Due to the pathology of Alzheimer's disease (AD), it is not uncommon for patients to struggle with cognitive tasks involving figure copying. However, figure copying requires involvement from multiple other domains, including visuospatial and frontal-executive abilities. Less clear is whether figure copying performance, as measured by the Rey-Osterrieth Complex Figure (ROCF), is more affected by visuospatial or frontal-executive compromise in patients with AD. This study aims to discover whether performance on the ROCF varies more with executive or visuospatial abilities in patients with AD.

A total of 156 patients (79 women, M age = 77.82, M education = 14.21) diagnosed with Alzheimer's disease (AD) participated in comprehensive neuropsychological assessment as part of outpatient neurology evaluations. In addition to the ROCF Copy trial, participants completed measures of visuospatial function (WAIS-IV Block Design & Picture Completion) and frontalexecutive functioning (Trails B, DKEFS Inhibition, WAIS-IV Similarities).

Canonical correlations revealed that ROCF Copy was significantly and positively related to the set of tests measuring visuospatial functioning, p < .001, and frontal-executive functioning, p < .001. Post-hoc bivariate correlations showed significant positive correlations between ROCF Copy and each visuospatial, ps < .01, and each frontalexecutive, ps < .04, measure. The relationship between ROCF Copy and each visuospatial measure was significantly stronger than the relationship between ROCF Copy and every frontal-executive measure except WAIS-IV Similarities, ps < .01. Those with visuospatial impairment (>1.5 SD) performed significantly worse on the ROCF Copy than those without impairment, p <.01. This difference persisted even when the effects of frontal-executive measures were controlled, p < .01. In contrast, those with frontal-executive impairment did not perform significantly worse on the ROCF Copy than those without frontal-executive impairment, p < .01. This did not change when controlling for the effects of visuospatial measures. Finally, the significant difference in ROCF Copy between those with mild cognitive impairment (MCI) and those with dementia (p < .01) disappeared when visuospatial measures were controlled, but it remained when frontal-executive measures were covariates, p < .03.

These findings suggest the figure copying of those with AD is significantly related to both their visuospatial and frontal-executive functioning, but the relationship with visuospatial functioning is stronger. Impairment in visuospatial, but not frontal-executive, functions, seems to have a negative impact on the figure copying of those with AD, and performance on visuospatial compared to frontal-executive measures better accounts for the weaker ROCF Copy scores among those with dementia relative to MCI. Therefore, the pathological effects of AD on figure copying appear to occur predominantly through visuospatial rather than frontalexecutive channels, and interventions for offsetting decline in figure copying may be most effective when targeting visuospatial abilities, such as visual perception and visual construction.