Assuming observations are statistically sound...

Explanation 1: being overweight leads to brain deficiency.

Explanation 2: brain deficiency leads to being overweight.

Explanation 3: a hidden variable, such as excessive blood sugar or nutrient deficiency, lead to both.

Explanation 4: there is no brain deficiency since there is no cognitive impact.

The article states as much, which is very refreshing to see in a mainstream paper.

Explanation 5: People who are in shape are more likely to be active and exercise regularly, which have very noticeable benefits on the brain.

That would be a subset of Explanation 3.

>a hidden variable, such as excessive blood sugar

Well we do know that eating a lot of sugar increases insulin production and insulin enlarges cells. I am not sure about the whole causation/correlation part of it, but for example with increased insulin production, fat cells become larger and their will be a greater likelihood of obesity even in skinny people. Moreover, I understand it is common for people with high insulin levels to have enlarged organs.

It is an outrageous theory, but maybe the difference in the brain is simply cell size due to insulin since that might affect appearance but not function.

In reading the paper, I appreciate the statistical efforts to adjust for age differences between the groups to make their claim, but at least optically it doesn't help that the average age of each cohort was already 10 years different. Lean group M = 48 yrs SD = 16; Overweight M = 57 years SD = 17; Obese M = 61 years SD = 16. The latter two groups were combined for analysis.

Is it still contentious to use the BMI to define these groups? http://www.npr.org/templates/story/story.php?storyId=1062684...

Doesn't BMI that uses electrical current to check for fat percentage in body negate the conetentiousness of BMI? I ask because I thought that was the case but now I'm not so sure.

There is no BMI measurement that use's electrical current. That's a body fat index which is very different and much more precise. BMI is a very poor and ultimately very hard to justify formula that only takes into account weight/height sorted by gender. Things like muscle index and body frame are completely ignored and often lead to very poor BMI's for people who shouldn't have them.

Maybe bargl means Bioelectric Impedance Analysis? A cursory internet search also indicates BIA is suspect as a one-time measure.

OK so I have this and I thought that the BMI it computed was based on tracking the fat to everything else based on BIA. But I may be way off here.

https://www.fitbit.com/aria

I see that Fitbit tracks this as BodyFat % not BMI. Not sure if that's accurate.

BIA is pretty bad. Here are ~1300 Fitbit Aria measurements matched against 7 DXA scans. The Fitbit bodyfat data (teal) is just random noise and can barely detect considerable changes in body composition.

http://i.imgur.com/J4Ls2bQ.png

I stopped paying attention to be BMI, I'm 6ft, 190lbs and have a 32" waist and it has me right on the edge of been overweight.

I think given that many people don't realise how inaccurate it is I might even class it as harmful.

You are on the edge of overweight, and you don't realize it because of all the fat people around you.

If you look at previous research on exercise, inflammation, and brain volume:

http://dx.doi.org/10.1016/j.neuroscience.2014.05.005

and the mountain of evidence showing that excess visceral fat tissue produces greater levels of chronic inflammation:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3437966/

Then it seems that inflammation would be the first place to start looking for mechanisms linking weight with white matter volume.

Certainly more visceral fat is associated with higher risk of all of the common forms of dementia, and most age-related conditions are sped up in their progression by inflammation.

It does not seem to affect cognitive ability (per the article). However I am curious if it is reversible. If you lose weight and get back to 'normal', does the brain layout look normal [for that given age group]?

> It does not seem to affect cognitive ability

Is that the case long term though? It would be interesting to study the progression of people over the next decade or few to see if it seems to speed up the onset of age related dementia or increase susceptibility to other conditions and illnesses.

Also what about long term affects? Maybe it doesn't affect my cognitive ability now, but will it make me more susceptible to dementia?

I'd be interested in seeing if the effect is continuous or happens more at certain ages. Like take a group of people who are obese and say in there 40s, half were obese since they were children, and half became obese after 25 or so. See if there is any difference in brain development if the person was obese while they were still developing.

