Nonlinear Lifespan Trajectories of Memory and Fluid Reasoning: A Longitudinal GAMM Study

Area of Science:

• Neuroscience
• Cognitive Aging
• Neuroimaging

Background:

• Longitudinal studies are crucial for understanding cognitive aging.
• Flexible nonlinear approaches are needed to model complex lifespan neural changes.
• Previous research often relies on linear or cross-sectional methods, limiting insights into dynamic brain changes.

Purpose of the Study:

• To characterize linear and nonlinear trajectories of brain activation across memory and fluid reasoning domains using longitudinal fMRI data.
• To identify brain regions with age-varying coupling between cognition and brain activation.
• To leverage generalized additive mixed models (GAMMs) for advanced analysis of lifespan neural changes.

Main Methods:

• Utilized three waves of fMRI data from the Reference Ability Neural Network study (N=431-441, ages 20-80+).
• Applied generalized additive mixed models (GAMMs) with tensor-product smooths to model age and time effects on brain activation.
• Analyzed age-varying cognition-brain coupling by linking within-person activation fluctuations to in-task performance.

Main Results:

• Cognitive functions showed age-related declines, but brain activation revealed pronounced nonlinear Age × Time effects, primarily in posterior association cortex.
• The right POS2 (precuneus/cuneus) showed a midlife reduction in activation, suggesting early functional vulnerability.
• Cognition-activation coupling analyses identified age-dependent shifts in coupling strength and sign across posterior midline, dorsal parietal, left rostral Area 6, and IFJp regions.

Conclusions:

• Posterior association cortex is a key area for functional reorganization throughout adulthood.
• Flexible GAMM approaches uncover developmental inflection points and brain-behavior coupling shifts missed by linear or cross-sectional methods.
• This study provides novel insights into the dynamic neural mechanisms underlying cognitive aging.