Title: Why Does Similar Alzheimer's Pathology Lead to Different Outcomes? A Multi-Modal Framework for Cognitive Resilience

Cognitive resilience to Alzheimer's disease, understood as the preservation of cognitive ability despite accumulating amyloid and tau pathology, is one of the most clinically consequential yet mechanistically underexplored phenomena in aging neuroscience. Amyloid and tau burden account for under 40% of the variance in cognitive outcomes. Synaptic integrity, myelination, and neuroinflammatory factors act as biological buffers of pathology-driven cognitive dysfunction, but buffer capacity arises from their joint regional configuration and interaction rather than any single marker. Measuring that configuration currently requires postmortem tissue. Cortex is organized into columns and layers whose distinct biological substrates map onto dissociable morphological measures that volumetry conflates: the surface area of the gray matter-white matter boundary indexes column number, and cortical thickness indexes laminar microcircuitry. Morphology therefore compresses this variation into a phenotype-aware, low-dimensional representation obtainable in living individuals. We test the hypothesis that morphological features bridge molecular composition and cognitive resilience through biological buffer capacity. The Seattle Alzheimer's Disease Brain Cell Atlas provides repeated lifetime cognitive assessment, quantitative neuropathology, and spatial transcriptomics in the same donors; the Adult Changes in Thought study contributes postmortem and antemortem structural MRI in overlapping subsets. Resilience is measured as better-than-expected cognition given each donor's position on a continuous index of neuropathologic burden. This work first asks whether cortical surface area and thickness predict resilience, then whether molecular tissue composition predicts the morphology that does so. Confirming that morphological features index biological buffer capacity would make brain morphometry a tractable proxy for molecular tissue state in living individuals, with direct implications for clinical biomarker development and mechanistic Alzheimer’s disease research.