 years, a shorter protocol was used, comprising six 437 runs of 3.5 minutes each, totaling 21 minutes. 438 Image preprocessing 439 In our study, we utilized preprocessed T1-weighted and fMRI images obtained from 440 the NIMH Data Archive (NDA). The data were processed using the HCP -441 recommended preprocessing pipeline, fMRIVolume, which includes realignment, 442 correction for spatial distortio ns, and registration of the functional images to the 443 structural images with minimal smoothing (only due to interpolation, with no 444 explicit smoothing). Artifact removal was performed using sICA+FIX, which 445 regresses out motion - and artifact -related ICA compo nents. Finally, fMRISurface 446 was used to project gray matter voxels onto the cortical ribbon and register them to 447 the fs_LR 32k surface space. 448 Precision mapping 449 We employed a template -matching approach similar to that used in previous 450 studies45 . Compared with other precision mapping methods—such as Infomap, non-451 negative matrix factorization, and Overlapping MultiNetwork Imaging mapping —452 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint template matching requires less scan time to obtain reliable personalized maps 45. In 453 this study, we used Yeo’s 17-network parcellation as the template. For each vertex, 454 we computed its functional connectivity profile with all the other vertices a nd 455 retained the top 5% of connections. We also examined the top 1% and 3% of 456 connections and compared the consistency of parcellation across different thresholds 457 using the Adjusted Rand Index, to ensure that the results were not driven by the 458 specific choi ce of threshold. We then calculated the Dice similarity coefficient 459 between each vertex’s connectivity profile and each of the 17 Yeo networks, 460 generating a vertex -by-network similarity matrix. Each vertex was assigned to the 461 network with which it showed t he highest similarity. Subsequently, we applied the 462 Connectome Workbench (https://www.humanconnectome.org/- 463 software/connectome-workbench) command to identify and remove small isolated 464 clusters smaller than 40 mm2. These small clusters were then spatially reassigned to 465 the nearest neighboring network to ensure spatial contiguity. Finally, we obtained 466 the individual network parcellation for each subject. 467 Homogeneity of functional networks 468 To evaluate the effectiveness of our precision mapping approach, we calculated the 469 within-network homogeneity, following procedures used in previous studies 40, 46. 470 Specifically, we computed the Pearson correlation coefficients among all vertices 471 within each network and then averaged these values across all networks for both the 472 precision-mapped parcellation and the group-level Yeo's 17-network parcellation. A 473 paired t -test was conducted to assess whether within -network homogeneity was 474 significantly higher for individual -specific parcellations compared to the group -475 level parcellation. 476 Functional segregation of DMN subnetworks 477 To investigate the functional organization of DMN subnetworks, we calculated the 478 functional segregation index, a metric that has been used in previous studies 34. We 479 first extracted the time series of all vertices and computed pairwise Pearson 480 correlations, resulting in a vertex-wise functional connectivity matrix. Next, we 481 applied Fisher’s z-transformation to the connectivity values. For each network, we 482 calculated the segregation index as the difference between its average within -483 network connectivity (FC W) and its average between-network connectivity (FC B), 484 normalized by the within -network connectivity, as shown in the formula. We 485 computed the within -DMN segregation (each DMN subnetwork vs. the other two 486 DMN subnetworks) and whole-brain segregation (each DMN subnetwork vs. all 16 487 other networks). 