5 ms·

Nit picks from former vestibular researcher : > the liquid in your cochlea sloshes around The vestibules are not the cochlea. The cochlea is responsible for s

Nit picks from former vestibular researcher : > the liquid in your cochlea sloshes around The vestibules are not the cochlea. The cochlea is responsible for sound. The vestibules are just about the sensation of movement, the structures are called Semicircular canals. You have one is each axis of rotation, more or less (x-y-z) > feel the liquid as it passes by them The hair cells are embedded in a crystal matrix, the otolith. The inertia of the 'heavy' crystal pulls the hair cells (which only look like hair and are not actually hair) and that sends a signal to the brain. Not the liquid. The Spins from drunkenness are due to the change in the density of the liquid in the vestibules (from the additional alcohol) causing a different buoyancy of the otoliths and thereby making the hair cells change resting position. > build an image of your position and acceleration Only the acceleration, not the position. You can only feel, via these organs, the 2nd derivative of position (0th derivative - position, 1st derivative - velocity, 2nd derivative - acceleration) > When your inner-ear accelerometer signal fails to cohere with your overall computed sense of motion, you get the feverish-dizzy, lacking-air, mentally-induced-nausea feeling of carsickness. This feeling is very unique to people and their lived experiences. Seasickness will subside for most people, and sailors/figure skaters/ballerinas/skateboarders/etc generally have a higher tolerance to motion sickness. Overall, very good overview! Thank you!

eddyg 3y ago
Thank you for taking the time to reply! Comments like yours are why I enjoy reading (and learning from!) HN.
derefr 3y ago
Glad to be corrected re: the composition of the vestibules! I've not only tried to read up on the topic, but also had chronic labyrinthitis and had a doctor try to explain what was happening there — and in neither case was the explanation anywhere near enlightening as the one you just gave :) > You can only feel, via these organs, the 2nd derivative of position (0th derivative - position, 1st derivative - velocity, 2nd derivative - acceleration) Well, sure, but the thing that I meant by "builds a picture" isn't just what you're feeling (which, indeed, is acceleration, not position), but also the information the brain has available to predict with. IIRC, I've read about "hardwired neural logic" (i.e. a brain function outside of our conscious awareness that we have from birth) for flow-state obstacle avoidance in known environments, which involves the brain doing internal dead reckoning based on integrating over its acceleration-picture of the body — thus allowing a person to e.g. make their way home through a forest they've walked through a million times, with their eyes closed, yet stepping over tree roots at just the right time. Dead reckoning over an integration of the acceleration-picture also seems to come into play with proprioception. Thus why people who are given a local nerve block in an arm, might end up smacking themselves in the face with said arm when trying to e.g. adjust their glasses. They're missing the feedback signals† from the arm that would allow them to compute the acceleration of the arm, and therefore the moment-to-moment position of the arm. --- † I'd love to know how the brain figures out what the acceleration forces being applied to the extremities are. I'm guessing you'd probably be almost the right person to know the answer to that question! (The perfect person would be one of those "prosthetic arm you can feel" researchers.) I know the brain is getting signals back from the muscles in the extremity — how flexed they are, how much work they're having to do to keep in position — and that's probably a majority of the information needed... presuming you are flexing the muscle. But we still know where our limbs are in space even when we have them "loose." So there's likely an additional component — the one that allows people without numbness in their extremities, to avoid being smacked in the face by their own hand when someone picks their arm up and drops it above their face while they're asleep. My naive guess for what that data source could be: pressure on blood vessels. As you move the extremity, the blood drags behind the limb, and so some blood vessels feel an temporary increase in pressure, and others a temporary decrease in pressure. The brain then takes the derivative of the blood pressure data; builds a map of blood pressure deltas; and then "recognizes", in each extremity's picture, the accelerations upon the extremity.
- Balgair 3y ago
  I'm not active in prostheses anymore, but have done a smidge of work in them. The propioceptive positioning does occur in muscles, but, as far as I remember, it's mostly in the tendons. In that, your brain/spine knows the stretch of all the tendons and then back calculates that to a position based on experience. So, people with very recent tendon injuries or differential loads on the muscles will have a somewhat hard time knowing the position of their body (by very recent, I mean within a few minutes/seconds). Blood pressure would be an interesting way to do the math, but as blood pressure changes a lot, it would be doing a lot of work here. Also, most of these systems are inherited from very ancient times when blood was a bit different and before hearts like ours. Animals want to know where they are more than they care to have much of a heartbeat. 'Fun' little ways that we have discovered these things are mostly when someone has a disease that causes them to do wacky stuff. Phantom limb diseases are a bit like that. There are some other ones, but I can't remember then, sorry! If you'd like to learn more about propioception and all the little biological fun times, I'd suggest just about any intro neuroscience textbook. They're kinda fun once you get into them. For grad level learning, Kandel's Principals of Neural Science is the go-to book. Lots of case studies in there.

