ABSTRACT
Executive functions come into play when two or more sources of information, streams of thought, or behavioral tendencies need to be processed in the interests of establishing, maintaining, and appropriately terminating planful, goal-directed action. It is underpinned by protocognitive or fundamental processes such as attentional focus and selectivity, inhibition and suppression of irrelevant, misdirected, or superfluous behaviors, and ultimately the sequential organization of output. Demands on executive function are greatest when remembered or habitual solutions to external challenges are inadequate, necessitating the elaboration of novel solutions. At this level, executive processing is a controlled process, involving the ability to represent the elements of the problem in a temporary and accessible ‘‘online’’ store, to manipulate them strategically, and to shift flexibly from approaches that are unlikely to be effective or appropriate to those which have a greater chance of success. Controlled processing occurs within a capacity-limited, and frequently refreshed group of subsystems collectively referred to as working memory. The neurocognitive architecture of executive function is inescapably hierarchical.
Much of the current edifice of our understanding of executive process was built on
a study of cases with primary injury to frontal cortex with little or no involvement of basal forebrain, diencephalic, or brain stem mechanisms of behavior. In parallel with this dominant trend, however, there was a growing view that the motor functions of the basal ganglia are complex enough to merit descriptors, such as programming, sequencing, and motor planning (Delwaide & Gonce, 1993; Marsden, 1982). Anatomical studies began to overturn the classical notion that output from the basal ganglia terminates in a single nucleus of the thalamus, which in turn projects to premotor cortex. It is now clear that the basal ganglia are in a position to gain access to much wider regions of the frontal cortex on the medial and lateral aspects of the cerebrum. Recent years have seen an explosion of interest in the behavioral implications of the frontostriatal circuitry. Parkinson’s disease (PD), together with its clinical and pathological congeners, has been one of the major drivers of this quest, and much of what is now known about the frontostriatal basis of executive dysfunction has come from research on this condition. In an ironic counterpoint to James Parkinson’s original assertion (Goldman &
Goetz, 2005) that the ‘‘intellect is uninjured,’’ PD has been the subject of a voluminous body of neuropsychological research (Henry & Crawford, 2004; Levin, Tomer, & Rey, 1992; Rashkin, Borod, & Tweedy, 1990). Psychometric studies, predominantly from the 1970s onwards showed that patients with PD performed poorly on a wide range of tasks when compared with healthy controls (Pirozzolo, Swihart, Rey, Mahurin, & Jankovic, 1993), suggesting multiple cognitive impairments, even at a relatively early stage of the disease. With a growing recognition of the possibility that this very general performance failure might be an artifact of the cardinal motor features of the disease, attempts were made to define the nature of the putative cognitive disorder more precisely. The most commonly reported impairment fell within the visuospatial domain. Nevertheless, convincing evidence of a fundamental spatial disturbance was not forthcoming. Before long the visuospatial impairment of PD came to be seen as a by-product of executive disturbances such as attentional set shifting or disturbances in the sequential execution of movement in the absence of external guidance (Brown & Marsden, 1990; Freeman et al., 2000). Increasing attention was drawn to the fact that executive-like disturbances were not restricted to tasks with visuospatial content, bolstering the notion that higher cognitive impairment in PD was indeed of an executive type (Brown & Marsden, 1990; Taylor, Saint-Cyr, & Lang, 1986), with greater similarities to frontal than to posterior cortical disorders. This picture has emerged against a background of disagreement (Saint-Cyr,
Taylor, & Nicholson, 1995). PD is a heterogeneous and progressive condition. Variations in the pattern of motor signs and stage of disease have, in all likelihood, played a major role in the complexity of the neuropsychological findings over the past 40 years. Cognitive impairments become more extensive as the disease progresses, and around 20% of patients develop a dementia. By contrast, the early phases of the disease are characterized by circumscribed losses of cognitive function, often restricted to particular forms of executive dysfunction. This is paralleled by restriction of the anatomical pathology to the nigrostriatal system, and involvement of nondopaminergic systems is not as extensive as it is in later stages of the disease. As two authorities have remarked, ‘‘Parkinson’s disease has become the prototypic