Not sure how you would account for effects of just being obese longer in general tho.

Given that most research findings are false (see http://journals.plos.org/plosmedicine/article?id=10.1371/jou... for confirmation), I'm going to need to see replication of this study before I believe it one way or the other.

There is more and more agreement that obesity, diabetes, hypertension, heart disease and dementia are all in the end manifestations of various forms of metabolic issues, most prevalent of which is (frequently undiagnosed) insulin resistance, which, in turn is the result of high sugar consumption coupled with inflammation (from consuming all kinds of strange substances like food coloring and artificial sweeteners).

Our bodies are not very well adapted for continously running on sugar, but ever since early 19th century when sugar became abundant, that's what we've been doing, especially the developed countries.

>coupled with inflammation (from consuming all kinds of strange substances like food coloring and artificial sweeteners)

Citation?

> > coupled with inflammation (from consuming all kinds of strange substances like food coloring and artificial sweeteners)

> Citation?

Immune reactivity to food coloring: http://www.ncbi.nlm.nih.gov/pubmed/25599186

>There is more and more agreement that obesity, diabetes, hypertension, heart disease and dementia are all in the end manifestations of various forms of metabolic issues, most prevalent of which is (frequently undiagnosed) insulin resistance

Citation needed.

> Citation needed.

You are confusing HN with wikipedia. ;)

On a serious note - I purposely avoided including links because this is such a big subject (and neglected to mention that in my post). Just google for it, draw your own conclusions.

I have. My conclusion is that your assertion is incorrect. That is why I asked for a citation.

Edit in Reply:

I would like you to cite gold-standard scientifically verified evidence, not "common knowledge." As far as I know, insulin resistance is not known to cause the myriad of diseases you mention. It is merely an early warning sign of diabetes.

If you wish to challenge the text books, you better come with peer-reviewed and independently verified studies.

Also, you're getting the scientific method backwards. If you assert a fact, you're the one who has to prove it. The burden of proof lies on you to demonstrate that the current medical evidence is wrong.

As said two thousand years ago, "The proof lies upon him who affirms, not upon him who denies; since, by the nature of things, he who denies a fact cannot produce any proof." [0]

[0] Just in Case - http://www.revistaamicac.com/Insulin%20Resistance%20-%20What...

-----------

Huh. You're right. Thanks for the sources.

It's not definitive research, but it's certainly very convincing. Doctors often act on less, including single patient case studies.

> As far as I know, insulin resistance is not known to cause the myriad of diseases you mention.

IR and cardiovascular: https://www.jci.org/articles/view/10762

IR and dementia: http://www.ncbi.nlm.nih.gov/pubmed/12467491

IR and obesity: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4038351/

IR and hypertension: http://www.ncbi.nlm.nih.gov/pubmed/7512468

That quote is regard to where burden of proof lies, not whether there is any burden.

> My conclusion is that your assertion is incorrect.

Citation needed. ;)

But seriously - what would you like me to cite? That we are consuming incredible amounts of sugar or food coloring? Or that high sugar consumption leads to type 2 diabetes, which usually starts with un-daignosed insulin resistance? Or that the world population is consuming more sugar in the last two centuries? Or that there is a link between metabolism and obesity, or between obesity and all aforementioned symptoms and diseases?

It's all mostly common knowledge really...

Sugar consumption is probably not all that different from tobacco smoking - I wouldn't be surprised if some 10 years from now we will see lawsuits against sugared drink vendors claiming that they didn't adequately warn their consumers of the dangers of sugar consumption.

Citations are still common in HN. You see them in bracketed form [1], [2], [3], etc. all the time. If you can't be bothered to provide even a couple of sources, I wouldn't expect people to invest any time into "Just googling it" (which sounds a bit condescending).

> Scans show greater shrinkage in volume of white matter in overweight and obese people, although it does not appear to affect cognitive performance.

Perhaps this is a demonstration that the brain is over-engineered and the extra matter is insurance against damage. If the white matter is being lost slowly, the brain can continuously adapt asymptomatically.

