The Aging Brain: Is It Less Connected?

Age-related cognitive decline may be caused by a breakdown in connections between different brain systems.

Time can wreak havoc on the brain. Age-related cognitive decline comes with a wide range of symptoms, from memory loss to problems with concentration. But what causes these symptoms? What happens in the brain of people as they age?

Jessica Andrews-Hanna and her colleagues at Harvard University, the University of Michigan and Washington University School of Medicine have explored the possibility that cognitive decline during aging results in part from a loss of coordination and communication among large-scale brain systems. Taking off from the concept of “disconnection syndromes” put forth by the late neurologist Norman Geschwind many years ago, they used magnetic resonance imaging to examine the brain activity of young (ages 18 to 34) and old (ages 60 to 93) subjects when they were either resting or engaged in certain tasks, such as classifying words into different categories. The researchers then analyzed the fMRI data to compare fluctuations in activity between brain regions in response to the experimental task, thus providing a measure of coordination between different systems in the cerebral cortex. The authors compared the strengths of these correlations with measures of cognitive performance based on several tests of intellectual fluency and memory. 

The most important finding was a “dramatic reduction” in the functional correlation, or interaction, between two well-defined large-scale brain networks in the aged brain. (It’s worth noting, however, that about half of the older subjects in the Andrews-Hanna experiment overlapped in the level of synchrony between brain areas with the younger population, so this isn’t a universal feature of the aged brain.) The first network is associated with a resting “default” state of the brain, whereas the other network is responsible for the control of goal-directed behavior and “executive” brain functions, such as concentration on a particular task. The observed weakening of these brain network interactions was significantly correlated with a measured decline of cognitive functions as a result of aging.
A Matter of Arousal?

Among the most outstanding developments in functional MRI in recent years has been the elucidation of a “default mode” of brain function. This default mode refers to the consistent network of brain areas that show high levels of activity when subjects are asked to rest and not perform any tasks. Interestingly, these observations of a default mode persist across a wide variety of state changes, from wakefulness to deep anesthesia. This new study provides important evidence that the integrity of this default mode degrades with aging and is associated with cognitive impairment.

The question, of course, is what degenerative mechanisms may account for the breakdown in cortical “crosstalk” in the older subject population? One possibility is a potential decline in the neurotransmitter dopamine, which could produce cognitive slowing and impaired performance on measures of executive function as seen in their older subjects. This possibility leads to the following hypothesis.

To us, it appears likely that the Andrews-Hanna results also likely relate to well-characterized changes in generalized arousal associated with aging. As people get older, they become less responsive to novel and surprising stimuli, such as a car that’s driving in a dangerous manner. (Perhaps it has run a red light…) It seems possible that one important aspect of these ascending arousal systems is to provide mechanisms for keeping the resting brain in a state of near readiness for action, a “mental engine idling,” so to speak. Thus, you are able to quickly act and avoid a collision.

One idea would state that “default mode” activities actually depend, in part, on activities originating in primitive brainstem mechanisms, which are located as far posterior as the medullary reticular formation. This area has been characterized as the “first responders” of the brain, because it is often the first area to respond to relevant stimuli, such as that recklessly driven car.

Furthermore, contributions from many ascending arousal systems have been shown to maintain cerebral activation, which may play an important in the maintenance of proper interactions between various brain networks in the default state. In this context, it is interesting to note that severe cognitive impairment following brain injury can be treated through electrical stimulation of the central thalamus, an integrative brain area that receives inputs from the brainstem arousal systems and has strong connections to both the dorsal attention and the default-state networks, which are active when the brain is at rest. The findings of the Andrews-Hanna study are extremely provocative and suggest many lines of thought to follow-up and advance our understanding of how the human brain changes as we age.

Further Questions

People familiar with the technologies used in the Andrews-Hanna paper might ask about other useful measures that the authors did not employ, such as EEG? These technologies measure other aspects of brain function that might shed further light on the interaction of large-scale systems in the brain, especially as that interaction is affected by the aging process.
A second question, which is related to our arousal proposal: Could the results reported by Andrews-Hanna et al. really be generated entirely by white matter changes in the cerebral cortex, or could there also be important contributions from lower regions of the brainstem that, for technical regions, are difficult to image with fMRI? Here’s a metaphor: if a window in the third story of a house is jammed, that fact might be due to a flaw in the window, but it could also be the case that the foundation of the house is settling. Those primitive brain areas involved with arousal, such as the brainstem, are analogous to the foundation of the house. Andrews-Hanna and her colleagues have been concentrating on the “third floor window” and have provided the field with an important contribution.  We suggest a “foundation-related” proposal to add to the authors’ interpretation.


Mind Matters is edited by Jonah Lehrer, the science writer behind the blog The Frontal Cortex and the book Proust was a Neuroscientist.

Rights & Permissions
Share this Article:


You must sign in or register as a member to submit a comment.
Scientific American MIND iPad

Give a Gift & Get a Gift - Free!

Give a 1 year subscription as low as $9.99

Subscribe Now >>


Email this Article