Noradrenergic gain failure and cognitive-state instability in aging and Alzheimer's disease

Abstract

Aging and Alzheimer's disease are usually characterized by average cognitive decline, yet early impairment may also appear as fluctuating attention, variable memory access, inconsistent reaction time, and state-dependent performance. This hypothesis-driven narrative review integrates adaptive gain theory, locus coeruleus–norepinephrine (LC–NE) anatomy and pathology, LC-sensitive MRI, pupillometry, EEG/fMRI state dynamics, sleep research, behavioral variability, and Alzheimer's disease biomarkers to examine how disease-related vulnerability becomes expressed as unreliable cognition. We formulate a reduced control-layer model in which LC–NE dysfunction may weaken the stability of task-relevant neural states. Three candidate latent operating patterns are considered: hypogain, hypertonic noise, and gain mistiming. These patterns occupy continuous and potentially overlapping regions rather than discrete biological subtypes. Individual readouts are nonspecific and correlated; inference therefore requires temporally aligned convergence across dynamic arousal measures, neural-state consequences, behavioral instability, repeated observations, and prespecified alternative explanations. A supplementary forward simulation maps imposed differences in gain amplitude, tonic volatility, and timing onto model outputs but does not estimate latent modes from patient data. The framework predicts that LC-related measures will add explanatory value for cognitive variability and task-state persistence beyond mean performance and Alzheimer's disease biomarker burden. Whether the candidate patterns can be recovered reliably and provide incremental explanatory value remains an empirical question.