Neuroanatomical correlates of selected executive functions in middle-aged and older adults: a prospective MRI study
Introduction
Cognitive aging is a selective process. Advancing age is associated with significant declines in performance on tasks that demand substantial mental effort, rely heavily on processing speed, and are characterized by novelty and complexity of the stimuli, whereas the tasks for which success depends on overlearned skills, knowledge, and expertise are relatively immune to the ill effects of senescence (Horn, 1986). For example, executive functions and working memory show significant decline with advancing age (see Salthouse, 1994, West, 1996 for reviews); vocabulary appears relatively stable in middle to later adulthood (e.g. Alwin & McCammon, 2001).
This selective aging of mental abilities is contemporaneous with differential age-related changes in the brain (for reviews see Kemper, 1994, Raz, 2000). In vivo studies of normal human aging suggest the greatest age-related volume reduction in the prefrontal cortex (PFC), and the smallest in sensory cortices (Raz, 2000). Furthermore, while the preponderance of in vivo studies suggests that age-related volume reductions are greater in the cortex relative to the subcortical white matter, abnormal foci of white matter hyperintensity (WMH) in periventricular regions and the centrum semiovale are frequently observed on brain MR images of asymptomatic persons (De Leeuw, De Groot & Breteler, 2000). Pathophysiological origins of WMH are diverse, and they represent a manifold of cerebrovascular and neuropathological events. These events include cerebral hypoperfusion coupled with greater vulnerability of the watershed zones (Brant-Zawadzki, 1992), subclinical ischemia (Pantoni, Inzitari & Wallin, 2000), and état criblé that begins with age-related neuronal loss, results in axonal degeneration, and finally produces an extensive network of fluid-filled peri vascular spaces (Ball, 1989). The most frequently observed pathological correlates of WMH include gliosis (Chimowitz, Estes, Furlan & Awad, 1992; Fazekas et al., 1993; Fazekas, Schmidt & Scheltens, 1998), myelin pallor (Awad, Johnson, Spetzler & Hodak, 1986; Fazekas et al., 1993, 1998; Takao et al., 1999), atrophy of the neuropil (Fazekas et al., 1998), and breakdown of the ependymal ventricular lining (Leifer, Buonanno & Richardson, 1990; Scarpelli et al., 1994). The significance of WMH in understanding the relationship between the aging brain and behavior is underscored by the fact that white matter abnormalities observed on neuroimaging are associated with attenuations in performance on tasks of processing speed, immediate and delayed memory, and executive functions (Gunning-Dixon & Raz, 2000).
The term “executive functions” is somewhat vague and ill-defined. It encompasses a broad range of skills and abilities, such as planning, self-monitoring, inhibiting prepotent responses, and altering behavior in response to changing task demands. A frequently used example of a failure of an executive function is perseveration on the Wisconsin Card Sorting Test (WCST). Although perseveration on the WCST may occur for a host of reasons, structural equation modeling of executive functions suggests that perseverative errors are primarily related to the ability to shift cognitive sets (Miyake, Friedman, Emerson, Witzki & Howerter, 2000). Neuroanatomical evidence suggests that age-related increases in perseverative errors on the WCST are associated with shrinkage of the prefrontal cortex (Raz, Gunning-Dixon, Head, Dupuis & Acker, 1998) and loss of axonal integrity in the prefrontal white matter (Valenzuela et al., 2000) however, these skills also depend on distributed cortical networks including multiple prefrontal regions and posterior association areas (Anderson, Damasio, Jones & Tranel, 1991; Berman et al., 1995; Eslinger & Grattan, 1993).
Working memory (WM), an age-sensitive ability that allows simultaneous short-term storage and processing of information (Salthouse, 1994), is also linked to perseverative errors on the WCST (Dunbar & Sussman, 1995; Hartman, Bolton & Fehnel, 2001; Kimberg & Farah, 1993; Raz et al., 1998). Working memory tasks vary in complexity and processing demands. In some tasks, success depends on simple storage and maintenance of information, in others, active manipulation of information is necessary. Functional imaging studies in young adults most frequently implicate parietal association cortices in the temporary storage of information (Cabeza and Nyberg, 2000, Paulseau et al., 1993) and, prefrontal regions, to varying degrees, in the storage, maintenance, manipulation and updating components of working memory (D’Esposito et al., 2000, Smith and Jonides, 1999). Regarding aging of working memory, limited neuroanatomical evidence suggests that age-related declines on complex span tasks, which presumably rely on all of the aforementioned working memory components, may be linked to shrinkage of the prefrontal and visual association cortices (Raz et al., 1998).
