How does human echolocation work? (2017) | Hacker News Reader

How does human echolocation work? (2017)

smithsonianmag.com

125 pointsexolymph54 comments

One thing you don't need to be blind to do is to learn how to walk down a hallway with your eyes closed. First, walk past an open door, in each direction, with your eyes closed, and you'll learn how to recognize the sound of an open door. Then do it with a closed doorway, as long as the doorways are recessed a little like they are in a hotel. You'll then be able to identify the sound of an closed doorway. Finally you can practice walking down the hallway with your eyes closed... you can definitely hear yourself getting too close to one side or the other and you will be able to walk an entire hallway with your eyes closed.

I'd imagine, like echolocation, this depends entirely on the acoustics of the environment

Detecting whether doors are closed or open through the sound when you pass them is echolocation.

Might be a mix of sound and touch sensation, too. As you move through the hallway, you push the air in front of you, creating patterns of airflow around you. As you walk past them, closed doors will impact the airflow roughly the same way as hallway wall. Open doors, however, will have a significantly different reaction - perhaps strong enough to feel it on your skin.

there is a certain correlation between blind people who are self-proclaimed echolocators and mobility and employment. The majority of blind people are caught up in this social construction whereby they are restricted and limited. All you have to do is look at the unemployment rates among blind people, and you have an unemployment rate of upwards of 70 percent. So that’s pretty dire. But unnecessary.

Somewhere, they reserve massage as a career for blind people so they will be employable. There are blind developers and I co-own a not active group for such that was begun because some of the tools developers need aren't very blind-friendly.

We absolutely could do a better job of helping people with various impairments make their lives work. Up to 60 percent of people have some degree of impairment but only about 15 or 20 percent self-identify as disabled. The people who are less impaired and are able to make their lives work without much accommodation tend to hide it or at least downplay it and don't call it a disability.

He is absolutely correct that impaired people are impaired at least as much by social climate shaping expectations as by the physical limitation itself. Many things can be accommodated fairly readily in logistical terms IF you can just get people on board with doing what's necessary.

> We absolutely could do a better job of helping people with various impairments make their lives work. Up to 60 percent of people have some degree of impairment but only about 15 or 20 percent self-identify as disabled. The people who are less impaired and are able to make their lives work without much accommodation tend to hide it or at least downplay it and don't call it a disability.

Doesn't the article, ie. the bit you quoted, suggest that this is good for them? How would identifying as disabled improve their life?

I'm not suggesting they should identify as disabled. I'm suggesting that there is no clear, bright line between able-bodied and disabled, some degree of impairment is actually the norm rather than the exception and life could be better for a lot of people if this were more widely understood and accommodated for.

I worked at a company making devices for blind and visually impaired users - screen magnifiers and OCR devices that scan whatever you put there and read it out to you. There are some very low hanging fruits in that domain. It was a very small family company, we had like 20 people total and just 2 developers, and none of us was blind, so we made some UI decisions that were very stupid in hindsight. Fortunately the company also has a distribution network for blind/visually impaired people stuff and we got feedback pretty quickly.

But I think that's one place where a blind developer would be more productive than average.

What I found surprising is the realization that computer hardware gets less accessible for blind people over time - touch screens are much worse than knobs and buttons, and graphical interfaces are much worse than text interfaces.

Web clients could help, but they also evolved from simple semantic DOM + some styling to dynamic js+virtual dom hybrid and "div-34gs34grtt" says nothing to the blind user.

The div-34gs34grtt isn't supposed to be read literally. The name isn't even exposed unless you're poking at the console or something.

All of the visual and hierarchical information abstracted away for visual users in a website that conveys information.

Abstracted visual semantic clues:

  * the type of data in said div

  * the indentation

  * the typography

  * the animations

the layout on the page

  * the interactivity
 with touch and mouse and keyboard events

These clues arw completely lost to blind users. Semantic attributes can help build up that hierarchical information for non-visual users. And yes, the name of a division is semantic information. It serves the purpose of defining an identifier for a section of a document, much like the title of the list above.

Unfortunately true. My biggest gripe with cars is the move to touchscreens. I remember being able to train myself to use a normal car radio with buttons. Now, the only way for me to do that would be to use my iPhone with my screen reader, and I usually need the driver to put the car stereo into Bluetooth mode for me

>self-identify as disabled

and herein lies the problem. to get any kind of support (money, housing, respect) it's ultimately a doctor that decides whether you are disabled or not

Depends heavily on your jurisdiction. In the UK if one has any condition lasting longer than 12 months which would affect the ability to do day to day tasks this would count as a disability automatically. The "test" here is applied without accommodations, so if one needs a cane to walk that would count too.

