pith. sign in

arxiv: 2510.12751 · v2 · submitted 2025-10-14 · 🧬 q-bio.NC

Non-linear associations of amyloid-β with resting-state functional networks and their cognitive relevance in a large community-based cohort of cognitively normal older adults

Junjie Wu , Benjamin B Risk , Taylor A James , Nicholas Seyfried , David W Loring , Felicia C Goldstein , Allan I Levey , James J Lah
show 1 more author
Deqiang Qiu
This is my paper

Pith reviewed 2026-05-18 07:16 UTC · model grok-4.3

classification 🧬 q-bio.NC
keywords amyloid-betafunctional connectivityresting-state fMRIdefault mode networkprecuneusAlzheimer's diseasecognitive performancecommunity cohort
0
0 comments X

The pith

Amyloid-beta shows inverted U-shaped links to connectivity in the precuneus network and ventral default mode network among cognitively normal older adults.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper investigates non-linear relationships between cerebrospinal fluid amyloid-beta levels and resting-state functional connectivity across 14 large-scale brain networks in 968 cognitively normal older adults drawn from the community. It reports inverted U-shaped patterns specifically in the precuneus network and ventral default mode network, patterns absent in the dorsal default mode network. These connectivity measures in amyloid-related networks also associate with better performance on visual memory, visuospatial, and executive tasks. The work contrasts with prior clinic-based samples by emphasizing early, asymptomatic changes detectable in a more representative group, with no comparable links observed for tau markers.

Core claim

In a community-based cohort of 968 cognitively normal older adults, CSF amyloid-β exhibits inverted U-shaped associations with functional connectivity in the precuneus network and ventral DMN but not the dorsal DMN. Higher connectivity within these and related networks correlates with superior visual memory, visuospatial, and executive function, while CSF tau shows no significant relationships with connectivity measures.

What carries the argument

Group independent component analysis of resting-state fMRI data to extract 14 large-scale functional networks, paired with CSF amyloid-β 1-42 measurements.

Load-bearing premise

Group independent component analysis isolates the same functional networks consistently across participants without bias from cohort composition or scanner effects.

What would settle it

A replication study in a comparable community cohort that finds only linear associations, absent associations, or associations in the dorsal DMN as well would undermine the network-specific inverted U claim.

read the original abstract

Background: Non-linear alterations in brain network connectivity may represent early neural signatures of Alzheimer's disease (AD) pathology in cognitively normal older adults. Understanding these changes and their cognitive relevance may help clarify early network vulnerability associated with AD pathology. Most prior studies recruited participants from memory clinics, often with subjective memory concerns, limiting generalizability. Methods: We examined 14 large-scale functional brain networks in 968 cognitively normal older adults recruited from the community using resting-state functional MRI, cerebrospinal fluid (CSF) biomarkers (amyloid-$\beta$ 1-42 [A$\beta$], total tau, phosphorylated tau 181), and neuropsychological assessments. Functional networks were identified using group independent component analysis. Results: Inverted U-shaped associations between CSF A$\beta$ and functional connectivity were observed in the precuneus network and ventral default mode network (DMN), but not in the dorsal DMN, indicating network-specific vulnerability to early amyloid pathology. Higher connectivity in A$\beta$-related networks, including dorsal and ventral DMN, precuneus, and posterior salience networks, was associated with better visual memory, visuospatial, and executive performance. No significant relationships were observed between CSF tau and functional connectivity. Conclusions: Using a large, community-based cohort, we demonstrate that non-linear alterations in functional connectivity occur in specific networks even during the asymptomatic phase of AD. Moreover, A$\beta$-related network connectivity is cognitively relevant, highlighting early network vulnerability and its functional consequences in amyloid pathology.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript reports an analysis of non-linear (quadratic) associations between CSF amyloid-β (Aβ) levels and resting-state functional connectivity in 14 large-scale networks derived from group ICA in a community-based sample of 968 cognitively normal older adults. It finds inverted U-shaped relationships specifically in the precuneus network and ventral DMN (but not dorsal DMN), positive associations between connectivity in Aβ-related networks and cognitive performance (visual memory, visuospatial, executive), and no significant CSF tau-connectivity links.

