Network hierarchy in Alzheimer's and Lewy body spectrum disorders: functional gradients, cognitive phenotypes, and translational gaps.

Dementia syndromes alter cortical hierarchy in addition to individual brain regions. Functional connectivity gradients offer a continuous, low-dimensional framework for examining these hierarchy-level changes. This focused structured narrative review combines a transparent evidence audit with a candidate-target comparative framework for evaluating gradient evidence in Alzheimer's disease (AD) and Lewy body spectrum disorders, including dementia with Lewy bodies and Parkinson's disease dementia. The evidence is asymmetric. AD has comparatively direct evidence for metric-specific change at the default-mode/transmodal end of the principal cortical gradient, whereas Lewy body disease has sparse and partly negative direct functional-gradient evidence. The review defines evidence tiers and eligibility rules for translating conventional functional-connectivity findings into testable gradient hypotheses, and applies them to the Lewy body literature. This framework is intended to improve evidence appraisal when emerging neuroimaging methods are applied to heterogeneous disease populations. Current data support metric-specific principal-gradient change and transmodal dedifferentiation in AD. In Lewy body disease, posterior cortical, attention, salience, and control-network findings remain FC-derived candidate targets for direct testing. The review also addresses gradient-estimation methods, disease-specific confounds, mixed pathology, brain-behavior associations, and validation standards. Gradient measures are not proposed as ready clinical biomarkers. Instead, the review specifies the evidence required to determine whether hierarchy-axis phenotyping adds value beyond conventional connectivity, atrophy, and molecular biomarkers.