RESUMO
BACKGROUND: The purposes of this study were (1) to investigate the abilities of word definition in patients with Alzheimer's disease (AD) according to the severity, and (2) to examine the error patterns in patients with Alzheimer's disease. METHODS: Eight individuals with MCI (CDR=0.5) and 16 patients with AD (eight for probable AD mild group of CDR=1 and eight for probable AD moderate group of CDR=2) participated in the study. Eight normal age-, gender-, and education-matched elderly adults served as a control group for the MCI and AD groups. As stimuli for the word definition, eleven semantic categories were used, and two concrete words were selected from each category, resulting in a total of 22 items. Prior to the task, four definition categories were provided: 1) functional, 2) relational, 3) perceptual, and 4) categorical. Statistical analyses were performed using Kruskal-Wallis test, and Bonferroni analyses were used as a post-hoc comparison for any significant results. RESULTS: There were significant differences in word definition scores among four groups. The probable AD moderate group showed the lower definition score than the probable AD mild group. And the probable AD moderate group showed the lower definition score than MCI group. Each group defined words in different ways. While the control group employed four definition different categories equally, the probable AD moderate group used a functional definition category mainly. However, relational and categorical definition categories were rarely observed in the probable AD moderate group. The analysis of error pattern showed that inadequate definition was frequently observed in all groups. CONCLUSIONS: The results from this study suggest that word definition task could be a sensitive indicator of the impairment of semantic knowledge in patients with AD.