Significance. If the network-specific quadratic findings and their cognitive correlations hold after methodological validation, the work would be a useful contribution by extending preclinical AD network studies to a large, community-recruited cohort rather than clinic samples. The direct regression of measured biomarkers onto ICA connectivity values and the focus on asymptomatic participants are strengths; the absence of tau effects is also informative. However, the central dissociation among networks depends on the stability of the group ICA decomposition.

major comments (2)
  1. [Methods (network identification)] Methods section on network identification: The headline dissociation (inverted U in precuneus/ventral DMN but not dorsal DMN) is load-bearing for the claim of network-specific vulnerability. In a heterogeneous community cohort (n=968) that includes varying subclinical vascular risk, head motion, and scanner effects, a single group ICA without reported split-half stability, dual-regression reliability metrics, or subject-specific ICA comparisons risks mixing or misaligning components; this could artifactually produce the observed ventral/precuneus versus dorsal pattern rather than reflect true differential vulnerability.
  2. [Results] Results: The abstract states statistically significant quadratic terms and cognitive associations, yet the manuscript provides no error bars, exact post-correction p-values, effect sizes, or sensitivity analyses (e.g., outlier removal, model assumption checks, or alternative linear/quadratic model comparisons) for the quadratic coefficients in the precuneus and ventral DMN models. These details are required to substantiate the inverted-U claim as the primary novel result.
minor comments (2)
  1. [Abstract] Abstract and throughout: Notation for amyloid-β is inconsistent (Aβ vs. A$β$); standardize to improve readability.
  2. [Methods] The manuscript would benefit from a supplementary table listing all 14 networks with their peak coordinates or spatial overlap metrics to allow readers to evaluate the ICA decomposition.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which help improve the rigor and transparency of our work. We address each major point below and have revised the manuscript to incorporate additional methodological validation and statistical reporting.

read point-by-point responses

  1. Referee: Methods section on network identification: The headline dissociation (inverted U in precuneus/ventral DMN but not dorsal DMN) is load-bearing for the claim of network-specific vulnerability. In a heterogeneous community cohort (n=968) that includes varying subclinical vascular risk, head motion, and scanner effects, a single group ICA without reported split-half stability, dual-regression reliability metrics, or subject-specific ICA comparisons risks mixing or misaligning components; this could artifactually produce the observed ventral/precuneus versus dorsal pattern rather than reflect true differential vulnerability.

    Authors: We appreciate the referee's emphasis on ICA stability, particularly in a community-based sample with potential confounds. Group ICA followed by dual regression is a standard and widely validated approach for large cohorts to derive consistent network maps across participants. To directly address this concern, the revised manuscript now includes a split-half stability analysis (randomly dividing the sample into two subsets of ~484 participants each, recomputing group ICA, and quantifying component similarity via spatial correlation coefficients). We also report dual-regression reliability metrics (e.g., test-retest spatial correlations) specifically for the precuneus, ventral DMN, and dorsal DMN components. These metrics confirm high stability (>0.85 spatial correlation) for the key networks, supporting that the observed dissociation is not artifactual. We have clarified in the methods that dual regression provides subject-specific connectivity estimates while preserving the group-level network definitions, and we have added discussion of motion, vascular risk, and scanner effects as potential limitations. revision: yes

  2. Referee: Results: The abstract states statistically significant quadratic terms and cognitive associations, yet the manuscript provides no error bars, exact post-correction p-values, effect sizes, or sensitivity analyses (e.g., outlier removal, model assumption checks, or alternative linear/quadratic model comparisons) for the quadratic coefficients in the precuneus and ventral DMN models. These details are required to substantiate the inverted-U claim as the primary novel result.

    Authors: We agree that detailed statistical transparency is essential for substantiating the quadratic findings. In the revised manuscript, we have added error bars (representing standard error) to all figures showing the quadratic associations. We now report exact post-correction p-values for the quadratic terms, along with effect sizes (standardized regression coefficients and partial eta-squared). Additionally, we have included comprehensive sensitivity analyses: (1) outlier detection and removal using Cook's distance with re-estimation of models; (2) formal checks of regression assumptions (residual normality via Shapiro-Wilk, homoscedasticity via Breusch-Pagan test); and (3) direct model comparisons between linear and quadratic fits using likelihood ratio tests and AIC/BIC values, confirming superior fit for the quadratic model in the precuneus and ventral DMN. These results are now presented in the main text and supplementary materials. revision: yes

Circularity Check

0 steps flagged

No significant circularity; results are direct empirical regressions on measured data.

