A computerized version of the Short Form of the Face-Name Associative Memory Exam (FACEmemory (R)) for the early detection of Alzheimer's disease

Background Computerized neuropsychological tests for early detection of Alzheimer's disease (AD) have attracted increasing interest. Memory for faces and proper names is a complex task because its association is arbitrary. It implicates associative occipito-temporal cerebral regions, which are disrupted in AD. The short form of the Face-Name Associative Memory Exam (FNAME-12), developed to detect preclinical and prodromal AD, asks individuals to learn the names and occupations associated with 12 faces. The current work advances this field by using voice recognition and touchscreen response format. The purpose of this study is to create the first self-administered episodic memory test, FACEmemory (R), by adapting the FNAME-12 for tablet use with voice recognition, touchscreen answers, and automatic scoring. The test was minimally supervised by a psychologist to avoid technological problems during execution and scored manually to assess the reliability of the automatic scoring. The aims of the present study were (1) to determine whether FACEmemory (R) is a sensitive tool for the detection of cognitive impairment, (2) to examine whether performances on FACEmemory (R) are correlated with those on the S-FNAME (paper-and-pencil version with 16 images), and (3) to determine whether performances on FACEmemory (R) are related to AD biomarkers in the cerebrospinal fluid (CSF) (A beta 42, p-tau, and A beta 42/p-tau ratio). Methods FACEmemory (R) was completed by 154 cognitively healthy (CH) individuals and 122 subjects with mild cognitive impairment, of whom 61 were non-amnestic (naMCI) and 61 amnestic (aMCI). A subsample of 65 individuals completed the S-FNAME, and 65 subjects received lumbar punctures. Results Performance on FACEmemory (R) was progressively worse from CH to the naMCI and aMCI groups. A cutoff of 31.5 in total FACEmemory (R) obtained 80.5% and 80.3% sensitivity and specificity values, respectively, for discriminating between CH and aMCI. Automatically corrected FACEmemory (R) scores were highly correlated with the manually corrected ones. FACEmemory (R) scores and AD CSF biomarker levels were significantly correlated as well, mainly in the aMCI group. Conclusions FACEmemory (R) may be a promising memory prescreening tool for detecting subtle memory deficits related to AD. Our findings suggest FACEmemory (R) performance provides a useful gradation of impairment from normal aging to aMCI, and it is related to CSF AD biomarkers.