488 Functional segregation = FC1 − FC3 FC1 489 Density map of DMN subnetworks 490 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint To examine individual differences in the topography of DMN subnetworks, we 491 computed the density map of the DMN subnetworks. The density map was generated 492 by calculating the proportion of individuals for whom a given vertex belonged to a 493 specific functional network. Higher values indicate that a greater proportion of 494 participants assigned that vertex to the given network. To better visualize age-related 495 changes in the spatial distribution of functional networks, subjects were divided into 496 three age groups based on age: children (≤12 years, n=147), adolescents (12 -18 497 years, n=262), and adults (≥18 years, n=138). For each group, we generated density 498 maps to visualize the spatial location of the network. 499 Sharpness of the boundary between DMN subnetworks 500 To quantify the topographical distinctiveness of DMN subnetworks, we calculated 501 the boundary sharpness between adjacent networks. First, we used the Connectome 502 Workbench co mmand to identify spatially contiguous clusters within each 503 functional network. Then, we generated borders outlining each cluster, and extracted 504 the vertex indices corresponding to these borders. For each cluster, we computed its 505 mean time series and ident ified nearby vertices within a 5 mm distance of each 506 border vertex. These nearby vertices were classified as either inside vertices 507 (belonging to the same cluster) or outside vertices (belonging to different clusters). 508 We then calculated the Pearson correl ation between the cluster’s mean time series 509 and the time series of each nearby border vertex. To assess boundary sharpness, we 510 computed Cohen’s d for the difference in correlation values between inside and 511 outside vertices. The outside vertices were furth er categorized into different 512 functional networks based on individual -level precision mapping. This allowed us 513 to compute boundary sharpness between each cluster and each adjacent functional 514 network. If the number of outside vertices belonging to a given n etwork was fewer 515 than 5, the sharpness was not calculated for that network pair. Finally, we averaged 516 the sharpness values of all clusters belonging to the same network to obtain network-517 to-network sharpness estimates. We separately computed the sharpness between 518 each DMN subnetwork and the other two DMN subnetworks. 519 Spatial overlap between DMN subnetworks 520 To quantify spatial overlap between DMN subnetworks, we built upon the 521 individualized precision mapping results. While precision mapping assigns each 522 vertex exclusively to a single subnetwork through a winner -take-all procedure, 523 functional overlap between subnetworks may exist in practice. To capture this 524 potential overlap, we adopted the following approach. For each subnetwork, we first 525 computed the mean BOLD time series by averaging all vertices within its subject -526 specific precision mapping -defined spatial extent. We then calculated Pearson 527 correlation coefficients between this mean time series and every cortical vertex 528 across the brain. The resulting correlation maps underwent z-score normalization to 529 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint standardize the distribution. To allow vertices to be associated with multiple 530 subnetworks based on their functional connectivity strength, each subnetwork was 531 redefined as the set of vertices exceeding a functional-connectivity threshold, which 532 we systematically varied from 0.0 to 2.0. Primary results are reported at a threshold 533 of 1, with additional results provided in the Supplementary Materials. We selected 534 this threshold because, after z -score normalization, a value of 0 corresponds to the 535 mean correlation and 2 represents two standard deviations above the mean —536 indicative of high connectivity. Higher thresholds retain very few vertices and 537 produce minimal overlap across regions. For each subnetwork pair, overlapping 538 vertices satisfying the threshold criterion for both subnetworks were identified. We 539 calculated the surface area of each cortical vertex to determine both the overlap area 540 and the combined area of the two subnetworks. The overlap ratio, compute d as the 541 overlap area divided by the combined area, provided a normalized measure of spatial 542 differentiation. 