full rationale

The paper applies standard group ICA to identify 14 functional networks from rs-fMRI data in a community cohort, then fits regression models (including quadratic terms) to test associations between CSF Aβ levels and network connectivity values, plus links to cognitive scores. These steps consist of preprocessing, decomposition, and statistical hypothesis testing on observed variables without any reduction of a 'prediction' to a fitted parameter by construction, without load-bearing self-citations that define the target result, and without ansatzes or uniqueness claims imported from prior author work. The central claims rest on falsifiable regressions against external biomarkers and neuropsychological data rather than self-referential definitions.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard statistical assumptions for quadratic regression and ICA decomposition plus the domain premise that CSF Aβ42 levels serve as a valid proxy for brain amyloid burden in asymptomatic adults.

free parameters (2)
  • quadratic coefficient for CSF Aβ in precuneus network model
    Fitted parameter capturing the downward turn of the inverted-U relationship.
  • quadratic coefficient for CSF Aβ in ventral DMN model
    Fitted parameter capturing the downward turn of the inverted-U relationship.
axioms (2)
  • domain assumption CSF Aβ42 concentration accurately indexes cerebral amyloid deposition in cognitively normal older adults
    Invoked to interpret the observed connectivity changes as early signatures of AD pathology.
  • domain assumption Group ICA yields stable, biologically meaningful network maps that are comparable across participants
    Required for treating the extracted connectivity values as valid network measures.

pith-pipeline@v0.9.0 · 5847 in / 1382 out tokens · 43938 ms · 2026-05-18T07:16:13.761522+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

  • IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    Inverted U-shaped associations between CSF Aβ and functional connectivity were observed in the precuneus network and ventral default mode network (DMN), but not in the dorsal DMN... using multiple regression including quadratic terms for CSF biomarkers

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

  1. [1]

    Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, et al. Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimer's & Dementia. 2011;7(3):280–92

    work page 2011

  2. [2]

    NIA-AA Research Framework: Toward a biological definition of Alzheimer's disease

    Jack CR, Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer's disease. Alzheimer's & Dementia. 2018;14(4):535–62

    work page 2018

  3. [3]

    What is normal in normal aging? Effects of aging, amyloid and Alzheimer's disease on the cerebral cortex and the hippocampus

    Fjell AM, McEvoy L, Holland D, Dale AM, Walhovd KB. What is normal in normal aging? Effects of aging, amyloid and Alzheimer's disease on the cerebral cortex and the hippocampus. Progress in Neurobiology. 2014;117:20–40

    work page 2014

  4. [4]

    Preclinical Alzheimer's disease: Definition, natural history, and diagnostic criteria

    Dubois B, Hampel H, Feldman HH, Scheltens P, Aisen P, Andrieu S, et al. Preclinical Alzheimer's disease: Definition, natural history, and diagnostic criteria. Alzheimer's & Dementia. 2016;12(3):292–323

    work page 2016

  5. [5]

    Secondary prevention of Alzheimer’s dementia: neuroimaging contributions

    ten Kate M, Ingala S, Schwarz AJ, Fox NC, Chételat G, van Berckel BNM, et al. Secondary prevention of Alzheimer’s dementia: neuroimaging contributions. Alzheimer's Research & Therapy. 2018;10(1):112

    work page 2018

  6. [6]

    Resting State Functional Connectivity in Preclinical Alzheimer’s Disease

    Sheline YI, Raichle ME. Resting State Functional Connectivity in Preclinical Alzheimer’s Disease. Biological Psychiatry. 2013;74(5):340–7. Wu et al. 13

    work page 2013

  7. [7]

    Amyloid- β deposition in mild cognitive impairment is associated with increased hippocampal activity, atrophy and clinical progression

    Huijbers W, Mormino EC, Schultz AP, Wigman S, Ward AM, Larvie M, et al. Amyloid- β deposition in mild cognitive impairment is associated with increased hippocampal activity, atrophy and clinical progression. Brain. 2015;138(4):1023–35

    work page 2015

  8. [8]

    Cascading network failure across the Alzheimer’s disease spectrum

    Jones DT, Knopman DS, Gunter JL, Graff-Radford J, Vemuri P, Boeve BF, et al. Cascading network failure across the Alzheimer’s disease spectrum. Brain. 2015;139(2):547–62

    work page 2015

  9. [9]

    Where is hippocampal activity in the cascade of Alzheimer’s disease biomarkers? Brain

    Aizenstein HJ, Klunk WE. Where is hippocampal activity in the cascade of Alzheimer’s disease biomarkers? Brain. 2015;138(4):831–3

    work page 2015

  10. [10]

    Phases of Hyperconnectivity and Hypoconnectivity in the Default Mode and Salience Networks Track with Amyloid and Tau in Clinically Normal Individuals

