Abstract
Verbal memory decline is an early marker of dementia. This study aims to identify significant lifestyle correlates of verbal learning and memory among older adults to inform intervention targets for maintaining cognitive health in late life. The sample included 221 community-dwelling Chinese older adults (60–80 years). Verbal learning and memory were assessed using the Hong Kong List Learning Test (HKLLT). Lifestyle factors, emotion, demographics, and medical history were assessed using questionnaires. Body mass index (BMI) and waist-to-hip ratio (WHR) were measured. Pairwise associations were examined using Spearman correlation and polynomial regression to screen linear and non-linear associations, followed by generalized additive models (GAMs) to examine partial effect of each lifestyle factor while controlling for others factors. At the pairwise level, poorer sleep quality, less moderate-to-vigorous housework, and several dietary structure indices (e.g., grain-to-vegetables and fruits ratio, plant protein-to-total protein ratio, and pungent food intake frequency) were significantly associated with poorer verbal learning and memory. In fully adjusted GAMs, poorer sleep quality remained significantly associated with poorer 10-min and 30-min delayed recall, while more moderate-to-vigorous housework remained significantly associated with better Recognition. The plant protein-to-total protein ratio retained a significant nonlinear (inverted U-shaped) association with Recognition, with performance peaking at a moderate proportion of plant protein (approximately 0.35–0.40). Higher pungent food intake frequency was associated with poorer 30-min delayed recall. None of the lifestyle factors remained significantly associated with total learning in GAMs sleep quality, moderate-to-vigorous housework engagement, and dietary structure, particularly a moderate plant protein-to-total protein ratio and reduced pungent food intake, emerged as the most robust lifestyle correlates of verbal memory, though effect sizes were modest and largely specific to memory rather than learning. These modifiable targets may guide lifestyle interventions to support memory maintenance in late life.