DYNOMIGHT ABOUT RSS SUBSTACK

Grug on diet soda and autism

Grug on diet soda and autism

Oct 2023

grug try to not yell about bads in science news too much because why make same yells over and over?

and grug have no new learns, just feel people maybe sometimes not use old learns

and grug family often plead to someday have conversation about alternate topic

but rare for one news to make all bads at once so grug think maybe useful to apply old learns together in one place

originally try to write in english but very boring so eventually manifest grug mode

so today grug yell about recent diet soda and autism news from fox news and hill and new york post and toronto sun and others

grug organize big yell into four sub-yells

1. Some fact paper not mention

each news come from same press release for same paper which grug try to read

in introduction see words:

Each of the three first-phase metabolites of aspartame has been studied with regard to its impacts on neurologic function. Aspartame is metabolized in the intestine into aspartic acid, an excitatory neurotransmitter; phenylalanine, which is involved in neurotransmitter regulation; and methanol, the metabolites of which include formaldehyde, formate, and other toxins. The adverse neurological impacts following the consumption of aspartic acid, phenylalanine, and/or aspartame include changes in neurotransmitter levels and excitotoxicity, with adverse impacts on neuron function/survival. Primates, including humans, are uniquely vulnerable to methanol, the blood levels of which rise following aspartame consumption. Exposures to methanol and formaldehyde resulted in increased neuronal apoptosis, neurodegeneration, and cognitive problems.

words basically true! or at least grug not want to argue

but grug feel weird to not mention:

that most protein in food have some phenylalanine and aspartic acid

and that eat one day normal food give 100x more phenylalanine and aspartic acid than drink one diet coke

and that if somehow not eat any phenylalanine, then die

and that cell walls of all fruits and vegetables have pectin which human guts convert into methanol

and that eat one day normal food give 10-100x more methanol than drink one diet coke

2. Big statistics learn

so grug try to read paper try hard many words

Offspring born in 1984 or thereafter, with complete early-life DS exposure data and neurodevelopmental diagnostic data, were included in analyses. The proportions of off- spring with DSearly and ASPearly exposures were separately calculated, by sex, for the following diagnostic groups: controls, all ASD cases, ASD cases without autism, autism cases, and the non-regressive version of each case definition. Differences in the proportions of exposed controls and cases, in each category, were calculated using the Pearson χ2 statistic in Stata/IC 14.2 for Windows.

in case you not read, grug try to summarize actual paper activities

  • find 235 autism human born after 1983 when aspartame approved

  • also find 121 neurotypical human

  • ask mom of each human “While you were pregnant or breastfeeding your child, how often did you drink diet drinks containing artificial sweeteners?”

  • take data (mom drink diet soda during pregnant) and (find human on internet or local media) and (birth year) and (ethnicity) and (mom learning level) and (family make ≥ $100k per year) and sometimes (sex) and do regression on output of (human has autism)

  • regression coefficient for first variable is positive, noooooo!

grug confused why paper make “survey plus regression” so complicated but never mind

on internet, many big brain see paper and make fancy yells about multiple hypothesis correction or confidence interval size or no preregister analysis or reverse effect in girls or no control for autism of parent or drink or smoke

grug not really disagree and feel internet big brain heart good but still not really support fancy yells

because grug do thought experiment

what if survey have 356,000 human instead of 356 and tiny confidence interval and p=.0000013 would grug believe paper prove diet soda make autism?

no

so grug feel these yells not reach core issue:

correlation bad

apex predator of science is correlation

grug sure mega brains already give you big learn about correlation and causation, and if you only have one statistics learn you have this one

so why not use big learn?

if researchers take moms and ask half drink more diet coke and those mom have more autism child then grug very worried even if confidence interval big

if confidence interval small and study replicated, grug immediately retract all aspartame yells and probably reconsider whole life

until then grug say: correlation very easy even if aspartame no make autism for example if autism human drink more diet coke then data make big correlation just because autism have large genetic component

if do similiar study and ask if mom drink chai, grug sure child skin color correlated because more south asian mom enjoy chai

FDA have many mega brains

if ask FDA to approve drug because correlation FDA laugh at you many humiliation

if ask FDA to recall drug because correlation FDA also laugh very hard

grug sometimes yell about FDA but FDA better than 95% of news because FDA always use most important learn

3. Claim cascades

grug notice something else

paper not want rejected so careful words:

While our findings do not establish a causal relationship between daily early-life exposure to diet sodas/aspartame and autism risk in males, they nonetheless raise concerns that justify further research, especially given the current widespread use of diet products among pregnant women.

but for press release, no annoying reviewers!