    Schultz AP, Chhatwal JP, Hedden T, Mormino EC, Hanseeuw BJ, Sepulcre J, et al. Phases of Hyperconnectivity and Hypoconnectivity in the Default Mode and Salience Networks Track with Amyloid and Tau in Clinically Normal Individuals. The Journal of Neuroscience. 2017;37(16):4323–31

    work page 2017

  11. [11]

    Imaging the evolution and pathophysiology of Alzheimer disease

    Jagust W. Imaging the evolution and pathophysiology of Alzheimer disease. Nature reviews Neuroscience. 2018;19(11):687–700

    work page 2018

  12. [12]

    Functional network segregation is associated with attenuated tau spreading in Alzheimer's disease

    Steward A, Biel D, Brendel M, Dewenter A, Roemer S, Rubinski A, et al. Functional network segregation is associated with attenuated tau spreading in Alzheimer's disease. Alzheimer's & Dementia. 2023;19(5):2034–46

    work page 2023

  13. [13]

    Amyloid-β and tau pathologies relate to distinctive brain dysconnectomics in preclinical autosomal-dominant Alzheimer’s disease

    Guzmán-Vélez E, Diez I, Schoemaker D, Pardilla-Delgado E, Vila-Castelar C, Fox- Fuller JT, et al. Amyloid-β and tau pathologies relate to distinctive brain dysconnectomics in preclinical autosomal-dominant Alzheimer’s disease. Proceedings of the National Academy of Sciences. 2022;119(15):e2113641119

    work page 2022

  14. [14]

    Synergistic association of Aβ and tau pathology with cortical neurophysiology and Wu et al

    Gallego-Rudolf J, Wiesman AI, Pichet Binette A, Villeneuve S, Baillet S, Group P-AR. Synergistic association of Aβ and tau pathology with cortical neurophysiology and Wu et al. 14 cognitive decline in asymptomatic older adults. Nature Neuroscience. 2024;27(11):2130– 7

    work page 2024

  15. [15]

    Regional amyloid burden and intrinsic connectivity networks in cognitively normal elderly subjects

    Lim HK, Nebes R, Snitz B, Cohen A, Mathis C, Price J, et al. Regional amyloid burden and intrinsic connectivity networks in cognitively normal elderly subjects. Brain. 2014;137(12):3327–38

    work page 2014

  16. [16]

    Effects of Beta-Amyloid on Resting State Functional Connectivity Within and Between Networks Reflect Known Patterns of Regional Vulnerability

    Elman JA, Madison CM, Baker SL, Vogel JW, Marks SM, Crowley S, et al. Effects of Beta-Amyloid on Resting State Functional Connectivity Within and Between Networks Reflect Known Patterns of Regional Vulnerability. Cerebral Cortex. 2016;26(2):695–707

    work page 2016

  17. [17]

    β-Amyloid affects frontal and posterior brain networks in normal aging

    Oh H, Mormino EC, Madison C, Hayenga A, Smiljic A, Jagust WJ. β-Amyloid affects frontal and posterior brain networks in normal aging. NeuroImage. 2011;54(3):1887–95

    work page 2011

  18. [18]

    Amyloid Plaques Disrupt Resting State Default Mode Network Connectivity in Cognitively Normal Elderly

    Sheline YI, Raichle ME, Snyder AZ, Morris JC, Head D, Wang S, et al. Amyloid Plaques Disrupt Resting State Default Mode Network Connectivity in Cognitively Normal Elderly. Biological Psychiatry. 2010;67(6):584–7

    work page 2010

  19. [19]

    Disruption of Functional Connectivity in Clinically Normal Older Adults Harboring Amyloid Burden

    Trey H, Koene RAVD, Becker JA, Angel M, Reisa AS, Keith AJ, et al. Disruption of Functional Connectivity in Clinically Normal Older Adults Harboring Amyloid Burden. The Journal of Neuroscience. 2009;29(40):12686

    work page 2009

  20. [20]

    Onami S, Greicius MD, Rabinovici GD, et al

    Mormino EC, Smiljic A, Hayenga AO, H. Onami S, Greicius MD, Rabinovici GD, et al. Relationships between Beta-Amyloid and Functional Connectivity in Different Components of the Default Mode Network in Aging. Cerebral Cortex. 2011;21(10):2399–407

    work page 2011

  21. [21]

