Development of the Figure Memory Test and the Word Accentuation Test for the assessment of cognition in Brazil

Abstract Background In low‐ and middle‐income countries, the educational level of the population tends to vary significantly and persons with no literacy are in high numbers. In addition, countries with vast territories, such as Brazil, present significant differences in culture and language. These facts pose challenges for cognitive assessment. To this date, most cognitive tools used in Brazil have been developed in high‐income nations, and later, culturally and linguistically adapted for use in the country. Yet, some instruments, such as the Figure Memory Test (FMT) of the Brief Cognitive Screening Battery and the Brazilian Word Accentuation Test (WAT‐Br) have been developed in Brazil. Method FMT: The FMT was developed to assess episodic memory among individuals with low or no education. Twenty black‐and‐white figures were selected to compose the test because they were easily recognized and named in pilot testing with individuals with low education. Ten figures were selected as target and are named and encoded, and ten others were included as distractors in the recognition phase. In an initial study, 51 cognitively unimpaired individuals, 23 having no education, completed the FMT and other cognitive tests. WAT‐Br: The WAT‐Br was developed to assess pre‐morbid intelligence. An initial pool of 60 items was constructed and a final version of 40 items was derived by item response theory. A sample of 206 older adults underwent the WAT‐Br and a standardized neuropsychological battery. Independent ratings were performed by two observers in 58 random participants. Result FMT: The FMT delayed recall scores did not differ between participants with (7.26 ± 1.56) and without education (6.87 ±1.52) (p = 0.458). No differences in FMT scores were observed when the literate group was divided according to mean years of schooling. WAT‐Br: The items showed moderate to high discrimination (α between 0.93 and 25.04) and spanned a wide range of difficulty (β between ‐2.07 and 1.40). The WAT‐Br was shown to have an excellent internal consistency (Kuder‐Richardson Formula 20 = 0.95) and inter‐rater reliability (intraclass correlation coefficient = 0.92). Conclusion The FMT and the WAT‐Br proved to be valuable tools for estimating cognitive performance in Brazil.