543 Surface area of DMN subnetworks 544 To quantify the spatial extent of each DMN subnetwork, we first calculated the 545 surface area (in mm²) for each cort ical vertex. The relative proportion of each 546 network was then calculated by summing the surface areas of its constituent vertices 547 and dividing by the total cortical surface area. After examining age effects on surface 548 area proportion, we further examined d evelopmental changes in the topography of 549 DMN subnetworks. To characterize how the spatial distribution of individualized 550 network parcellations changes with age, we compared individual maps with the Yeo 551 template as a reference, which represents the canonic al spatial distribution of each 552 network. By comparing individual network maps against this reference on a vertex-553 by-vertex basis, we identified vertices that showed different subnetwork 554 assignments across development. For example, by comparing an individua l's 555 DMNC map against the group-level DMNA and DMNB maps, we could determine 556 whether regions that contract from DMNC during development are subsequently 557 occupied by other DMN subnetworks (i.e., DMNA or DMNB) in the reference map. 558 559 Assessing Age-Related Effects on Brain Measures 560 We employed a general linear model (GLM) to examine the relationship between 561 age and brain measures, including functional segregation, boundary sharpness, 562 spatial overlap and surface area of DMN subnetworks. In these models, age was 563 treated as the independent variable, while the brain measures served as dependent 564 variables. Sex and scanning sites were included as covariates. To account for 565 multiple comparisons, we applied false discovery rate (FDR) correction to the p -566 values associated with the age variable. 567 568 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint DMN subnetwork development and cognitive development 569 We examined whether developmental changes in the segregation and spatial extent 570 of DMN subnetworks are associated with cognitive development. We used scores 571 from the Picture Seq uence Memory (PSM) test as a cognitive measure of episodic 572 memory performance. We expected the PSM scores to increase with age, and 573 correlate with the memory -linked DMNC network. In the PSM task, a series of 574 illustrated objects and activities were presente d on an iPad in a specific order. The 575 sequence length varied from 6 to 18 pictures depending on the participant's age. 576 Participants were asked to recall the correct order of the pictures and received points 577 for each correctly recalled adjacent pair. This t est assesses the acquisition, storage, 578 and effortful recall of newly learned information. Higher scores indicate better 579 episodic memory performance relative to age-matched peers. 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint 612 613 614 615

616 1. Buckner RL, Andrews-Hanna JR, Schacter DL. The brain's default network: Anatomy, function, 617 and relevance to disease. Ann N Y Acad Sci 1124, 1-38 (2008). 618 619 2. Menon V. 20 years of the default mode network: A review and synthesis. Neuron 111, 2469-2487 620 (2023). 621 622 3. Anticevic A, Cole MW, Murray JD, Corlett PR, Wang XJ, Krystal JH. The role of default 623 network deactivation in cognition and disease. Trends Cogn Sci 16, 584-592 (2012). 624 625 4. Padmanabhan A, Lynch CJ, Schaer M, Menon V. The Default Mode Network in Autism. 626 Biological Psychiatry: Cognitive Neuroscience and Neuroimaging 2, 476-486 (2017). 627 628 5. Konrad K, Eickhoff SB. Is the ADHD brain wired differently? A review on structural and 629 functional connectivity in attention deficit hyperactivity disorder. Hum Brain Mapp 31, 904-916 630 (2010). 631 632 6. Gordon EM, et al. Default-mode network streams for coupling to language and control systems. 633 Proc Natl Acad Sci 117, 17308-17319 (2020). 