An active social life can stave off dementia in the elderly, but just how this works is not clear. Does having lots of friends and a full calendar actually retard the accumulation of plaques and tangles, and in that way preserve brain function, or does frequent human contact somehow mitigate the unrelenting accumulation of damaging proteins? To answer that question, David Bennett and colleagues at Rush University, Chicago, Illinois, and the University of Pennsylvania, Philadelphia, are following elderly subjects as they age, measuring cognitive decline right up to death, and then performing postmortem measurements of plaque and tangle pathology. Their initial results on the first 89 subjects, published in the May issue of Lancet Neurology, currently available online, support the buffer model: Elderly folks with larger social networks had better cognitive function as they aged, without any diminution of plague and tangle accumulation.

The finding that social networks seem to modify the effect of pathology on brain function echoes Bennett’s previous work showing a similar phenomenon with education, another factor known to decrease the risk of dementia (Bennett et al., 2003). While only correlative, the findings are consistent with the theory that people have a neural or cognitive reserve, which allows them to tolerate the build-up of plaques and tangles to a point before impairment shows itself, and identifies social contact as a potentially important factor in establishing or maintaining that reserve in the aging brain.

The subjects, part of Rush Memory and Aging Project, were tested at baseline and then annually with a battery of 19 tests covering a wide variety of memory and processing tasks. All subjects were pre-dementia at the beginning of the study, and at that time the researchers counted up the number of children, family, and close friends that the subjects reported seeing at least once a month to arrive at a number for social network size. At death, plaques and tangles were quantified in brain sections after silver staining or immunostaining with amyloid or phospho-tau antibodies. Taking all the data, the researchers used linear regression to determine the relationship between cognitive function (expressed as a global z score of all tests), disease pathology (global, or amyloid- or tau-specific) and social networks.

In the tests of cognition, the study subjects showed average performance at baseline, and a decline over the test period. When researchers compared cognitive scores obtained just before death with pathology measures, linear regression revealed a relationship between lower global cognition and higher pathology detected at autopsy. In contrast, the extent of social engagement did not correlate with the number of plaques and tangles, suggesting that higher social activity did not result in a reduction in brain pathology. But when the statistical model was altered to allow interaction between pathology and social networks, the researchers discovered a significant effect of social networks on the relationship between cognition and global pathology. The net result was that for each unit increase in plaque and tangle pathology, there was a smaller decline in cognitive tests in the people with more extensive social networks. In comparing plaques and tangles, the researchers found that cognitive decline most strongly correlated with tangle pathology, and this effect showed the strongest interaction with social network size.

People with wide social circles are more likely to engage in activities that require physical and mental activity, all of which may protect against AD. But the association with networks was not changed by controlling for these factors or by taking into account the presence of depressive symptoms or chronic medical conditions in the subjects. In the subdomains of cognitive function, the most pronounced effects of social network showed up on the correlation of overall pathology to semantic memory, and of tangles to episodic, semantic, and working memory.

To illustrate the protective effects of social networks, the authors modeled cognitive function versus pathology for high versus low social scorers. Their results showed a dramatic difference: In the top 10 percent of social networkers (13 members), mental function remained nearly steady, even in the presence of the most severe global pathology. Subjects in the lowest 10 percent of social networks displayed a significantly stronger decline in global cognitive scores with increasing pathology. The latter group consisted of only two members, and more data from the ongoing study will be helpful to nail down the magnitude of the protective effect. So far, more than 1,100 people have enrolled in the aging study, so we can look forward to a wealth of data in the coming years.

What might separate the social butterfly from the wallflower in terms of sensitivity to AD? As the authors conclude, “These data provide evidence that the extent of social networks, or something related to social networks, provides some type of reserve which reduces the deleterious effect of Alzheimer’s disease pathology on cognitive abilities in old age.” Different people vary widely in their ability to make and maintain social ties, and the authors point out that many of the regions of the brain involved in social cognition also support episodic and semantic memory. It is possible that the very social have a stronger underlying brain circuitry in these regions, or that they more readily access alternative circuits when the need arises. Either way, the message is the same—it can’t hurt to get out there and mingle early and often.—Pat McCaffrey


No Available Comments

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Education modifies the relation of AD pathology to level of cognitive function in older persons. Neurology. 2003 Jun 24;60(12):1909-15. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . The effect of social networks on the relation between Alzheimer's disease pathology and level of cognitive function in old people: a longitudinal cohort study. Lancet Neurol. 2006 May;5(5):406-12. PubMed.