eddyg 3y ago

Thank you for taking the time to reply! Comments like yours are why I enjoy reading (and learning from!) HN.

derefr 3y ago

Glad to be corrected re: the composition of the vestibules! I've not only tried to read up on the topic, but also had chronic labyrinthitis and had a doctor try to explain what was happening there — and in neither case was the explanation anywhere near enlightening as the one you just gave :) > You can only feel, via these organs, the 2nd derivative of position (0th derivative - position, 1st derivative - velocity, 2nd derivative - acceleration) Well, sure, but the thing that I meant by "builds a picture" isn't just what you're feeling (which, indeed, is acceleration, not position), but also the information the brain has available to predict with. IIRC, I've read about "hardwired neural logic" (i.e. a brain function outside of our conscious awareness that we have from birth) for flow-state obstacle avoidance in known environments, which involves the brain doing internal dead reckoning based on integrating over its acceleration-picture of the body — thus allowing a person to e.g. make their way home through a forest they've walked through a million times, with their eyes closed, yet stepping over tree roots at just the right time. Dead reckoning over an integration of the acceleration-picture also seems to come into play with proprioception. Thus why people who are given a local nerve block in an arm, might end up smacking themselves in the face with said arm when trying to e.g. adjust their glasses. They're missing the feedback signals† from the arm that would allow them to compute the acceleration of the arm, and therefore the moment-to-moment position of the arm. --- † I'd love to know how the brain figures out what the acceleration forces being applied to the extremities are. I'm guessing you'd probably be almost the right person to know the answer to that question! (The perfect person would be one of those "prosthetic arm you can feel" researchers.) I know the brain is getting signals back from the muscles in the extremity — how flexed they are, how much work they're having to do to keep in position — and that's probably a majority of the information needed... presuming you are flexing the muscle. But we still know where our limbs are in space even when we have them "loose." So there's likely an additional component — the one that allows people without numbness in their extremities, to avoid being smacked in the face by their own hand when someone picks their arm up and drops it above their face while they're asleep. My naive guess for what that data source could be: pressure on blood vessels. As you move the extremity, the blood drags behind the limb, and so some blood vessels feel an temporary increase in pressure, and others a temporary decrease in pressure. The brain then takes the derivative of the blood pressure data; builds a map of blood pressure deltas; and then "recognizes", in each extremity's picture, the accelerations upon the extremity.

Balgair 3y ago
I'm not active in prostheses anymore, but have done a smidge of work in them. The propioceptive positioning does occur in muscles, but, as far as I remember, it's mostly in the tendons. In that, your brain/spine knows the stretch of all the tendons and then back calculates that to a position based on experience. So, people with very recent tendon injuries or differential loads on the muscles will have a somewhat hard time knowing the position of their body (by very recent, I mean within a few minutes/seconds). Blood pressure would be an interesting way to do the math, but as blood pressure changes a lot, it would be doing a lot of work here. Also, most of these systems are inherited from very ancient times when blood was a bit different and before hearts like ours. Animals want to know where they are more than they care to have much of a heartbeat. 'Fun' little ways that we have discovered these things are mostly when someone has a disease that causes them to do wacky stuff. Phantom limb diseases are a bit like that. There are some other ones, but I can't remember then, sorry! If you'd like to learn more about propioception and all the little biological fun times, I'd suggest just about any intro neuroscience textbook. They're kinda fun once you get into them. For grad level learning, Kandel's Principals of Neural Science is the go-to book. Lots of case studies in there.