634 635 7. Andrews-Hanna JR, Reidler JS, Sepulcre J, Poulin R, Buckner RL. Functional-Anatomic 636 Fractionation of the Brain's Default Network. Neuron 65, 550-562 (2010). 637 638 8. Braga RM, Buckner RL. Parallel Interdigitated Distributed Networks within the Individual 639 Estimated by Intrinsic Functional Connectivity. Neuron 95, 457-471.e455 (2017). 640 641 9. Sherman LE, Rudie JD, Pfeifer JH, Masten CL, McNealy K, Dapretto M. Development of the 642 Default Mode and Central Executive Networks across early adolescence: A longitudinal study. 643 Dev Cogn Neurosci 10, 148-159 (2014). 644 645 10. Chai XJ, Ofen N, Gabrieli JDE, Whitfield-Gabrieli S. Selective development of anticorrelated 646 networks in the intrinsic functional organization of the human brain. J Cogn Neurosci 26, 501-647 513 (2014). 648 649 11. Gu S, et al. Emergence of system roles in normative neurodevelopment. Proc Natl Acad Sci 112, 650 (2015). 651 652 12. Fan F, et al. Development of the default-mode network during childhood and adolescence: A 653 longitudinal resting-state fMRI study. Neuroimage 226, 117581-117581 (2021). 654 655 13. Wang C, Hu Y, Weng J, Chen F, Liu H. Modular segregation of task-dependent brain networks 656 contributes to the development of executive function in children. Neuroimage, 116334-116334 657 (2019). 658 659 14. Sun L, et al. Human lifespan changes in the brain's functional connectome. Nat Neurosci, 660 (2025). 661 662 15. Baum GL, et al. Modular Segregation of Structural Brain Networks Supports the Development of 663 Executive Function in Youth. Curr Biol 27, 1561-1572.e1568 (2017). 664 665 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint 16. Kraus B, Zinbarg R, Braga RM, Nusslock R, Mittal VA, Gratton C. Insights from personalized 666 models of brain and behavior for identifying biomarkers in psychiatry. Neurosci Biobehav Rev 667 152, 105259-105259 (2023). 668 669 17. Gratton C, et al. Defining Individual-Specific Functional Neuroanatomy for Precision Psychiatry. 670 Biol Psychiatry 88, 28-39 (2020). 671 672 18. Demeter DV, et al. Precision functional mapping reveals less inter-individual variability in the 673 child vs. adult human brain. In: bioRxiv). 20250724 edn (2025). 674 675 19. Gordon EM, et al. Precision Functional Mapping of Individual Human Brains. Neuron 95, 791-676 807.e797 (2017). 677 678 20. Seitzman BA, et al. Trait-like variants in human functional brain networks. Proc Natl Acad Sci 679 116, 22851-22861 (2019). 680 681 21. Cui Z, et al. Individual Variation in Functional Topography of Association Networks in Youth. 682 Neuron 106, 340-353.e348 (2020). 683 684 22. Keller AS, et al. Personalized functional brain network topography is associated with individual 685 differences in youth cognition. Nat Commun 14, (2023). 686 687 23. Lynch CJ, et al. Frontostriatal salience network expansion in individuals in depression. Nature 688 633, 624-633 (2024). 689 690 24. Dickie EW, et al. Personalized Intrinsic Network Topography Mapping and Functional 691 Connectivity Deficits in Autism Spectrum Disorder. Biol Psychiatry 84, 278-286 (2018). 692 693 25. Andrews-Hanna JR, Smallwood J, Spreng RN. The default network and self-generated thought: 694 component processes, dynamic control, and clinical relevance. Ann N Y Acad Sci 1316, 29-52 695 (2014). 696 697 26. Smallwood J, Bernhardt BC, Leech R, Bzdok D, Jefferies E, Margulies DS. The default mode 698 network in cognition: a topographical perspective. Nat Rev Neurosci 22, 503-513 (2021). 699 700 27. Buckner RL, DiNicola LM. The brain's default network: updated anatomy, physiology and 701 evolving insights. Nat Rev Neurosci 20, 593-608 (2019). 702 703 28. Thomas Yeo BT, et al. The organization of the human cerebral cortex estimated by intrinsic 704 functional connectivity. J Neurophysiol 106, 1125-1165 (2011). 705 706 29. Bathelt J, Geurts HM. Difference in default mode network subsystems in autism across childhood 707 and adolescence. Autism 25, 556-565 (2021). 708 709 30. Doucet G, et al. Brain activity at rest: a multiscale hierarchical functional organization. J 710 Neurophysiol 105, 2753-2763 (2011). 