Dr. Kennedy is professor in the Department of Psychiatry and Behavioral Sciences at Albert Einstein College of Medicine, and director of the Division of Geriatric Psychiatry at Montefiore Medical Center in the Bronx, New York.
Disclosure: Dr. Kennedy reports no affiliation with or financial interest in any organization that may pose a conflict of interest.
Please direct all correspondence to: Gary J. Kennedy, MD, Director, Division of Geriatric Psychiatry, MMC, 111 East 210th St, Klau One, Bronx, NY 10467; Tel: 718-920-4236; Fax: 718-920-6538; E-mail: email@example.com.
Criteria for Alzheimer’s disease and preclinical dementia have been proposed recently, which include potential biomarkers of the illness. Nonetheless, the etiology of the illness remains uncertain despite consistent associations described for cerebral amyloid and hyperphosphorylated tau pathologies. As a result, further progress toward understanding age-related changes in cognition that are not related to dementia is critical both to characterize healthy aging but also to develop interventions that will sustain cognitive performance. This will be the case even if proposed biomarkers become powerful predictors of the presence of disease.
Preliminary criteria for diagnosis of Alzheimer’s disease in both its clinical and pre-clinical forms have appeared in the proposed Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition,1,2 and subsequently by work groups convened by the National Institute on Aging and the Alzheimer’s Association.3-5 Both suggest that bio-markers related to amyloid or the microtubule protein tau soon may be incorporated into the diagnostic process. The appeal of putative biomarkers of dementia is their promise of signaling the presence of a disease process before cognition or brain topography become noticeably impaired. In addition to anteceding the onset of symptoms, biomarkers might be less subject to variations seen in healthy cognitive performance related to education, vocation, or innate intelligence. The proponents of biomarker research assume that amyloid is the signal event in Alzheimer’s disease. Amyloid and hyperphosporylated tau are pathology consistently associated with Alzheimer’s disease. But how distal are these pathologies from the pathoetiology of the illness?
Uncertainty about the primary pathophysiology of Alzheimer’s disease has raised skepticism about the use of biomarkers.6 Yet, interventions applied once synaptic atrophy and neuronal death are manifest as cognitive impairment may be too late to be effective. Further, intervention trials to test the amyloid hypothesis among patients selected on the basis of these preliminary criteria could take a decade to complete. Even if the spread of amlyoid in the brain can be reduced or reversed, the demonstration of preserved or improved cognition will be required to establish efficacy. As a result, attention to progress in the characterization age-related changes in cognition is critical to refining criteria for preclinical dementia. The study of working memory and executive function now underway in the Research Domain Criteria initiative of the National Institute of Mental Health7,8 may further advance the assessment of cognitive processes and neural circuits sensitive to the earliest signs of Alzheimer’s disease.
Cognitive constructs refer to mental processes which have both scientific and clinical utility yet are approximations of reality based on observation and theory. Functional imaging has revealed age-related changes relevant to the theoretical constructs, but no unified phenomenon which might represent a simple theory of aged cognition has emerged. Nonetheless, awareness of these constructs and how they are changed for better or worse with age will have clinical implications when genuinely therapeutic agents arrive for the treatment of dementia. Prominent constructs recently reviewed by Reuter-Lorenz and Park9 appear in bold followed by descriptions.
Working memory, that component of the cognitive system which retains information for immediate use, is an area of intense interest for clinicians caring with older adults but for neuroscientists as well. Memory screening tests most often rely on working memory by requiring the patient to register, retain (learn), and recall (remember) new information such as a recited list of words or set of image presentations. Older adults perform as well as younger provided the memory load does not exceed four items or require marked executive function to inhibit, reorder, or refresh the list. Thus, when older adults are asked to select from the assortment of recently learned memories or to alternate categories of items, working memory becomes fatigued. Unlike consolidated memory, which seems to have infinite shelf space, working memory is volume dependent and vulnerable to overload. As the memory load increases, both younger and older adults recruit prefrontal cortical areas to manage the load. Older adults reach overload sooner and show a drop in prefrontal activation, suggesting the system has met its limit.
Inhibitory control deficits appear when older adults are given memory tasks in which distractors are also presented. In comparison to a younger adult, the older person is less likely to ignore, screen out, or delete irrelevant stimuli. When instructed to remember a sequence of words or images followed by items to be ignored, older adults show greater brain activation for the latter than do younger adults. When instructed to generate a list of words starting with the letter “S” followed by an instruction to list words starting with the letter “A,” older adults are more likely to insert S-words into the A-word list. Thus, age-related inhibitory dysfunction mediated by prefrontal processes results in impairments in the initial stage of information processing, placing further limits on working memory capacity and efficiency.
Processing speed decrements are the most widely accepted explanation for decline in cognitive processes during late life. Changes in white matter structure and integrity are largely responsible. It is as though age and cardiovascular illness fray the insulation in neural circuits. However, the effects are not uniform. Some neural circuits and the cognitive processes they serve may be more intact and more capable of compensating for those with less integrity. As a result, slowed processing speed is not considered a sufficient explanation for cognitive decline during aging.
Long-term memory deficits have been ascribed to a number of age-related changes in brain structure and cognitive function. Older adults are less effective at encoding new information for memory tests. However, when given contextual or categorical cues associated with the memory item, their performance improves. Such tests of episodic memory are also sensitive to loss of volume and under-activation of the hippocampus and parahippocampus, two areas affected early in Alzheimer’s disease. Implicit, automatic, or procedural memory functions out of awareness and is related to previously learned material that can be applied to current tasks with little conscious effort. This form of effortless recall, particularly when associated with semantic processing, involves left inferior prefrontal regions and is relatively preserved in older people.