“Maternal consumption of these products during periods of heightened offspring vulnerability represents a modifiable potential risk factor, the elimination of which might help to protect susceptible offspring in the next generation.”

and news company need readers for make more shiney rocks! so news company say:

‘Major warning flag’ raised about chemical that moms consume when pregnant or breastfeeding

and then internet human scared by scary science news, so internet human say:

aspartame in diet soda apparently triples the risk of having an autistic son.

grug not want beat up internet friend, reaction understandable because news have extra bads everyone think in paper even though not

sometimes big brains yell “always read paper!”

grug not make this yell

because understand few read paper, instead spend time feed little human or look for partner for make little human or make shiney rock or stare at favorite glowing rectangles

grug also understand paper always have many big word even when not needed, punish human reader

so grug say OK, no read paper! but remember not trust news too much

4. Secret bads

why so many bads in news?

grug think like this

every day many big brain write many science but few become news, how chosen?

answer is journalist need Big Science for make everyone stare at glowing rectangles and get shiney rocks

but much easier to make Big Science with bads than by find new big learns

and journalist job not to read science or find bads but mostly to quote big brains

but call big brains take time, decrease shiney rock intake, so often just pretend to call big brains using quotes from press release

and if do call big brain and learn science bad, journalist often not make news at all

and grug tell you secret

big brains not like to tell journalist about bads

grug observe big brains in same micro-niche do read paper and do see bads and gleefully talk about bads at lunch but when write or talk to journalist very gentle unless old and famous and not care anymore

because high cost for point out bads: bad maker remember and maybe later reject paper or not cite paper or refuse to give shiney rocks or say bads in tenure letter or just have big awkward at conference

each big brain micro-niche is tiny gossipy village, unbelievable

so grug advise to look look hard for quotes from other big brains, if say bads then probably many other big brain agree and bads even worse than stated

for diet soda autism news grug find quote

Dr Deirdre Tobias, a nutritionist at Harvard University who was not involved in the research, told DailyMail.com it was ‘shocking that the authors would feel confident enough in this design to draw those conclusions.’

[…]

Dr Tobias added that the three artificial sweeteners they examined are ’completely different compounds, metabolized very differently in humans, and have been extensively evaluated for their safety. 

‘Thus, the fact that the signal for aspartame was essentially the same as the other chemicals further points to the bias in this study, perhaps due to errors in the mothers’ recall or other factors related to women who chose diet sodas.

‘It is extremely unlikely that any association they are observing has anything to do with the chemical aspartame itself.’

grug salute brave big brain

maybe some people think mean

but grug say not mean: purpose of big brain is find new learns for normal human, if bads not corrected then mean to normal human!

Bonus yell

grug not blame anyone too much

blame authors some for bads but grug also make many bads many times so expect there will always be bads somewhere

and not blame journalist too much because news economics brutal and journalist starving

and not blame human for interest in Big Science

so grug sad system promote bads but each agent understandable so system persist long time

but recently some big brains make big math “community notes” feature to show bads for news and already work better than grug think possible

so grug optimistic that system evolve and bads not promoted so much in future

maybe this work so well that grug successfully contained henceforth, but probably not

Comments at substack.

new dynomight every thursday
except when not

(or try substack or rss) ×
Fahren-height

(celsi-pour?)

The Internet is well into middle-age, and yet doesn’t seem to have answered humanity’s most pressing question: If you pour boiling hot water from various heights, how much does it cool in flight?

Fancy math doesn't make simple math stop being true

on butts and instrumental variables

What are you supposed to do when someone disagrees with you using a bunch of math you can’t understand? I’ve been thinking about that recently because of the NordICC colonoscopy trial. It took 85k Europeans aged 55-64, invited a third...

Are language models good at making predictions?

politics more than science

To get a crude answer to this question, we took 5000 questions from Manifold markets that were resolved after GPT-4’s current knowledge cutoff of Jan 1, 2022. We gave the text of each of them to GPT-4, along with these...

My stupid noise journey

A tale of bad choices

Interested in how to be a big dumb idiot and over-complicate things and waste time and money and endure tons of stress and some real physical pain all by thinking that you’re cleverer than you actually are? (No?) Looking back,...

The second system problem

Building a safe AI ≠ preventing all unsafe AI

In The Vulnerable World Hypothesis, Nick Bostrom imagines we found a technological "black ball"—say a way to make a nuclear weapon with just some glass, some metal, and a battery. He concludes that society in our current "semi-archic default condition"—could...

I still think it's very unlikely we're observing alien aircraft

They'd have to be messing with us.

Some suggest there might be alien aircraft on Earth now. The argument goes something like this: A priori, there’s no reason there shouldn’t be alien aircraft. Earth is 4.54 billion years old, but the universe is 13.7 billion years old,...

Why didn't we get GPT-2 in 2005?

We probably could have

The ancient Romans were never great at building ships and never tried to explore the Atlantic. The basic reason seems to be—why bother? The open ocean has no resources and is a vast plane of death. But imagine that in...

First-principles on AI scaling

How likely are we to hit a barrier?

It's hard not to feel blinkered by recent AI progress. Every week there seems to be an AMAZING NEW SYSTEM with UNPRECEDENTED CAPABILITIES. It's impossible not to wonder what the future holds. Until recently, I thought progress was so dizzying...

Your friend the language model

The world is running out of khakis

I originally wrote this as part of a much longer post on LLM scaling laws and possible barriers/trajectories for progress. The idea was to provide the minimal background necessary to understand all that stuff. But in retrospect, many people probably...

Winner take all science

Is it helpful for things to work this way?

By the early 1950s, it was known thanks to people like Miescher, Levene, and Chargaff that genes were carried by long polymers in the cell nucleus. It was also known that those polymers had a sugar-phosphate backbone and were composed...