    Rationale and Design of the Emory Healthy Aging and Emory Healthy Brain Studies

    Goetz ME, Hanfelt JJ, John SE, Bergquist SH, Loring DW, Quyyumi A, et al. Rationale and Design of the Emory Healthy Aging and Emory Healthy Brain Studies. Neuroepidemiology. 2019;53(3-4):187–200. Wu et al. 15

    work page 2019

  22. [22]

    Semiblind spatial ICA of fMRI using spatial constraints

    Lin Q-H, Liu J, Zheng Y-R, Liang H, Calhoun VD. Semiblind spatial ICA of fMRI using spatial constraints. Human Brain Mapping. 2010;31(7):1076–88

    work page 2010

  23. [23]

    Decoding Subject-Driven Cognitive States with Whole-Brain Connectivity Patterns

    Shirer WR, Ryali S, Rykhlevskaia E, Menon V, Greicius MD. Decoding Subject-Driven Cognitive States with Whole-Brain Connectivity Patterns. Cerebral Cortex. 2012;22(1):158–65

    work page 2012

  24. [24]

    Group ICA for identifying biomarkers in schizophrenia: ‘Adaptive’ networks via spatially constrained ICA show more sensitivity to group differences than spatio-temporal regression

    Salman MS, Du Y, Lin D, Fu Z, Fedorov A, Damaraju E, et al. Group ICA for identifying biomarkers in schizophrenia: ‘Adaptive’ networks via spatially constrained ICA show more sensitivity to group differences than spatio-temporal regression. NeuroImage: Clinical. 2019;22:101747

    work page 2019

  25. [25]

    Technical performance of a novel, fully automated electrochemiluminescence immunoassay for the quantitation of β-amyloid (1–42) in human cerebrospinal fluid

    Bittner T, Zetterberg H, Teunissen CE, Ostlund Jr RE, Militello M, Andreasson U, et al. Technical performance of a novel, fully automated electrochemiluminescence immunoassay for the quantitation of β-amyloid (1–42) in human cerebrospinal fluid. Alzheimer's & Dementia. 2016;12(5):517–26

    work page 2016

  26. [26]

    L'examen clinique en psychologie

    Rey A. L'examen clinique en psychologie. Paris: Presses Universitaries De France; 1958. 222– p

    work page 1958

  27. [27]

    Contributions to neuropsychological assessment: A clinical manual

    Benton AL, Abigail B, Sivan AB, Hamsher Kd, Varney NR, Spreen O. Contributions to neuropsychological assessment: A clinical manual. 2nd ed. New York: Oxford University Press; 1994

    work page 1994

  28. [28]

    Initial letter and semantic category fluency in Alzheimer's disease, Huntington's disease, and progressive supranuclear palsy

    Rosser A, Hodges JR. Initial letter and semantic category fluency in Alzheimer's disease, Huntington's disease, and progressive supranuclear palsy. Journal of Neurology, Neurosurgery & Psychiatry. 1994;57(11):1389–94

    work page 1994

  29. [29]

    Manual of directions and scoring

    Battery AIT. Manual of directions and scoring. Washington, DC: War Department, Adjutant General’s Office; 1944. Wu et al. 16

    work page 1944

  30. [30]

    The human connectome in Alzheimer disease — relationship to biomarkers and genetics

    Yu M, Sporns O, Saykin AJ. The human connectome in Alzheimer disease — relationship to biomarkers and genetics. Nature Reviews Neurology. 2021;17(9):545–63

    work page 2021

  31. [31]

    Primary Disruption of the Memory-Related Subsystems of the Default Mode Network in Alzheimer’s Disease: Resting-State Functional Connectivity MRI Study

    Qi H, Liu H, Hu H, He H, Zhao X. Primary Disruption of the Memory-Related Subsystems of the Default Mode Network in Alzheimer’s Disease: Resting-State Functional Connectivity MRI Study. Frontiers in Aging Neuroscience. 2018;Volume 10 - 2018

    work page 2018

  32. [32]

    Association of Phosphorylated Tau Biomarkers With Amyloid Positron Emission Tomography vs Tau Positron Emission Tomography

    Therriault J, Vermeiren M, Servaes S, Tissot C, Ashton NJ, Benedet AL, et al. Association of Phosphorylated Tau Biomarkers With Amyloid Positron Emission Tomography vs Tau Positron Emission Tomography. JAMA Neurology. 2023;80(2):188–99. Wu et al. 17 Figures/Tables Fig. 1. Averaged connectivity maps across all healthy older adults for the auditory network,...

    work page 2023