Abstract : Subjects with mild cognitive impairment (MCI) have disturbances in their spatial navigation abilities and exhibit early deficits in visuospatial short-term memory. The purpose of the present study was to determine whether a quantitative (span score) and qualitative (evaluating navigation strategies used) analysis of the Corsi test (usual condition and complex navigation task) would be useful to reveal cognitive decline. We evaluated the performance of 15 young adults, 21 healthy elderly subjects and 15 subjects with MCI using the electronic version of the Corsi test (the Modified Corsi Block-Tapping Test, MCBT) and the complex navigation task (the Modified Walking Corsi Test, MWCT). The MWCT, which is an adaptation of the Corsi test, assesses spatial memory when the subject walks in a complex environment. We used Richard et al.'s model [Cogn Sci 1993;17:497-529] to investigate problem-solving strategies during the Corsi tests. The span scores obtained on the MCBT and the MWCT were significantly lower in the healthy elderly subjects (MCBT = 5.0 ± 0.7; MWCT = 4.0 ± 0.7) and the subjects with MCI (MCBT = 4.7 ± 0.8; MWCT = 4.1 ± 0.9) than in the younger adults (MCBT = 6.2 ± 0.6; MWCT = 5.3 ± 1.0). The visuospatial working memory was more impaired in the complex navigation task (MWCT = 4.3 ± 0.9) than in the modified Corsi test (MCBT = 5.3 ± 0.8). Finally, the subjects with greater cognitive impairment were more likely to have inadequate or absence of problem-solving strategies. Investigating the problem-solving strategies used during the MWCT appears to be a promising way to differentiate between the subjects with MCI and the healthy elderly subjects.
