We investigated the effects of aging on lexical diversity (LD) and lexical informativeness (LI) in 251 adults aged 25 to 89. Multiple LD measures were analyzed, including traditional (Number of Different Words, NDW; and Type-Token Ratio, TTR), length-independent (Moving-Average Type-Token Ratio, MATTR; Measure of Textual Lexical Diversity, MTLD; Hypergeometric Distribution Diversity, HD-D), and entropy-based (Word Information Measure, WIM) metrics. An age-related decline was found in LD, although not all metrics captured this trend equally. WIM was the most sensitive indicator. MATTR and MTLD showed smaller or inconsistent age effects. Regarding LI, age and education were significant predictors. WIM was the strongest predictor of informativeness, followed by NDW. WIM is a robust and sensitive metric for capturing age-related discourse changes.