The complex nature of perseveration and working memory calls for examination of multiple aspects of brain structure and function. It is plausible, for instance, that regional age-related shrinkage of the prefrontal cortex and structural alterations in the cerebral white matter exert additive influence on these cognitive functions. Thus, a simultaneous examination of those neuroanatomical parameters may yield significant non-redundant information. Such an assessment of the relative contribution of select cerebral cortical regions and subcortical white matter to an age-sensitive function is the focus of this report.
Our objective was to examine the role of specific brain regions in age-associated differences in working memory and perseveration. On the basis of the reviewed literature, we predicted that both working memory and perseveration would be associated with shrinkage of the prefrontal cortex, and increased WMH volume in subcortical frontal regions. Whether the influence of the WMH and the PFC volume on perseveration would be additive or redundant was a question to which we sought an answer in the data without preconception. We expected to observe a number of dissociations between the PFC and the control region indexed by the volumes of the fusiform gyrus (FG) and of the temporal lobe WMH. We predicted that neither the fusiform gyrus volume nor temporal lobe WMH burden, both of which evidence significant age-related differences (Raz et al., 1998, also see Gunning-Dixon & Raz, submitted, for a review), would mediate age-related differences in perseveration. While some of our previous MRI findings indicate that inferior parietal cortical volumes do not evidence age-related shrinkage and correlate only modestly with working memory, age-related volume reductions in the FG mediate age-related differences in working memory performance (Raz et al., 1998). Thus, we chose the FG as a control region and predicted that the FG volume differences would contribute to the age-related declines in working memory.
Section snippets
Participants
The data for this study were collected in an ongoing investigation of neuroanatomical correlates of age-related differences in cognition. Participants completed a mail-in health questionnaire and underwent a telephone interview as the means of screening for history of neurological and psychiatric conditions, head trauma with loss of consciousness exceeding five minutes, alcohol and/or drug abuse, hypo- or hyperthyroidism, and diabetes. None of the participants included in this sample reported
Results
To reduce the number of variables in the models (in keeping with the customary 1:10 variable per observation ratio) and for the lack of specific hypotheses about hemisphericity, we combined hemispheric volumes for each region. To compensate for body-size differences between the sexes, all cortical ROIs were adjusted for height using a formula: . In this formula, ROI volumeraw is subject’s raw volume of a given ROI, ROI volumeadj is subject’s
Discussion
The main result of this study is that both the prefrontal cortical volume and the integrity of the subcortical frontal regions contribute almost equally to increase in perseveration observed in middle-aged and older adults. This finding refines the classic view that perseverative behavior on the WCST is linked to the frontal lobe integrity (Milner, 1963). Although a quarter of the covariation between perseveration and age was due to unidentified causes (in accord with Salthouse, Mitchell,
References (60)
“Leukoaraiosis” explained
The Lancet
(1989)- et al.
Physiological activation of a cortical network during performance on Wisconsin Card Sorting Test: A positron emission tomography study
Neuropsychologia
(1995) - et al.
Brain imaging of the central executive component of working memory
Neuroscience and Biobehavioral Reviews
(2002) - et al.
Frontal lobe and frontal-striatal substrates for different forms of human cognitive flexibility
Neuropsychologia
(1993) - et al.
The unity and diversity of executive functions and their contributions to complex frontal lobe tasks: A latent variable analysis
Cognitive Psychology
(2000) The assessment and analysis of handedness
Neuropsychologia
(1971)- et al.
Magnetic resonance imaging shows orientation and asymmetry of white matter fiber tracts
Brain Research
(1998) - et al.
Brain volume estimation from serial section measurements: A comparison of methodologies
Journal of Neuroscience Methods
(1990) - et al.
Dual voxel proton magnetic resonance spectroscopy in the healthy elderly: Subcortical-frontal axonal N-acetylaspartate levels are correlated with fluid cognitive abilities independent of structural brain changes
Neuroimage
(2000) - et al.
Aging, cohorts, and verbal ability
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences
(2001)