Visuographic tests of set shifting and inhibitory control. The contribution of constructional impairments

Luria's graphical or alternating sequences test (the 'rampart' design) is a well-known procedure to examine attentional processes such as set shifting and inhibition of competing responses. Although the test requires attentional control over inappropriate responses, it also relies on constructional abilities that may hamper the specificity of the test. In order to control for the influence of constructional abilities, the structural equation modelling (path analysis) was used to examine the correlations between the performance on alternating sequences and an independent criterion measure of executive functioning (the Behavioural Dyscontrol Scale), corrected for the influence of visuoconstructional ability as reflected by performance on a task of design copying. The tests were administered to 950 elderly participants in a psychogeriatric day-care programme. Performance on the Behavioural Dyscontrol Scale (BDS) and the copying tasks jointly predicted performance on the alternating sequences test (AST) (multiple R(2)=.41). Though the effect of BDS on AST was 38.9% mediated by constructional ability, BDS scores predicted performance on graphical sequences above and beyond the influence of drawing ability. In a non-parametric classification tree analysis, the performance on the AST was reliably predicted by drawing ability. When BDS performance was added to the analysis, the classification accuracy increased for participants with poor drawing, but remained unchanged for those who performed in the normal range on the copying tasks. As a task that relies on shifting and inhibitory control, the alternating sequences test is a feasible measure of executive functioning that is relatively independent of the influence of the component drawing skills.