Influence of educational level in aphasia testing: experiences from standardizing the Portuguese version of the AAT (Aachener Aphasie Test)

Martin Lauterbach (mlauterbach@fm.ul.pt)

Language Research Laboratory

Neurological Clinical Investigation Unit

Faculty of Medicine

Lisbon, Portugal

The stratification of normative data for demographic variables is frequently used in order to obtain reliable results in neuropsychological testing. In Portugal with its heterogeneous population in terms of education the normative data of the Portuguese version of the AAT for healthy controls (N=156: male=58, female=98) showed a pronounced effect of education. The Token Test turned out to be the subtest that is less influenced by educational level. Contrarily in the Naming subtest less educated subjects made a significantly more errors. When the stimulus material consisted of photographs instead of line-drawings the number of errors decreased significantly. The qualitative analysis of the naming errors turned out that 40% of the errors were perceptive ones; thus they were caused by a failure outside the linguistic system. Using the same stimuli in an auditory comprehension task the same number of errors as in the naming task was observed; the errors were mostly related to semantic-perceptive distracters.

There is no ceiling effect in the performance of healthy subjects due to educational differences. Educational level influences not only linguistic but also perceptive capacities, or the mediation between these two stages of processing. The premorbid knowledge and capacities of the aphasic subject seem to exert a strong influence on the test outcome and have to be taken in account. The stratification of AAT scores for educational level might be useful, not only in the Portuguese population.

Key words: aphasia testing, normative data, educational effect