What a bunch of dieting monkeys teaches us about some human scientists

The notion of caloric restriction (CR) as a means to prolong human life is intriguing.

The basic idea of CR is that if you eat a lot less than the average person you’ll live longer. You might even live a higher quality, healthier life.

But is this true?

A whole bunch of scientists and even more dieting monkeys have been part of a very large, very expensive (as in millions of dollars; see image below from the National Institute on Aging) research effort to answer this question. A paper in 2009 on monkeys suggested CR worked.

But yesterday we heard that the latest study on the monkeys pretty conclusively indicated that at least in monkeys, which are very similar to humans, CR to put it simply failed to work. It did not prolong life.Caloric Restriction

The best write up I’ve seen on this is from one of my favorite science writers, Gina Kolata, over at the NY Times. I highly recommend it.

Perhaps the “good news” coming out of this study is that humans can enjoy eating in moderation and do not have to adapt some extreme CR approach to life that leaves them forever hungry.

What I find most fascinating coming out of this arena has nothing to do with CR itself, but rather I think this bunch of dieting monkeys has more to teach us about human scientists than anything else.

The CR studies reveal some sobering realities about the traps that we human scientists can fall into if we are not extremely careful.

Let’s go back 3 years to that first major monkey study, published in Science in 2009 by a team from the University of Wisconsin.

The study concluded that CR prolonged life and it made huge international news. Iit was a bombshell of a story.

However, many people believe the study came to the wrong conclusion, in fact an exaggerated or hyped conclusion.


Gina Kolata hits the nail on the head:

But even that study had a question mark hanging over it. Its authors had disregarded about half of the deaths among the monkeys they studied, saying they were not related to aging. If they had included all of the deaths, there was no extension of life span in the Wisconsin study, either.

In other words, the scientists doing the monkey CR aging study published in 2009 (what Kolata refers to in this passage as “that study”) eliminated more than half of the death data and specifically that data which didn’t fit with their hypothesis that CR lengthened life. As a result, the scientists most likely in my opinion came to the wrong conclusion.

Why would they do this?

In this regard, Kolata quotes Steven Austad of Univ. of Texas in San Antonio in this passage:

the University of Wisconsin study “was not nearly as conclusive as it was made out to be” and that the new study casts further doubt on the belief that caloric restriction extends life.

There is a potential trap in science of becoming married to a hypothesis, becoming emotionally invested in it. Another way of putting it is that some scientists sometimes make the mistake of deciding in advance what their studies should prove rather than letting the data itself tell the story.

I think this has happened in the CR research field to some scientists and that 2009 study is a telling example. Others in the CR field have remained more objective.

Part of the reason I think some scientists are too emotionally attached to CR is that they themselves started CR on themselves too.

So amazingly, some of the scientists studying CR in monkeys, also in essence started doing concurrently personal experiments on themselves also with CR. At some point one has to wonder if they lost their objectiveness.

In the cancer field it would be akin to scientists who deeply believe that a certain new drug, Drug X, prevents cancer and who are studying Drug X in rodents predisposed to get cancer, also starting taking Drug X themselves at the same time they were studying Drug X in rodents. When put that way, to me it seems particularly concerning.

Kolata digs deeper and finds that some earlier rodent students on CR, which the CR scientists had so often referenced, perhaps were also showing CR had no effect on lifespan. When the mouse studies showed no benefit of CR on lifespan (or even shortened lifespan in some mice), some of the CR scientists tried to undermine those studies as Kolata describes:

The response to that study was “absolute disbelief,” Dr. Austad said. “Even though the authors are well-respected calorie restrictors, people said the result was not interesting, that there was something weird about the mice.”

The weight of the evidence suggests that some groups in the CR field had become emotionally attached to CR and lost some of their objectivity. When the data didn’t fit with their hypothesis that CR extended life they ignored it.

It’s pretty clear that quite a number of CR scientists have become too attached to CR and it has changed the way they do science.

One CR scientist even described himself as “a hopeless caloric-restriction romantic”.

The story on CR and lifespan is not resolved and perhaps there is some subtle effect of CR on lifespan and aging, but if there is an effect of CR it is unlikely to be major for most primates including humans. We’ll see.

However, what is clear from the CR studies is that scientists are just as human as any one else and sometimes fall into the trap of wanting their hypothesis to be true so badly that they veer off course from rigorous scientific method.


