The icon indicates free access to the linked research on JSTOR.

What is retirement? A time of leisure and relaxation, or the continuation of work at a slower pace? While the definition of retirement in the popular culture has typically been the former, the statistics are beginning to tell a different story.

JSTOR Daily Membership AdJSTOR Daily Membership Ad

In a study with Age Wave, a research firm that focuses on trends in aging, financial services company Merrill Lynch found that 47 percent of American retirees have continued to work or plan on doing so, while 72 percent of pre-retirees anticipate they will keep working in retirement. As the study’s authors put it, “In the near future it will become increasingly unusual for retirees not to work.”

At first glance, this may seem like a response to the recent financial crisis and its aftermath. In reality, the causes are more nuanced. While reduced pensions and financial uncertainty are two key factors affecting work in retirement, fully 80 percent of the study’s respondents agreed with the statement, “I work in retirement because I want to” (emphasis the author’s).

In other words, working retirees are not just looking for income. They are also often looking for purpose, stimulation, and personal growth. Nuanced indeed—but what if work could provide even more?

What if work also happened to be a key ingredient in preserving the brain?

Your Brain on Retirement

Much like good looks and good health, fluid intelligence is, like the saying goes, wasted on the young. The typical person sees her abilities peak around age 20, with a steady decline thereafter. While there is a great deal of variation among individuals, the trend is horribly irresistible: someone who scores 50 percent on a common fluid intelligence test at age 20 could expect to score only 10 percent on the same test at age 60.

It sounds rather grim. However, some people’s skills do not decline as rapidly as others. From this modest origin the “use it or lose it” hypothesis of cognition was born. It’s an aptly named idea that means what it says: more mental stimulation leads to better mental preservation, while a lack of stimulation results in the opposite—a faster rate of deterioration.

And unfortunately for anyone looking to spend their last decades feet up in a Barcalounger, the evidence seems to bear it out.

For example, in a cross-sectional analysis of cognitive function among older men in the United States, England, and 11 other European countries, Susann Rohwedder and Robert J. Willis found that nations with an earlier retirement age saw the most rapid declines in fluid intelligence.

The relative performance difference between men in their 60s and those in their 50s was about twice as great in countries with an early retirement age. In other words, performance on the test nosedived for men who retired earlier. In general, retirement was associated with a memory score nearly five points lower (on a 20-point scale) than remaining at work, representing nearly 1.5 times the standard deviation for the sample as a whole.

The authors summarized their analysis rather unequivocally: “On average, retirement causes a decrease in a person’s cognitive ability relative to staying in the labor force.”

Of course, teasing out causes and effects in a situation like retirement is far from simple. Some people retire because of cognitive declines, so there is necessarily a chicken and egg problem underlying the question. However, while an individual’s decision to retire might be affected by personal conditions, it is also heavily influenced by national policy. For example, the commonplace retirement age in the United States is 67, which is a function not of personal preferences but of Social Security policy.

Taking advantage of the sometimes striking policy differences between countries, Rohwedder and Willis were able to examine the effects of different retirement ages on a broad pool of retirees. Rohwedder explained that those policy differences allowed her and her co-author to isolate the causal effect that retirement has on the mind.

“That’s why we can interpret our results as causal estimates of the effect of retirement on cognition,” Rohwedder said.

So is postponing retirement—or at least keeping intellectually active—enough to stave off cognitive decline?

“It seems that staying active in retirement is an important part of maintaining one’s cognitive abilities… [though] we cannot yet say which aspects of the work process are the strongest contributors to staving off cognitive decline and how best to carry some of those over into retirement,” said Rohwedder.

There are a number of theories about what is needed to preserve mental acuity in one’s older years (more on that later). But there is one important lifestyle factor that arises again and again in the literature on cognitive health and aging, and no discussion of cognitive preservation would be complete without it.

That factor is education.

Education and the Echoes of the Past

Of course, education is, in itself, a complicated variable. It interacts with various lifestyle decisions, health and social behaviors, and even labor force participation. In other words, one’s education level might have an effect on variables like what to eat and how much to exercise; but then again, it may not. Educational attainment is also itself influenced by numerous personal characteristics, such as childhood intelligence or socioeconomic background. This makes it yet another complicated knot to untangle.

In an approach similar to that of Rohwedder and Willis, a Demography study used changes to the length of mandatory schooling in several European countries as a way to tease out the effects of educational attainment. As with differences in retirement age policy, legal changes around schooling laid the groundwork necessary for a natural experiment free from the confounding factor of individual differences.

The effects were noticeable. One extra year of schooling increased memory scores four decades later by about 0.26 (on a scale of 10), or about 15% of standard deviation, and similar effects were found for verbal fluency. The study supports a body of literature that underscores the significance of education to cognitive development, and it demonstrates the far-reaching implications that education can have over the course of our lives.

So, as with those other previously noted luxuries, “Education may be wasted on the young,” writes Jonathan W. King and Richard Suzman in an editorial for Psychological Science in the Public Interest. They explain, “early exposure to education and enriched environments, rather than late exposure… might have the most pronounced effects on both cognitive and non-cognitive skills in old age.” With fluid intelligence peaking around age 20, the implication is that the period leading up to early adulthood is the most vital for ensuring a slower long-term decline in cognitive capacity.

But that doesn’t mean our cognitive fates are irreversibly sealed by our second decade. Far from it.

Redirecting Cognitive Destiny

Education is important, but it isn’t destiny. Challenging the mind over time can enhance or reduce the benefits to schooling, which in turn implies that the health of your mind is, to an important extent, still in your hands.

For example, one curious aspect of education is that having too much of it can, in some circumstances, reduce its benefits. In a study on education, job level, and cognitive ability, a longitudinal survey in the Netherlands discovered that over- or under-education for one’s job can be surprisingly deleterious.

The researchers found “significant negative effects of the extent of overeducation on cognitive decline” in immediate and delayed recall and in verbal fluency. The more overeducated someone was for their position, the worse they performed on cognitive tests over time.

Interestingly enough, the opposite held true for those who were undereducated. The researchers found that these workers experienced less cognitive decline in delayed recall and cognitive flexibility relative to equally educated peers in more “typical” positions, though results for the latter were only weakly significant. Asked if this means there is more to the cognition story than simply “using it or losing it,” co-author Andries de Grip affirmed that having a challenging job above one’s education level seems to be the key ingredient.

“It’s important to remain [employed] in jobs at an appropriate level,” he said. The level being a cut above what one’s qualifications would imply.

Use it or Lose it — As Much as You Can

Thus, on balance, use it or lose it—or most importantly, challenge it—is an idea that carries merit, and, as Christopher Hertzog and his colleagues suggest, it falls under the umbrella of “cognitive enrichment.” Still, other factors can’t be discounted. There are countless variables that may influence one’s level of cognitive health at any given time, and these might be just as important as working or education. Some (like genetics) we don’t control, but others (such as smoking), we do.

Enriching your own cognitive environment could involve any number of these myriad factors, some of which may be contributing to or driving the demonstrated benefits of work. Social connections, exercise, and what we generally refer to as “a healthy lifestyle” are all examples.

In the end, whether or not you remain gainfully employed past the typical retirement age, the existing research has meaningful implications for how you spend retirement. It seems the last thing we should be looking for in our golden years is countless reruns of The Golden Girls; however, it seems that our cognitive health could benefit from close friends (or better yet, co-workers) just like them.


JSTOR is a digital library for scholars, researchers, and students. JSTOR Daily readers can access the original research behind our articles for free on JSTOR.

The Journal of Economic Perspectives, Vol. 24, No. 1 (Winter 2010), pp. 119-138
American Economic Association
Demography, Vol. 51, No. 2, April 2014
Springer on behalf of the Population Association of America
Vienna Yearbook of Population Research, Vol. 10, Education and the Global Fertility Transition (2012), pp. 95-126
Austrian Academy of Sciences Press
Psychological Science in the Public Interest, Vol. 9, No. 1 (October 2008), pp. i-iii
Sage Publications, Inc. on behalf of the Association for Psychological Science
Oxford Economic Papers, New Series, Vol. 60, No. 2 (Apr., 2008), pp. 237-253
Oxford University Press
Psychological Science in the Public Interest, Vol. 9, No. 1 (October 2008), pp. 1-65
Sage Publications, Inc. on behalf of the Association for Psychological Science