[Cognitive capacity in advanced age: initial results of the Berlin Aging Study]

Z Psychol Z Angew Psychol. 1995;203(4):283-317.
[Article in German]

Abstract

This study reports data on intellectual functioning in old and very old age from the Berlin Aging Study (N = 516; age range = 70-103 years; mean age = 85 years). A psychometric battery of 14 tests was used to assess five cognitive abilities: reasoning, memory, and perceptual speed from the broad fluid-mechanical as well as knowledge and fluency from the broad crystallized-pragmatic domains. Cognitive abilities had a negative linear relationship with age, with more pronounced age-based reductions in fluid-mechanical than crystallized-pragmatic abilities. At the same time, ability intercorrelations formed a highly positive manifold, and did not follow the fluid-crystallized distinction. Interindividual variability was of about equal magnitude across the entire age range studied. There was, however, no evidence for substantial sex differences. As to origins of individual differences, indicators of sensory and sensorimotor functioning were more powerful predictors of intellectual functioning than cultural-biographical variables, and the two sets of predictors were, consistent with theoretical expectations, differentially related to measures of fluid-mechanical (perceptual speed) and crystallized pragmatic (knowledge) functioning. Results, in general indicative of sizeable and general losses with age, are consistent with the view that aging-induced biological influences are a prominent source of individual differences in intellectual functioning in old and very old age. Longitudinal follow-ups are underway to examine the role of cohort effects, selective mortality, and interindividual differences in change trajectories.

Publication types

  • English Abstract

MeSH terms

  • Aged
  • Aged, 80 and over
  • Aging / psychology*
  • Berlin
  • Cognition*
  • Female
  • Humans
  • Individuality
  • Intelligence
  • Male
  • Neuropsychological Tests / statistics & numerical data*
  • Reference Values