Deterioration of controlled processes in the preclinical phase of dementia: a confirmatory analysis

Dement Geriatr Cogn Disord. 2000 Jan-Feb;11(1):46-52. doi: 10.1159/000017213.

Abstract

Preclinical deficits in dementia have recently been investigated in the PAQUID cohort by means of a principal component analysis (PCA) performed from various test scores. According to the interpretation of the first PCA factor, which was found to be an independent predictor of dementia 2 years later, the authors suggested that cognitive deficits occurring in the preclinical phase of dementia may reflect the deterioration of controlled processes. Because this hypothesis relies essentially on the interpretation of PCA factors, the present study was designed to test it within the same cohort. The analyses were focused on one of the tests administered to the subjects, the Wechsler Paired Associate Test (WPAT) divided into two subscores in order to provide a measure of controlled processes of memory functioning (i.e. retrieval of unrelated pairs) as well as of more automatic processes (i.e. retrieval of related pairs). The results showed that the future demented subjects, although significantly impaired compared to elderly controls, were able to increase their performance across trials. More importantly, a new PCA including the subscores of the WPAT showed that the unrelated pair scores were strongly correlated to factor 1, which remained the only factor associated with subsequent dementia, whereas the related pair scores were rather correlated to factor 2. Therefore, these results confirm our previous PCA interpretation, lending more weight to our hypothesis of a preclinical impairment of controlled processes in dementia.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Dementia / diagnosis*
  • Dementia / psychology
  • Female
  • Humans
  • Male
  • Paired-Associate Learning*
  • Predictive Value of Tests
  • Psychometrics
  • Reference Values
  • Wechsler Scales / statistics & numerical data*