Retinal cells that help stabilize our world view | Hacker News Reader

Retinal cells that help stabilize our world view

optometry.berkeley.edu

117 pointsgeox30 comments

People think that the brain is like a micromanager dealing with all parts of the body manually, but it seems like various levels of 'intelligence' in our body is very much decentralised.

Reflexes are part of the high school biology curriculum in my country. https://en.m.wikipedia.org/wiki/Reflex_arc

Since what year?

Not asking sarcastically, just curious when schools started moving forwards with these things. The decentralised nature of thinking is something that's only been presented to me with a biological basis in the past few years. Would love to know who was agead of the curve, how so, and why!

I attended high school in India around 2008, they had a good three chapters on CNS (incl. how reflexive actions can be routed within the spinal cord).

It was part of the German curriculum in the early 2000s.

In the US I learned about CNF in 2006 and a lot of related things such as specialization of different brain areas and the stomach having sort of a brain that affects you emotionally

It’s been in some high school US textbooks at least as far back as the 1990’s and I suspect it’s much older. But the implication of such didn’t get much attention. I vaguely recall one of those little blue box blurbs with a mention of headless chickens being able to run and a diagram of a reflex test.

We learned about this in 6th grade biology in the late 1980’s.

In my country there is a fairly common expression "det sitter i ryggmärgen" literally "it's in the spinal cord" which means that you have practiced something a whole lot and know it by heart. The implication being that the spinal cord can do the task on its own without involving the brain. It's so common that it wouldn't surprise me if 1st grade teachers used it without even reflecting on the literal meaning.

Now whether the spinal cord actually can learn motor skills seems to be a bit of an open question, but it can perform some instinctive tasks like shying away from painful stimuli.

Love that expression.

There’s a book, The Talent Code. One of the central premises is neuroplasticity and more precisely the process of muslin “wrapping” certain neuropathways making them more efficient.

This happens all the time - for connections deemed by the brain as important. While demoting “unimportant” ones.

So yes, the brain undergoes constant change - iterating on continuous improvement.

Intentional practice activates this process.

This, what we call “talent”, more often than not, is the result of intentional practice.

Mialin, not muslin obviously.

I learned about the reflex arc in the early 1980's via Charlie Brown's "Questions and Answers" series.

https://en.wikipedia.org/wiki/Charlie_Brown%27s_Super_Book_o...

It wasn't named, but it was presented as "if you accidentally touched a hot stove, your reflexes would move your hand before you felt the pain because the round trip to your spine is a shorter path." Which thinking about it now, doesn't quite make sense as an explanation because one is a round trip and the other is not, but oh well. There was an accompanying picture showing a round trip to the base of the skull/top of the spine (not the middle of the spine like the picture in wikipedia).

I did not learn about this in school, but it was apparently accepted enough to put into a children's "encyclopedia".

The knee-jerk effect was documented in the 19th century [0]. I was certainly aware of it when I was young (1970s) although I can't remember whether it was explicitly taught at school, either the effect or the underlying cause.

[0] https://en.m.wikipedia.org/wiki/Patellar_reflex#History

What part of that implies the signal does not go all the way to the brain before the reflex happens?

Was learning this in Ontario in the early 2000s.

The humunculean conceptualization of neurology needs to die.

It doesn't add any explanatory power and confuses rather than clarifies. Why not nest humunculae infinitely? No, the brain at all steps, processes information.

The visual system in particular; this was first revealed by the fun and seminal 1958 paper “What the frog’s eye tells the frog’s brain” from an impressive set of authors: https://labs.la.utexas.edu/gilden/files/2016/04/WhatTheFrogs...

Well, yeah - if the spinal cord is severed - body parts below the injury can't be moved any more, but they don't start to rot off.

The science:

Lovely study. Confirms and extends previous work in rabbits and mice to primates. This paper basically says in a rigorous way: We finally found what we knew should be there in primates. Hurray.

This makes a ton of sense because rabbits, mice, and primates are members of the mammalian clade called “superprimates” or more formally the euarchontoglires. Why throw away a perfectly good set of cells and a bunch of brain regions that all ither mammals have?

Rodents and rabbits are evolutionarily more closely related to humans than most other mammals—-dogs, cats, bats, cows, etc.

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

Given some of the comments here, it is worth pointing out that the retinas are part of the central nervous system. They are bilateral forward-facing out-pouching of the prosencephalon. The retinas and optic nerves you are using to read this are the only parts of the brain that can be directly inspected with an ophthalmoscope.

Retina it is also the ultimate biologically tractable neural network in that input and output channels are very accessible. If you want any biological inspiration for DNNs look no farther.

Highly recommend the first three reference of this new paper to anyone interested in how wet neural network can works. Focus on starburst amacrine cells.

1. A great review of this topic https://pubmed.ncbi.nlm.nih.gov/22314444/

2. Hard core analysis in mice. And yes, we have learned a great deal about vision by studying mice, arguably mouse work is now trend setting even in vision research. This is an example. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3818553/

3. Open article on link up between directionally selective retinal ganglion cells and the rest of the brain. https://journals.plos.org/plosone/article?id=10.1371/journal...

Speaking of world view/perception, after exclusively using a curved monitor for a couple of months, I switched back to a flat monitor, and it appeared curved for a week or so.

Wow I always thought the brain was the CPU, nice to know about the local specialist processing

world view is a very unfortunate choice of words for what they meant

Yep, this headline is similarly problematic:

> Gaze Stabilization in a Moving World

So less consistent philosophy, more steadycam

I took it to be a humorous play of words but still with the actual meaning unobscured. It's not always problematic to have a bit of fun, or is there something more harmful that I'm missing?

No, it's a nice pun and nothing more. People read too much into wordplay, perhaps we need retinal cells that stabilize that

To bring the metaphor back to its original reserve, since the world is highly dynamic, what stabilizes peoples view of it?

mostly ideology

World coordinates refer to points in the actual world versus on a 2d surface in computer vision so I think this language follows from that.

It might well be intentional. Scientists like those weird double entendre titles; it's especially noticeable in paper titles in academia, which often follows the format ("project name: punny phrase vaguely explaining the project").

It's frustrating because as a grad student I was explicitly taught to avoid using local slang, informal sayings and expressions, humor, etc. in my academic writing to make it as understandable and unambiguous as possible.

Reply on news.ycombinator.com