Abstract
Old age is an extended period of life characterized by major inter-individual differences and different life trajectories which have been synthesized in three modalities of ageing: usual, pathological and successful (Rowe and Khan, 1997). Analysis of the variables associated with these patterns and with cognitive evolution in old age has generated a large number of studies, many of which involve the analysis of learning potential or cognitive plasticity as an assessment tool which enables us to establish differences between old adults with greater precision than static assessments (Schreiber and Schneider, 2006). The present study analysed the performance of an initial sample of 312 old adults with different cognitive profiles in the AVLT-PA learning potential assessment task, with the objective of determining which components of the task may be used to classify participants most correctly, and which may be used to predict the long-term cognitive evolution of old adults. Results indicate that the most appropriate marker of the AVLT-PA task for classifying old adults and determining their cognitive evolution is the Delayed Recall measure.