DOI: 10.1111/tops.12729 ISSN: 1756-8757

Discourse Production Across the Adult Lifespan: Microlinguistic Processes

Hana Kim, Stephen Kintz
  • Artificial Intelligence
  • Cognitive Neuroscience
  • Human-Computer Interaction
  • Linguistics and Language
  • Experimental and Cognitive Psychology


Successful spoken discourse requires a speaker to be informative to deliver a coherent, meaningful message. The informativeness of discourse can be conveyed by the variety of vocabulary produced (i.e., lexical diversity [LD]), the typicality of vocabulary items used (i.e., core lexicon [CL]), and the amount of relevant content produced (i.e., information units). Yet, it is well documented that older adults produce less informative content compared to younger adults despite relatively subtle changes to LD. The typicality of core lexical items has not been assessed in healthy aging. Paradoxically, these results indicate that some aspects of discourse informativeness remain stable or even improve across the adult lifespan, while other aspects decline. The purpose of the current study is to understand how microlinguistic processes of informativeness change across the adult lifespan. The cross‐sectional study included narrative language samples from two wordless picture books collected from 420 healthy participants between 20 and 89 years old. LD and percent of correct information units (%CIUs) were analyzed, as well as CL nouns and verbs. The results indicate that %CIUs and CL nouns demonstrate a quadratic decline starting around the ages of 40 and 60, respectively. LD shows a slight linear decline as a function of age. CL verbs are resistant to age‐related changes but are influenced greater by education. The differing findings across the microlinguistic measures can be explained by the weakened connections within the language system and the differential characteristics of the measures. The findings contribute to the aging literature by systematically identifying the trajectory of how variables of informativeness change with age.

More from our Archive