Covariates of production and perseveration on tests of phonemic, semantic and alternating fluency in normal aging.

Standard measures of phonemic and semantic fluency, as well as fluency measures that impose additional demands upon set-switching were administered to younger and older adults. No age effect was found for total output on semantic fluency, while older participants generated significantly more responses relative to their younger counterparts on the phonemic measure. However, older adults made significantly more perseverative errors on both these measures. For tests of alternating fluency, only where participants were required to switch between two semantic dimensions, was an age deficit observed. Although the correlation between phonemic repetitions and age remained significant when performance on the Digit Symbol Test or a measure of fluid intelligence was partialled out, age effects for semantic repetitions and semantic alternating fluency were substantially reduced. The age benefit for number of correct responses upon phonemic fluency was significantly attenuated when a measure of crystallized intelligence was entered as a control variable. The theoretical and practical implications of these results are discussed.