There's a lot of people in the UK who fall very clearly on the "disabled" side of that equation in the UK who would not self identify as disabled.

I grew up with a blind kid who made a ticking sound with his tongue - he could navigate through the world quite remarkably. It was extraordinary to watch. I remember teaching him how to drive stick - we would drive around a big paddock and I’d tell him what I saw and he’d steer.

is there any reality where ticking sound with his tongue equals being able to drive stick around a paddock. I dunno. I feel like this doesn't work on this scale.

I didn’t mean he could echolocate while driving, it was just a memory of this kid I had, is all.

it is amazing when people make an unnecessary leap like that when reading. Some people do it a lot, and some people never do it.

It may be "unnecessary" but it was logical within the flow of the paragraph as written.

> It was extraordinary to watch. I remember teaching him how to drive stick

If it's unrelated to being extraordinary to watch, which references the clicking, there should be a better segue.

Of course, the response was unnecessarily harsh!

> is there any reality where ticking sound with his tongue equals being able to drive stick around a paddock. I dunno. I feel like this doesn't work on this scale.

I don't know if I was particularly harsh here. I just envision driving around, with the windows up (or down, doesn't really matter), and being able to click and have some kind of "vision". The image made me think it probably didn't work there, but in a different scenario, sure.

It's quite impressive that he had the spatial awareness to drive despite being blind

Echolocation probably helps with that. You don't get the high-frequency components like vision can give you but you get enough of the low-frequency components of the space around you to develop a spatial sense, I would think.

For a glimpse into the effectiveness of human echolocation, there's a restaurant in Singapore - Nox - that's pitch black and staffed by visually impaired hosts. It's a pretty incredible experience for those that have the opportunity.

How did you “watch” in the dark?

I do also watch TV, because I dont see why I should change my language to make pedants like you happy.

I was not trying to be pedantic. I was sincerely intrigued. I imagined you had night vision goggles on and such. It turned out much less interesting.

Sorry, I was obviously a bit grumpy and interpreted your comment wrongly.

As a blind guide, tracking sighted visitors is easier then you might think. Most sighted visitors will tend to be slow movers. If you ever loose track of one of them, and they dont verbalize on their own, you can always ping 'em by calling their name. For added smoothness, give them a small task like "can you figure out what is sitting on the table in front of you?" afterwards, and they will never even know you lost them for a second.

These are all over the place, in german speaking countries the keyword to search for is "Dialog im Dunkeln" and "Dinner im Dunkeln." I used to work as a guide in such a place. Its fun to watch sighted visitors having to adapt to moving in the dark... Can confirm. I am blind, and I do use echolocation a lot. As an example of what is possible, I typically can hear a tree trunk or light pole when walking. The accoustic shadow they cast is enough for detection. If you like, you can play with that yourself. Do you hear the tree?

Can you visualize something in your mind? I mean, there's stuff like the "vOIce" which translates different sounds to the visual cortex area (by training, not by surgery), allowing to "see" yourself over distances. BTW, as you are blind, you might like edbrowse: https://edbrowse.org

Things like "vOIce" would require a lot of training on your side, to adapt your nervous system to the (pretty random) simuli coming in from the artificial perception system. I never met anyone who really achieved anything with that sort of tool. Echolocation is already useful enough, trying to suplement this with current technology is pretty much hopeless, and a bit of hybris.

And yes, I know about edbrowse, but I still prefer lynx :-)

Seems like that would be a safety violation.

Why?

I’m blind. I use this on a daily basis to get around. It helps me cross the street, navigate hallways, and explore new or unfamiliar rooms. Not every blind person has it, however.

Fascinating! Did you always have it, or did you learn the skill? If the latter, could you talk about it?

We have a room at work that was recently remodeled. The acoustics were quite bad and so they added a lot of (quite well camouflaged) sound absorbers to the walls.

It's a super weird feeling to walk past them, and I think it's because the acoustics/echo are very unexpected.

Fantasy: a smart guy would make a discreet 360 degree camera necklace which encodes objects as sounds, and features for facial expressions and signs. And training would be in VR, so they can can learn about train platforms, narrow hiking tracks, dangerous footpaths, surrounded by snarling stray dogs etc in a safe environment.

Sounds like something someone who can see thinks up for the blind. If you do any study at all on the blind you will discover that a lot of gadgets that seem like they would be great are not very helpful. None of the things you bring up are issues most blind people worry about.

Train platforms are designed with bumps on the edge to the blind can feel where to go. (if your local trains station doesn't have this fire management and get it fixed).