711 712 31. Greicius MD, Supekar K, Menon V, Dougherty RF. Resting-state functional connectivity reflects 713 structural connectivity in the default mode network. Cereb Cortex 19, 72-78 (2009). 714 715 32. Schmahmann JD, et al. Association fibre pathways of the brain: parallel observations from 716 diffusion spectrum imaging and autoradiography. Brain 130, 630-653 (2007). 717 718 33. Wang C, Hu Y, Weng J, Chen F, Liu H. Modular segregation of task-dependent brain networks 719 contributes to the development of executive function in children. Neuroimage 206, (2020). 720 721 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint 34. Chan MY, Park DC, Savalia NK, Petersen SE, Wig GS. Decreased segregation of brain systems 722 across the healthy adult lifespan. Proc Natl Acad Sci 111, E4997-5006 (2014). 723 724 35. Fair DA, et al. Development of distinct control networks through segregation and integration. 725 Proc Natl Acad Sci 104, 13507-13512 (2007). 726 727 36. Lopez KC, Kandala S, Marek S, Barch DM. Development of Network Topology and Functional 728 Connectivity of the Prefrontal Cortex. Cereb Cortex 30, 2489-2505 (2020). 729 730 37. Ng KK, Lo JC, Lim JKW, Chee MWL, Zhou J. Reduced functional segregation between the 731 default mode network and the executive control network in healthy older adults: A longitudinal 732 study. Neuroimage 133, 321-330 (2016). 733 734 38. Keller AS, Sydnor VJ, Pines A, Fair DA, Bassett DS, Satterthwaite TD. Hierarchical functional 735 system development supports executive function. Trends Cogn Sci 27, 160-174 (2023). 736 737 39. Supekar K, Uddin LQ, Prater K, Amin H, Greicius MD, Menon V. Development of functional 738 and structural connectivity within the default mode network in young children. Neuroimage 52, 739 290-301 (2010). 740 741 40. Kong R, et al. Spatial Topography of Individual-Specific Cortical Networks Predicts Human 742 Cognition, Personality, and Emotion. Cereb Cortex 29, 2533-2551 (2019). 743 744 41. Vijayakumar N, et al. Brain development during adolescence: A mixed-longitudinal investigation 745 of cortical thickness, surface area, and volume. Hum Brain Mapp 37, 2027-2038 (2016). 746 747 42. Bethlehem RAI, et al. Brain charts for the human lifespan. Nature 2022, (2022). 748 749 43. Asato MR, Terwilliger R, Woo J, Luna B. White matter development in adolescence: A DTI 750 study. Cereb Cortex 20, 2122-2131 (2010). 751 752 44. Somerville LH, et al. The Lifespan Human Connectome Project in Development: A large-scale 753 study of brain connectivity development in 5–21 year olds. Neuroimage 183, 456-468 (2018). 754 755 45. Hermosillo RJM, et al. A precision functional atlas of personalized network topography and 756 probabilities. Nat Neurosci 27, 1000-1013 (2024). 757 758 46. Gordon EM, Laumann TO, Adeyemo B, Huckins JF, Kelley WM, Petersen SE. Generation and 759 Evaluation of a Cortical Area Parcellation from Resting-State Correlations. Cereb Cortex 26, 760 288-303 (2016). 761 762 763 764 765 766 Acknowledgments 767 Fonds de recherche du Québec FRQNT 2021-NC-283571 (XJC) 768 Canada First Research Excellence Fund (XJC) 769 Canada Research Chairs program (XJC) 770 Jeanne Timmins Fellowship (YH) 771 Author contributions: 772 Conceptualization: YH, JK, XJC 773 Methodology: YH, JK 774 Investigation: YH, JK, HJ, XJC 775 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint Visualization: YH 776 Supervision: XJC 777 Writing—original draft: YH 778 Writing—review & editing: YH, JK, HJ, XJC 779 Competing interests 780 The authors declare that they have no competing interests. 781

& Correspondence 782 Ying He ([email protected]) 783 784 Data and materials availability: 785 Raw data are freely available through the Human Connectome Project, upon completion of a 786 data-usage agreement: [https://nda.nih.gov/general-query.html?q=query=featured-datasets 787 :HCP%20Aging%20and%20Development]. All the analysis and visualization code are 788 available on GitHub: https://github.com/Chai-Neuro-Lab/Segregation-of-DMN-subnetworks 789 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted February 25, 2026. ; https://doi.org/10.64898/2026.02.24.707618doi: bioRxiv preprint