Comments on the Executive Control of Clock‐drawing

To the Editor: We would like to comment further on Lee and Lawlor's observation that the clock-drawing task (CDT) is impaired in geriatric depression.1 They suggest that the CDT's deterioration in depression might be understood on the basis of “frontotemporal dysfunction.” We agree. A subject's performance on any neuropsychological measure will reflect both the integrity of the specific domain being tested, in this case visuoconstructional praxis, and its executive control. Executive control functions (ECF) guide complex goal-directed behavior in the face of novel, irrelevant, or ambiguous environmental cues. The executive demands of clock-drawing are shared by the independent living skills, i.e., cooking, dressing, toileting, shopping, and medical treatment compliance. In this sense, clock-drawing performance may mirror the performance of more functionally relevant behaviors that are also under executive control.2 Executive dyscontrol is associated with cortical and subcortical frontal system lesions.3 Frontal system pathology is associated with both Alzheimer's disease (AD) and major depression. Frontal cortical synaptic density, for example, is the strongest reported pathological correlate of dementia severity in AD.4 Frontal cortical metabolic deficits correlate significantly with depression severity as measured by the Hamilton depression scale,5 and total frontal cortical volume by magnetic resonance imaging is reduced in depressed older adults.6 These changes are superimposed on selective reductions in total frontal cortical volume by magnetic resonance imaging in healthy older people.7 Relatively minor age differences are observed in total cerebral, temporal lobe, or amyg-dala/hippocampal volumes. The CDT has traditionally been viewed as a visuospatial task, sensitive to right hemisphere pathology.8 While AD is associated with signs of impairment in this region (visual agnosia, dyspraxia, and apraxia), major depression is not. CDT failures in noncortically impaired subjects challenge a chiefly visuospatial conceptualization of the CDT. We believe CDT failures can asrise from either right hemisphere constructional deficits, frontal system executive dyscontrol, or both. The relative variance in CDT performance explained by ECF independently of visuoconstructional praxis remains to be determined. This is because (1) current CDT rating schemes are designed to elicit constructional failures rather than ECF related failures, (2) bedside mental status exams are either indirectly sensitive to ECF failures or ignore them altogether, and (3) the possible qualitative differences in CDT failures arising from constructional rather than ECF-related pathology are not assessed routinely. We expect that a significant proportion of the variance in CDT failures is, in fact, the product of executive dyscontrol. This could explain the observation that clock drawing can be spared when subjects are asked to copy a design, but fail when they are given a blank sheet of paper to work from. We have been formulating a conceptual model of dementia that emphasizes the loss of ECF and its relationship to qualitative clinical differences that might distinguish cortical from noncortical dementias.3 The CDT provides an opportunity to observe both the executive control of clock drawing and qualitative differences in clock drawing failures. Qualitative differences have been observed in how AD and Huntington's disease (for example) fail the CDT after they have been matched to dementia severity.9 The potential of this approach is reflected in Figure 1. . Probit regressional model of “clockness” made by a blind rater as a function of EXIT25 and QED scores among 24 AD cases (R = .62). Model describes dichotomous rating of “clocklike” = 1 or “not clocklike” = 0. Blind assessments agree 87% [x2 (df · 1) 10.75, P = .001]. Figure 1 presents a probit model of the relative contribution of dementia severity [as measured by the Executive Interview (EXIT25)10] and qualitative dementia type [as measured by the Qualitative Evaluation of Dementia (QED)11] to predict whether a blind rater will assess the subject's CDT production as being grossly “clock-like.” EXIT25 scores > 15/50 best discriminate demented from non-demented adults.12 QED scores > 15/30 are associated with the cortical type clinical features associated with AD. Note how QED scores increase linearly with the EXIT25 in this AD sample. This reflects the invasion of multiple cognitive domains by cortical type clinical features. A QED score of 17/30 best discriminates AD from both noncortical dementias (i.e., the dementia of depression) and normal older controls.12 Our model would predict that mild to moderate ECF deficits alone will disrupt a clock's detail, but spare it's overall form. Visuospatial function must also be impaired before the figure loses its recognizibility as a clock. In this small sample (n = 24) of AD cases, the EXIT25 and the QED explain 62% (r = .79) of the variance in a blind rater's subjective dichotomous assessment of “clockness.” Note that the loss of “clockness” occurs in association with both executive dyscontrol (an EXIT25 score of 30/50 is near the mean for institutionalized residents of AD special care units) and a qualitatively cortical dementia type. The CDT's potential as an ECF measure is significant because executive dyscontrol predicts functional outcomes across a variety of conditions.13–15 There are, at present, only a few valid bedside measures of executive dyscontrol. An ECF-rated version of the CDT might prove to be a useful adjunct to the busy clinician's need to predict functional status and need for services from a minimal database.