The blind mostly are not going on narrow hiking paths. There are plenty of places to hike that are not narrow, dangerous or whatever. Even when they do, the cane works much better because they can feel exactly where the obsticals are, instead of having to figure it out from your sound thing.

I don't know what you are proposing to do about snarling dogs. Does this include a gun attachment to shoot them or something?

I used to go kayaking with my buddy in Monterey. He lost his eyesight due to war wounds, mine stayed-we bonded over that. He taught me something similar so we could go kayak in the sea together, all I had to do was make a noise every so often (~20-30secs) and he would be able to navigate that way, free. Good times.

A Sense of the World by Jason Roberts is a book about the man who invented the use of the cane as a device for the blind to use for echolocation. The book doesn't really focus much on that detail, but it's an incredible story of his life and the many adventures he had despite being blind.

The most interesting thing about human echolocation to me is that blind people (and all people, generally) do it automatically without consciously attempting to do so, and without realizing the source of their information about the world. It's evidence that we don't have access to at least some of our own logical thought processes.

How Well Do We Know Our Own Conscious Experience? The Case of Human Echolocation

https://faculty.ucr.edu/~eschwitz/SchwitzAbs/Echo.htm

"hey!"

"what?!"

"where are you?!"

"over here!"

"ok!"

https://en.wikipedia.org/wiki/Marco_Polo_(game)

I wonder how long the training takes.

Hang out in the shower and with eyes closed move your head around and try to figure out how far you are from the walls. This is very easy and anyone can get a sense for it in 15m or so.

Next, move to an area with reflective surfaces and little soft mass (e.g. a drywalled room without tons of upholstered furniture), walk around a little.bit, and listen for the boundaries with high bandwidth noise present (fans, white noise machines, cars driving by at high speed), or vocalizations like tongue clicking etc., as you walk around. It will become clear.

Finally, extrapolate from these experiences.

Eating a moderate amount of LSD or some functional equivalent will help a lot, especially at the beginning, but it is neither necessary nor sufficient. It helps a lot at the beginning but doesn't add much once one's figured out how to listen to things.

>> I wonder how long the training takes.

> Hang out in the shower and with eyes closed move your head around and try to figure out how far you are from the walls. This is very easy.

not really taking sides here, but given that it takes about 18 years of play and training for boys to fine tune their sensory-neuro-motor eye-hand-everything else coordination (and based on women's gymnastics, maybe slightly less time for girls? probably has more to do with adolescent growth spurts), along the way we pick up echo location like you speak of in the shower. It's subtle, but I imagine it takes a good amount of time/repeated trials to develop, maybe one of those 10,000 hours things. (I realize that book has been debunked, but still)

Sorry, not for development of anything close to sighted perception in normal adults but more an example of how some degree of non sighted perception is practically achivable. Fwiw my eyes work pretty well and I rely on them for everything, but also learned how to not break my toes in the dark etc. The shower thing is of course trivial though. Presumably people that rely on hearing exclusivelly for spatially oriented perception have a whole lot more nuance and detail on which to comment.

> .. but also learned how to not break my toes in the dark

With echolocation?

> Eating a moderate amount of LSD or some functional equivalent will help a lot

I imagine the hallucinatory effects of LSD outstrip its usefulness in creating neuroplasticity. If I had my eyes closed, even on a moderate dose, I would more than likely experience vivid enough visual and auditory hallucinations that figuring out the environment might not be possible.

Although one time on a dose of mushrooms I swear I could hear where people were by their heartbeats. It's unlikely I could actually hear their heart beat from that far away, but I was weirdly aware of where they were and could predict when they would come into the room long before I would normally be able to do so.

Heartbeats and not footsteps?

Ten weeks.

https://www.smithsonianmag.com/smart-news/study-shows-people...

When I was a kid I played hockey and I could hear the boards coming up behind me when skating at them backwards and know when to stop. I guess cold air and hard flat surfaces is a unique-ish acoustic environment but it was definitely a useful input.

Nowhere in the article have I found the words "correlation with a matched filter":

https://en.m.wikipedia.org/wiki/Matched_filter

If you don't know how much you are already aware of the acoustic environment you are constantly surrounded by, try going into an anechoic room. It's disorienting and mildly unpleasant when the sound goes away.

Just as a warning, there is a fair amount of inspiration porn and possible sensationality surrounding human echolocation. There is no doubt that humans can use echos to navigate to a certain extent however there have been reports such as one NPR story I recall hearing where a reporter interviewed am echolocation proponent and treated every claim he made as totally credulous. This included being able to echolocation whilst riding a bike.

That particular story really took off and caused a wave of attention towards echolocation. I will eat my hat if that story was reported with even a modicum of journalistic integrity.

Reply on news.ycombinator.com