There may come a point where there is so much loss/damage dementia is inevitable, implying obese people would have a higher long-term likelihood of dementia.

Edit: removed an off topic counter-reference.

Hmmm, I am so brainy I can waste a few brain cells on second helpings and still get into Mensa.

I always knew it ;-)

I went to the client subsidised canteen for lunch with my team today. I selected some nice salad from the buffet, and decided to weigh it before heading for the till. At today's conversion rates it was 12 bucks for green leaves and Greek salad.

I put it down and had a very nice pie and chips for half that.

We have a situation where a significant portion of our society is having to make choices at the margin that it is evolutionarily unable to do.

I think we have many challenges ahead of us before we reach that golden future of humanity. wars, climate change, to name two, but they come down to the same problem we have with obesity - arranging our society to ensure our inner chimp's less helpful evolutionary urges do the least damage.

> Brains of overweight people look ten years older

So, wiser?

More likely closer to dementia, alzheimer's disease, etc.

Ironically, this is linked to from the side of the article: https://www.theguardian.com/society/2015/apr/10/underweight-...

Age doesn't always correlate to wisdom.

And it correlates even less to people whose initial reaction to any discussion of age is "BUT WISDOM"

Well imagine the brain loses out over nutrients/resources in an obese person more so than a smaller person...

Another question is why does "society" honor and accept Obesity as a lifestyle choice, when its openly critical of tobacco addicts and heroin addicts? What's the difference?

It's widely seen as less of a simple choice to become obese, as it is to take up smoking for one thing. Presumably if you met someone who'd been regularly given cigarettes since they were toddlers, you would have a certain degree of sympathy for their issues as adults. There is also the simple fact that abstinence from tobacco is simple, and easy, and it works for everyone who doesn't choose to smoke. Most people don't attempt to have a "moderate" intake of smoking after all.

Meanwhile food is a bit harder to manage if you have a problem with it, don't you think? So, for people talking about a "lifestyle" ignore the crazy people. For people talking about realistic challenges beyond "suck it up fatty", you might want to pay attention.

Finally, "honor"? I don't think you have widespread discriminatory hiring practices against the "honored".

It doesn't, but at least in the US it might seem so because society makes it as hard as possible to get hooked on heroin and tobacco (prohibition, restriction & taxes) and easy as possible to get off of it (rehabilitation programs), whereas the ingredients that cause obesity (e.g. corn) are actively subsidized by the government and promoted by a large portion of the medical establishment. I'd bet that if the US government were actively subsidizing tobacco and opium farmers and processors, addiction to those drugs would be seen much more as a lifestyle choice.

>whereas the ingredients that cause obesity (e.g. corn)

Pardon my ignorance, but isn't corn just a vegetable? Are we talking about the same corn (sweetcorn) or something else?

My intuition tells me that such a vegetable can't be "bad" for you, though that's probably because I eat it as part of my "vegetable" section of my diet.

Is this vegetable really in the same ranks as cane sugar, salt, sweeteners etc.?

Unprocessed corn isn't exactly bad for you--yes, it is a cereal and thus has large quantities of carbohydrates, but it has plenty of fat, proteins and fiber as well. It's when the corn gets processed, and the fat, proteins, and fiber get removed that it starts being unhealthy. In the end it's the American preference for sweet, bland foods that causes most of these problems.

The US uses lots of HFCS - high fructose corn syrup. Partly this is because government subsidies make it a very cheap ingredient. Some people think HFCS is particularly bad, worse than other sugars.

I don't think the US weight loss market would be so large if society honored and accepted obesity ($64 billion in 2014 was one number I saw). There's certainly people who push for acceptance, but I've never gotten the feeling that it's the overwhelming sentiment.

It doesn't, but a vocal community are pushing this idea.

So we're going by looks now?

A ten-year-old can be obese. An 80-year-old can be obese. I'm not sure what it says about the corresponding "lean peer" when adding ten years to these ages.

The downvoting at Hacker News is often amusing.