Functional imaging studies with positron emission tomography or functional magnetic resonance imaging scans have provided a number of discoveries about regional differences in the aging brain. For example, compared to younger people, older adults will activate a greater number of brain regions to meet the same cognitive challenge. Hemispheric dominance, whether for verbal processing on the left or spatial processing on the right, is diminished such that functional asymmetry and localization are reduced. Over-activation is also seen in both posterior and anterior regions of cortex accompanied by a general posterior to anterior shift in activation. The phenomenon is thought not to be a result simply of cerebrovascular aging. Over-activation may be associated with superior cognitive performance and represent compensatory enhancement of neural circuitry. However compensation has its price.
The medial prefrontal, medial, and lateral parietal brain areas are known as the default network. These regions are highly interconnected, more active at rest than during purposeful activity, and associated with internal rather than external stimuli. The default network manages ongoing attention to the environment, self-focus, and reflective memories. However, with advancing age, the network loses interconnectivity and over-reacts to external stimuli. As a result, frontal areas are recruited to compensate, causing loss of efficiency and accuracy.
De-differentiation is the result of lost topographic specificity and decline in neural plasticity. Additional regions of cortex have to be recruited, not to reach a new equilibrium, but rather to respond to loss of specialization. For example, face recognition is specific to the ventral visual cortex, but as this area loses functionality with age, prefrontal areas are recruited to manage the work load placed on working memory. There is a general posterior to anterior activation in the cortex with the medial, lateral, and anterior prefrontal cortex being over-activated to compensate for under-activation in the medial temporal lobe and ventral visual cortex.
Frontal over-activation makes older adults vulnerable to age- and illness-related prefrontal deficits. In addition, there is an increased noise to signal ratio. As dopamine levels decline with age, the strength of synaptic signaling falls while the background neural noise does not.
Older adults are also more likely to exhibit difficulties with proactive versus reactive cognitive control. Because of changes in executive function and prefrontal structures, older adults are less able to benefit from cues and context that might precede a sequence of stimuli. Rather, they rely more than younger people on cognitive procedures that occur during stimulus presentation. This reduces processing speed as well as the stimulus load that can be successfully processed. Thus, their executive function is more reactive than proactive or anticipatory. They multi-task with difficulty.
How, then, do we explain the increasing proportion of older adults who maintain sufficient cognitive function to remain independent into late life?”10 Reuter-Lorenz and Park9 present two hypothetical mechanisms to account for the maintenance of cognitive performance in late life, shown in the Figure. The Compensation-Related Utilization of Neural Circuits hypothesis suggests that cognitive processes become rerouted to new or additional circuits as age and illness wear the brain down. As one area of mental hardware deteriorates, another is recruited to take its place. A virtual scaffold gradually emerges and is made possible by the distribution of cognitive processes to frontal and other areas, including both hemispheres, and to new neurons via neurogenesis. The Scaffolding Theory of Aging and Cognition adds the notion that the brain’s software may be enhanced at the same time that the circuitry is being upgraded. The scaffold is enhanced by learning, physical exercise, cognitive stimulation, and social engagement. The impact on cognitive performance will vary as a result of both the quality of the structure of the scaffold and personal behavior. In this way, not only do age and illness affect cognition, but so do personal history, ongoing mental activity, and the environment.
Advances in cognitive neuroscience buttressed by interest in dementia biomarkers and functional imaging techniques promise to increase the measurement of risk for the development of Alzheimer’s disease. However, equally important are insights into compensatory mechanisms and the plasticity of neural circuitry that may argue for interventions which might sustain if not improve cognitive performance to the end of the life span.11 This will be particularly important when treatments emerge to modify the disease process and slow the rate of decline among people with dementia. An understanding of how biomarkers might predict dementia will not obviate the need to advance our understanding of aging and cognition in healthy active older adults. PP
1. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Association. In press.
2. Kennedy GJ. Proposed revisions for the diagnostic categories of dementia in the DSM-5. Primary Psychiatry. 2010;17(5):26-28.
3. Alz.org. Proposed criteria for Alzheimer’s disease dementia. Available at: www.alz.org/research/diagnostic_criteria/dementia_recommendations.pdf. Accessed August 10, 2010.
4. Alz.org. Proposed criteria for mild cognitive impairment due to Alzheimer’s disease. Available at: www.alz.org/research/diagnostic_criteria/mci_reccomendations.pdf. Accessed August 10, 2010.
5. Alz.org. Proposed criteria for preclinical Alzheimer’s disease. Available at: www.alz.org/research/diagnostic_criteria/preclinical_recommendations.pdf. Accessed August 10, 2010.
6. Kolata G. In Alzheimer’s research, hope for prevention. The New York Times. August 5, 2010: A18.
7. Insel T, Cuthbert B, Garvey M, et al. Research domain criteria (RDoC): toward a new classification framework for research on mental disorders. Am J Psychiatry. 2010;167(7):748-751.
8. NIMH Research Domain Criteria (RDoC). Available at: www.nimh.nih.gov/research-funding/nimh-research-domain-criteria-rdoc.shtml. Accessed August 12, 2010.
9. Reuter-Lorenz PA, Park DC. Human neuroscience and the aging mind: a new look at old problems. Journal of Gerontology: Psychological Sciences. 2010;65B(4):405-415.
10. Fries JF. Aging, natural death, and the compression of morbidity. N Engl J Med. 1980;303(3):130-135.
11. Rae MJ, Butler RN, Campisi J, et al. The demographic and biomedical case for late-life interventions in aging. Sci Transl Med. 2010;2(40):40cm21.