Hacker News new | past | comments | ask | show | jobs | submit login
Changing gut bacteria through diet affects brain function (newsroom.ucla.edu)
54 points by georgecmu on June 16, 2013 | hide | past | favorite | 24 comments



Only 36 people in 3 groups of 12. Changes visible in fMRI but not (from the writeup) obviously in any direction of better/worse. Study funded by Danone yoghurt company.

Diet surely affects cognition and microbiome, and microbiome also affects health and cognition. But this is some pretty thin support for anything interesting.


It seemed they noticed people who had healthier gut (i.e took probiotics) had less activity in the part of the brain that processes internal sensations.... right, so stomach aches hurt, and not having them doesn't. *

* I think I'm being a bit too cynical...


Where did you read that they had observed a reduction of abdominal sensations in the probiotics group?

It is not mentioned in the press release, but I would be surprised if they had not assessed this.


On the other hand, the data is presented in a pretty objective manner, including the size of the sample ("small study"), and the connections to industry are correctly highlighted. Definitely not a conclusive study, but interesting nonetheless.


The differences in connectivity between the periaqueductal grey and the cortex are interesting.

The periaqueductal grey is a key nucleus in the pain pathways. The probiotics group shows an increased connectivity with the prefrontal cortex, wich suggests an improved ability to control and reduce pain perception.

The rest of the description is too vague for me to interpret.


The skepticism shown in other comments is healthy but IMO misinformed and/or misguided.

This study confirms past results such as [0], which shows that adding some Lactobacillus rhamnosus to the diet of mice modifies the expression of GABA receptors (the ones sensitive to bnezodiazepines, like Xanax), and reduces their anxiety (which can be assessed reliably). The effect is mediated by the vagus nerve. If the nerve is cut, the effect disappears.

In the same vein, [1] found that the pulmonary exposure to Mycobacterium vaccae promotes the growth of serotonin and norepinephin neurons in the brain of mice and makes them "less resigned" [2].

I'm sure there are plenty of similar results.

The sample is relatively small, but not inappropriate for a fMRI study. There are statistical tools adapted for these situations like SPM [3], provided they are used properly (and I wouldn't be able to assess whether or not they were).

-

[0] http://www.pnas.org/content/108/38/16050

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1868963/

[2] http://en.wikipedia.org/wiki/Behavioural_despair_test

[3] http://www.fil.ion.ucl.ac.uk/spm/


That's really interesting, thanks for sharing.

As an aside, why did you choose to use cardinal numbering rather than what I certainly would have reached for first (pun intended)?

Back on point, I can't see anywhere in [1] that 5-ht neurons grow, rather they just seem to be more active. Fascinating that this happens as a response to increased antigen load to something they have been immunised against. We really are a product of so many hidden interactions with our environment


Formally, ordinals start at zero too :-) [0]. We owe their latest definition to Von Neumann, but AFAIK, the former definitions were similar.

Regarding 5-ht, I quoted from memory, which was apparently fuzzy... But is is indeed fascinating.

--

[0] This is meta-meta-contrarianism.

- The layman counts from 1

- The uptight programmer counts from zero, because Dijkstra said so (or so he thinks).

- The meta-contrarian (I used to be one) says fuck it, ordinals start at one.

- The meta-meta-contrarian reads Wikipedia[1], realizes he was formally wrong, and goes one step further in pedanticity, back to zero [0].

That being said, my brain prefers 1-indexing programming languages like Lua, Julia, R and Matlab...

[1] http://en.wikipedia.org/wiki/Ordinal_number

[0] Help! I'm stuck in a Boolean algebra[1][0]!

[1] http://en.wikipedia.org/wiki/Boolean_algebras_canonically_de...

[0] ... where xor is an addition that wraps around ...


Thanks for the entertainment!


You're welcome!


> The women who ate no product at all, on the other hand, showed greater connectivity of the periaqueductal grey to emotion- and sensation-related regions, while the group consuming the non-probiotic dairy product showed results in between.

So basically the probiotics had nothing to do with it, and it was actually the dairy product? Or something else entirely? Lactic acid? Who knows.

If it was the probiotics why would consuming the non-probiotic have any effect whatsoever? The fact that it did shows that this study is not studying what it thinks it is studying.

Props for including a good set of control groups, but un-props for the interpretation.


> So basically the probiotics had nothing to do with it, and it was actually the dairy product?

No. The dairy had some effect in one of the tests (but not on the others), and the probiotics added to that.

That's the very point of having a dairy control group: exclude that the effect is solely due to taking dairy.

Had the results been equivalent in both groups, your conclusion would be correct.


> So basically the probiotics had nothing to do with it, and it was actually the dairy product? Or something else entirely? Lactic acid? Who knows.

Lactic acid as in "lactic acid that builds up in muscles and makes one's legs hurt when running fast for a long period of time"


Or lactic acid, as in what is found in dairy products. Though that would make saying dairy product and lactic acid sort of redundant (outside of being more specific). Same lactic acid though in either case, just how it manifests itself in the body.

https://en.wikipedia.org/wiki/Lactic_acid


The NSA ate my question mark :]

Thanks for clarification, I am now surfing Wikipedia on a sunny Sunday morning.


And note especially this: https://en.wikipedia.org/wiki/Lactic_acid#Brain_metabolism

Which is quite relevant to this study.


As an aside, like most of the health information we share, the idea of lactic acid building up and causing soreness is a myth. (Lactic acid levels return to baseline an hour after exercise)

http://en.wikipedia.org/wiki/Delayed_onset_muscle_soreness


If I had to guess, this could be related to our internal feedback loops mechanisms. Many emotions create peculiar sensations in our guts - that is an experience we all have - and those sensations are nothing else than signals to the brain.

It isn't a stretch of the imagination to think that the status of our guts could amplify or dampen the sensations, and thus the feedback to our brain of our emotions, especially the feedback to our conscious mind: If I remember correctly from my studies in psychology, we actually feel our own emotions mostly as physical reactions.


Has anything ever been shown to not affect brain function.


Consider x a distance between a place where an event took place at time t1 and a brain. At time t2, if x is larger than c * (t2 - t1), it for sure does not affect the brain.


Sure. Except then you need to define c. Back to square one.


c is commonly used to represent the speed of light[1]. The commenter was saying that an action sufficiently far away cannot affect you immediately, because the information of that action's occurrence cannot move faster than light. I believe this is generally regarded as true by physicists, although there are some ideas for circumventing this limitation.

I think they were making a joke.

[1] http://en.wikipedia.org/wiki/Speed_of_light


Thanks. :-)

(I thought "speed of light" at first, but didn't think of a way it could be applicable.)


c is speed of light?




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: