|
Cognitive Neuroscience of Human Memory
Summary: Randy Buckner is interested in determining brain systems that support human cognition and how these systems are altered in neuropsychiatric disorders.
Weighing about three pounds, the human brain contains billions of individual brain cells that are wired together in complex but highly structured networks. The resulting architecture rivals the world's largest computer in terms of processing capacity and is unique in its ability to adapt to changing situations. The human brain is also fragile. Development of abnormal function in childhood or disruption in advanced aging leads to disorders of cognition. Thus, in addition to the great intrinsic interest in understanding how the human brain works, understanding how brains become disordered will shed light on schizophrenia, autism, Alzheimer's, and other diseases with tremendous emotional and financial costs.
The Architecture of Human Brain Systems
Our work integrates multiple human imaging approaches to characterize human brain systems and determine abnormalities that associate with neuropsychiatric disorders. Until recently, the barrier to understanding the human brain was that invasive techniques could be used only with animals, causing a gap between the understanding of neural circuits being unraveled using animal models and the study of the human brain. Human imaging technologies changed that, and today, images of activity in the thinking brain are a familiar sight.
In the mid-1990s we developed an approach that allows indirect measurement of neural activity associated with individual cognitive events. Using this method, we discovered a distributed set of brain areas active when individuals successfully recognize items from long-term memory but not when they are attempting to remember and fail. Activity within these areas can predict whether items will be correctly retrieved and whether a person will report having a particularly vivid recollection. Using methods that can identify functional connections between distributed areas, we further found that these areas are part of a distributed brain system that is linked to areas specialized for memory function within the medial temporal lobe.
We are currently exploring how this distributed memory system is used during decision making when information from the past, and not information in the immediate environment, is most relevant. What has emerged from this research is the possibility that memory systems and perceptual systems provide competing sources of information. This unexpected observation suggests that normal brain function is characterized by dynamic shifts between thinking about the past and extracting information from sensory information available in the immediate environment.
Of particular interest is how this competition between systems is controlled. Our upcoming studies are specifically exploring whether a breakdown in the control of memory systems contributes to neuropsychiatric disorders.
Brain Aging and Alzheimer's Disease
Fifty percent of adults over the age of 85 experience some form of dementia, including Alzheimer's disease (AD). AD begins with memory lapses and progresses to severe cognitive and behavioral dysfunction. We began studying AD motivated by the possibility that our observations about normal memory systems might shed light on the disease. In collaboration with John Morris (Washington University in St. Louis), we discovered several dissociations that suggest brain aging results from multiple cascades, each with different rates of progression and brain correlates. AD reflects a particularly devastating cascade that markedly affects the distributed network of areas linked to the medial temporal lobe memory system. Other cascades fall within the domain of "normal" aging, including subtle changes to communication between regions. Effects associated with normal aging can be distinguished by the anatomic systems targeted and their slow rates of progression.
An important event occurred in 2004 when a safe molecular marker of fibrillar amyloid deposition became available. Fibrillar amyloid is the protein constituent of plaques—one of the neuropathological indicators of AD. It is now possible to image amyloid plaques in the living human brain, including in individuals who show no symptoms of the disease. Using this molecular marker, we have examined the influence of amyloid deposition on cognitively normal individuals and found that amyloid deposition is correlated with AD-like changes in the brain before symptoms appear. These findings demonstrate that amyloid deposition in clinically normal older adults is associated with brain dysfunction that is most likely preclinical AD. To test this hypothesis, we are currently following these individuals longitudinally.
We are also exploring why amyloid deposition preferentially targets certain brain areas. One surprising discovery is that the brain regions most vulnerable to AD are those with the highest activity levels in normal young adults, suggesting that local metabolism may accelerate the earliest stages of the disease. Of particular interest, the regions with preferentially high activity and metabolism include those linked to memory systems, suggesting a reason why memory abilities are affected early in the progression of AD.
Genetic Mechanisms
Many neuropsychiatric disorders run in families, suggesting a strong genetic component. For example, a child with an autistic sibling is 25 times more likely to develop the disorder than his peers. To better understand the underlying genetic mechanisms that influence brain function and risk for neuropsychiatric disorders, we and our colleagues have recently begun to study the link between genetic variation and brain function. This line of inquiry has led to the development of approaches that focus on the individual and methods that can specifically measure features of brain organization that indicate atypical brain development.
In the past, human neuroimaging techniques have had to combine measurements from many people, providing a fictional "average brain." But even the brains of normal individuals vary considerably, so abnormal function has to be distinguished from this diversity. Moreover, common disorders are likely products of abnormalities in multiple brain systems and gene variants. By harnessing advances in magnetic resonance imaging (MRI) scanner technology, including new coil designs and fast imaging protocols, we are measuring the detailed activity and structure in thousands of individual brains, with the goal of understanding the causes of common variation.
This research has resulted in methods able to estimate a number of markers of successful brain development. For example, the left hemisphere takes on language and verbal memory function in most individuals. In autism and schizophrenia, the brain does not successfully specialize to the typical degree. We developed a robust method to measure functional asymmetries across multiple brain systems. Atypical differentiation of brain systems may provide a window into abnormal development. We are presently using estimates of brain asymmetry and other measures of functional specialization to explore how genetic risk factors for neuropsychiatric disorders influence the functional architecture of the human brain.
This work was supported in part by grants from the National Institutes of Health and the Simons Foundation.
As of May 30, 2012
|
