The picture-naming task in the analysis of cognitive deterioration in Alzheimer's disease

Background: The picture-naming test has been widely used in determining cognitive deterioration in Alzheimer patients, because it has been found that the number of pictures patients can name, and the types of errors they commit, change as the illness advances. Aims: The objective of this study was to examine the lexical-semantic deterioration of Alzheimer's patients over a period of time by analysing the changes in their errors and the variables that predict these changes. Methods and Procedures: A total of 100 pictures were shown to a group of 10 probable Alzheimer's patients on two separate occasions spaced 2 years apart. The task of the patients was to name the pictures. For each picture, the values of name agreement, imageability, familiarity, frequency, age of acquisition, and length were available. Three types of analyses of the results of the naming test were performed: (a) relationship between stimulus variables and patient success, (b) analysis of errors from the first and second evaluations, and (c) relationship between stimulus variables and change in errors. Outcome and Results: There was a considerable reduction of correct responses from the first to the second evaluation, and the determining variable in both cases was age of acquisition. What changed in this 2-year period were the types of errors produced by the patients, with a higher percentage of semantic errors and circumlocutions in the first evaluation and a predominance of "I don't know" responses in the second. But the most interesting discovery was that the nature of those changes, which resulted from the patients' cognitive deterioration, is determined by the age of acquisition of the stimuli. Conclusions: The combination of these two indices-stimulus variables and types of errors-makes it possible to specify the level of lexical-semantic deterioration of Alzheimer's patients.