The aims of this study were (1) to describe and compare the performance of illiterate and low-educated older adults, without evidence of cognitive impairment, on different versions of the Boston Naming Test (BNT) original, Brazilian adapted, abbreviated 30-item (even and odd) and 15-item from the CERAD (Consortium to Establish a Registry for Alzheimer's Disease) battery; (2) to compare performance on the original versus adapted versions of the BNT.

A total of 180 healthy older adults (60 years or older) were stratified according to educational level (0, 1–2, and 3–4 years), and age (60–69, 70–79, and ≥ 80 years). The protocol comprised the following instruments: Mini-Mental State Examination (MMSE), Brief Cognitive Screening Battery (BCSB), Functional Activities Questionnaire (FAQ), Geriatric Depression Scale (GDS), and the BNT.

The illiterate participants had poorer performance than the educated participants. The performance of the two educated groups was similar on all versions of the BNT. A higher number of correct responses were observed on the adapted BNT than on the original BNT in all three education groups.

The adapted BNT appears to be the most suitable for use in the low-educated Brazilian population. The present study provided normative data for low-educated elderly on several different versions of the BNT, which may be helpful in diagnosing naming deficits among elderly in these strata of the population.