Table 3.

Association between cognitive measures and the demographic, clinical and neuroimaging variables.

These correlations were performed in the older group. Values for the global deviation index were available for 23 individuals. Values for visual abilities were available for the whole older group (n = 31). Values for language were available for 30 individuals. In addition, there were missing data for BDRS (one missing case), and the imaging measures (five missing cases). Due to variability in sample size, we replicated the correlations for visual abilities and language in the same sample than for the analyses on the global deviation index (n = 23). The pattern of results was the same (data not shown). Since depressive symptomatology was measured with BDI (n = 22) or GDS (n = 9), original scores were Z-transformed in order to combine both measures. Gender was coded as 0 female and 1 male. Significant BH corrected results (or trends) are highlighted in bold for easier reading. Patterns of significant uncorrected results are shadowed in light blue (or light brown when results remain significant after the BH correction). ^aPartial correlation, covariate: gender; ^bAll volumetric imaging variables are corrected for intracranial volume. BH corrected p-value = p-values are adjusted using the Benjamini–Hochberg’s correction for multiple testing; WAIS-III, Wechsler Adult Intelligence Scale - Third Edition; BDRS, Blessed Dementia Rating Scale; FAQ, Functional Activities Questionnaire.