Categorization in Context for Young and Older Adults

Abstract

Individuals make sense of the world by grouping items into categories, or clusters of concepts that share certain characteristics. Some research has indicated that older adults may organize concepts differently than young adults; however, findings have been inconsistent - dependent upon the tasks. Linguistic context influences word meaning. Although common categories (e.g., animals, furniture) are context-independent, exemplars are only activated by certain contextual cues within a message. Common categories are generally well-established in memory; however, it is unclear whether older adults use linguistic context as effectively as younger ones. The purpose of the current study was to investigate the effect of linguistic context on category structure in young and typical older adults. All participants passed hearing, reading, and category screening tests. They were administered the Peabody Picture Vocabulary Test - IV (PPVT-IV), yielding no significant differences between the groups on this measure as well as educational level. In a timed computer-based contextual categorization task, participants (20 young, 20 older) were provided with 150 stimulus sentences containing a superordinate category label. Using the context of the sentence, the participants were required to make a semantic decision relative to determining if a specific exemplar was the best example of the target category concept in the sentence by answering `Yes' or `No'. There were six exemplar categories (i.e., true related, true unrelated, false related, false unrelated, out-of-set related, out-of-set unrelated). Accuracy of response and reaction time were determined for each sentence for all participants. Results indicated that young adults were significantly more accurate and responded significantly faster than the older group. Both groups had similar patterns of errors for the six categories. Participant scores on the PPVT-IV correlated with reaction time for both age groups but not with accuracy. Logistic regression indicated that it was possible to predict a participant's accuracy based on age group, category of response, as well as the interaction between the two variables. It appears that categorization is vulnerable to the aging process, which may have further implications for communication effectiveness and cognitive processing.