  1. Thanks for this follow-up. I had been pretty skeptical of the original CR finding based on A) the formidable complexity of normal metabolism, B) findings that “organ reserve” in the form of plentiful muscle weight benefits people who fall seriously ill and C) the difficulties observed in trying to get reliable results from resveratrol (which is supposed to mimic CR). It seemed to be just too easy and pat of a story compared to the daunting complexity of normal metabolism.

    Of course, CR will show a benefit (even in humans) if you compare to chronic *overfeeding* and the diseases that spring from broken metabolism. However, this is a far cry from proving that CR, per se, has a salutary effect. From talking to various Calorie Restrictors, this point seems largely lost on them.

  2. My totally unscientific 2 cents worth is that psycho-emotional health is the most important factor in physical health and longevity.

  3. I totally agree. There are all kinds of scientific fads that sidetrack us. A big one in evolution/sexual selection in the 90s was “fluctuating assymetry.” Supposedly females of dozens of species could detect miniscule variations in symmetry of potential mates, and were picking the most symmetric. There were dozens of papers, my personal favorite being one that was on the professors undergraduate students in a class. Due to unconscious bias in reporting results and the usual dreaded negative-results-aren’t-published syndrome, this went on for years until it was finally clear that the emperor had no clothes.

    The good lesson is that despite scientists being as human as everything else, with their own biases, in the long run science is self-correcting and will get it right. (This is what creationists refuse to acknowledge when they insist that Darwinism is simply another “religion” like theirs.)

    I write on bad science a lot on my own blog (and there is a ton of bad science, unsurprisingly, on any topic that has to do with human sociology or health), and here is more of my take on the topic, using the horrendous vaccination/autism fraud as an example:


  4. Gizmo – not so fast. Any occlusion or rupture of an artery feeding oxygen, nutrients, hormones, etc. to your tissues (in particular the heart or brain) has undoubted potential to end life abruptly and independently of any psycho-emotional influence. Psycho-emotional health doesn’t go very far once the heart stops ticking. That said, I agree that it seems to factor into ‘general’ health.

  5. Interesting article and interesting comments. As Mark points out metabolism is complex, add to these how nutrition (and all the food that contains non nutrients) is metabolized and impacts on the various systems and you amplify the problem.

    There is however protein engineering going on. There may be variants from person to person but function does not happen by accident it is just we have lacked the tools to understand and interpret it at the necessary nuts and bolts level at the systems scale. We now see these tools coming and with it potential proper insights on the ageing process. This seems to involve multiple roles in an accelerating self fulfilling pattern of declining DNA repair, shortening telmoerase, free radical defense systems, mitochondrial function etc. Highly complex and the fixation of studying each part in minute detail is an understandable approach but flawed owing to the realities explained by complex systems science.

    Interesting that CR saw some effect on cancer incidence. This would tie into the rationale that external factors create mutating events on our cells and the systems within them and the wrong combination can trigger age related disease. A sort of sinister crab shoot theory. There are some fascinating articles on this and we have a blog post on the subject.

    What is does show is the crying need to address the realities of the complexities of biology especially with complex disease, of which ageing is arguably one. Some of the tools to do this have newly arrived and offer an exciting new dawn. Proper biomarkers with proper understanding of genetic variations in patients entering a study should result in profound statistical results, negative or positive, rather than poking in the long grass of the data seeing shadows you really want to see for commercial or reputation reasons.

  6. The authors of the 2009 study say the CR was not drastic at all but was akin to moderation .10% less than uncontrolled eating. They were not showing the primates lived longer and their mortality measures were crude at best plus there were confounding variables regarding animals brought into the study later, the MRI measures were inconsistent, genetics were not controlled for BUT it did show that moderation resulted in improved quality of life with less dementia, arthritis, diabetes and malignancies and the primates on the restricted diet moved better and looked better. I would rather be moderate most of the time to sustain quality if not quantity of life. I was interested to see if this would affect offspring too but no data on that. It was an interesting study whose main points were missed by media hype. I agree with pharma above me, nice post and thanks !

  7. Considering the side-by-side photograph of the two monkeys, it does not surprise me that you wouldn’t see a need to provide labels for the CR monkey and Non-CR monkey. It was extremely obvious: One looked much younger. That’s something not necessarily difficult to quantify (i.e. how aged each of these monkeys appear to people), but perhaps difficult to appreciate scientifically. This is just one study and there will be more.

Comments are closed.