Fewer Paris, parrots peaches: Interpreting cross-cultural semantic fluency deficits in Alzheimer's assessment

Abstract

Impaired semantic fluency has emerged as a robust neuropsychological indicator of Alzheimer's Disease (AD). Furthermore, semantic fluency deteriorates more rapidly than phonemic fluency and this discrepancy pattern is thought to be especially useful in the early assessment and differential diagnosis of AD. However, studies have implicated a role for culture in semantic task performance suggesting that certain cognitive processes are facilitated by cultural knowledge and the country's state of development. Since 60% of AD patients are estimated to come from developing countries, the implications of these assertions with respect to the interpretation of cognitive performance are crucial to valid AD assessment. The effects of age, education and ethnicity on performance on semantic (cities, animals, fruits) and phonemic (letters P, F, L) verbal fluency were investigated using multiple regression analyses in 124 healthy British and Caribbean adults. Caribbean and British persons had similar scores on phonemic fluency for all three letters (P, F, L). However, the Caribbean group produced significantly fewer words for each semantic category (cities, animals, fruits). Increasing years of education were associated with greater word production in both semantic and phonemic fluency. There was no interaction between ethnicity and education or age. The phonemic-semantic fluency scores in Caribbean's fall within the range of normal performance but the discrepancy between the two scores parallels the pattern observed in impaired AD patients. These results provide further psychometric substantiation for different cognitive processes being involved in phonemic versus semantic fluency. Lower word production in semantic fluency in the Caribbean group may be attributed to differences in societal development and access to education, both of which shape and facilitate abstraction, categorical and taxonomic thinking skills that are required for semantic fluency, but not for phonemic fluency. The findings also suggest that these cognitive skills may not be invariably present but are differentially influenced by culture. These skills are measured in semantic fluency and may be featured in routine neuropsychological assessment of AD. Differences in cognitive performance which are attributable to culture and not AD pathology has critical implications for valid AD assessment both within and across cultures.