3 ms·

Cats are susceptible to some optical illusions is what I take away. In a larger sense, optical illusions mean we can't "trust our eyes" as many exhort us to do

Cats are susceptible to some optical illusions is what I take away. In a larger sense, optical illusions mean we can't "trust our eyes" as many exhort us to do. Does this mean those urging us to trust our eyes are hucksters or trying to manipulate us?

bckr 3y ago
The reason we have a word for optical illusions is because they are outliers. They represent examples of when your optical sense needs to be checked against other perceptions.
- bediger4000 3y ago
  I'd buy that except that the umbrella term "optical illusion" covers a lot of ground from simple physical things like a blind spot, through those color inverting things, to weird ones that involve computation, where you can't see all 16 dots, or the jagged blocks "heal" themselves. Cats perceiving squares is clearly not optical, but rather computational.
xoa 3y ago
>In a larger sense, optical illusions mean we can't "trust our eyes" as many exhort us to do. No, that's not what it (necessarily) means. A lot of optical illusions are specifically engineered to exploit biology or evolutionary gaps, some of which are due to path dependency but others which are likely very sensible in the ultra typical non-synthetic non-adversarial environment. As the article here says: >“Many animals are evolved to perform this sort of perception,” said Smith. “It’s probably to do with navigating the environment. You need to know when not to walk into a tree or off a cliff.” Another way to think of this is "ability to perform high speed object inference from partial information", and it makes a lot of sense this would be pretty important. Much of the natural world is full of woods and grass land where important objects are only seen through a 3D overlay of grass/leaf/branch cover. The examples there might be real but even more so are probably biological ones, like "huh I think that's the shape of a predator hiding there". That's the sort of thing that'd have some real evolutionary pressure. I think we generally take for granted being able to infer what something is from highly minimal information with great reliability normally but it's not like it's a given. So "trust, but verify" while also recognizing sometimes it's better to err on the side of caution. For technologists doing human UI design there are probably usability and safety considerations related to some of this too, designers can make use of negative space in useful and harmful ways. Finally it's fine to just enjoy optical illusions and magic tricks and so on both for their cleverness and what they reveal about ourselves without leaping to some sort of silly "hucksters/manipulators" thing.

bckr 3y ago

The reason we have a word for optical illusions is because they are outliers. They represent examples of when your optical sense needs to be checked against other perceptions.

bediger4000 3y ago
I'd buy that except that the umbrella term "optical illusion" covers a lot of ground from simple physical things like a blind spot, through those color inverting things, to weird ones that involve computation, where you can't see all 16 dots, or the jagged blocks "heal" themselves. Cats perceiving squares is clearly not optical, but rather computational.

xoa 3y ago

>In a larger sense, optical illusions mean we can't "trust our eyes" as many exhort us to do. No, that's not what it (necessarily) means. A lot of optical illusions are specifically engineered to exploit biology or evolutionary gaps, some of which are due to path dependency but others which are likely very sensible in the ultra typical non-synthetic non-adversarial environment. As the article here says: >“Many animals are evolved to perform this sort of perception,” said Smith. “It’s probably to do with navigating the environment. You need to know when not to walk into a tree or off a cliff.” Another way to think of this is "ability to perform high speed object inference from partial information", and it makes a lot of sense this would be pretty important. Much of the natural world is full of woods and grass land where important objects are only seen through a 3D overlay of grass/leaf/branch cover. The examples there might be real but even more so are probably biological ones, like "huh I think that's the shape of a predator hiding there". That's the sort of thing that'd have some real evolutionary pressure. I think we generally take for granted being able to infer what something is from highly minimal information with great reliability normally but it's not like it's a given. So "trust, but verify" while also recognizing sometimes it's better to err on the side of caution. For technologists doing human UI design there are probably usability and safety considerations related to some of this too, designers can make use of negative space in useful and harmful ways. Finally it's fine to just enjoy optical illusions and magic tricks and so on both for their cleverness and what they reveal about ourselves without leaping to some sort of silly "hucksters/manipulators" thing.