3 ms·

I remember a talk (Ted maybe) about this, the theory being no one actually saw in pictures. As a demonstration the audience was asked to 'picture' a pirate wit

by cortic 8y ago

I remember a talk (Ted maybe) about this, the theory being no one actually saw in pictures. As a demonstration the audience was asked to 'picture' a pirate with a peg leg; Once the picture was firmly in their mind they were then asked what colour of trousers/shirt the pirate was wearing. IF they were actually picturing the pirate it would be as simple as looking but most people didn't initially know and had to 'paint' in trousers/shirt and colour them before answering. Presumably if their mind actually held a picture they would have noticed the lack of trousers, or the lack of colour. The talk went further to suggest that we don't actually 'see' with our eyes either, that our eyes and senses sends raw information, more like a description and the concepts are built in our mind, our experience might be like we're looking through our eyes and see a kind of picture of the world; But actually that is an illusion and our visual experience is more conceptually based which is why we are vulnerable to things like focus-blindness and other optical illusions where any kind of camera or picture taking device would not be.

wongarsu 8y ago
Interestingly, the clear image we think we see with our eyes is very much an illusion (or rather: best effort reconstruction). The area our eyes can see sharp in high resolution is tiny, and our peripheral vision is mostly greyscale. Yet we think we see everything sharp in full color. A similar thing may happen with mental images. You "see" the pirate, but unless you are looking at the trousers you can't tell their color (the same would work with a real life pirate in front of you)
- cortic 8y ago
  If vision was actually made up of conceptually 'tagged' zones, then it would stand to reason that the hierarchical structure would only be developed within our 'focus', so that makes sense. However, if i had a picture of a pirate, even a cursory glance would tell me if he had trousers on, even in the absence of hairy legs, censored, smudged, missing, i would notice it instantly and know it was missing before being asked.. but not if the image was in my mind, which means mental images don't act like actual images.
  - feanaro 8y ago
    Elsewhere in the comments, a person suggests this exact test as a way of telling whether someone is aphantasic or not (https://news.ycombinator.com/item?id=18805666 https://news.ycombinator.com/item?id=18805666). I also remain unconvinced since I seem to construct details such as these automatically. In other words, I do believe I see the colour of the shirt even before the question is asked (though not always).
tbirrell 8y ago
Interestingly enough, as I followed along with your description of the demonstration, I too "painted in" the pirate. But when analyzing the rest of the scene, I realized that I had already created the background of the picture (ship, deck, sea, sky, rigging) and just had to look at those elements.
- cortic 8y ago
  I wonder if the things like ship, deck, sea, sky, rigging have sub-hierarchical tags that are missing in the same way the shirt or trousers colors were missing from the concept of a pirate.
- culot 8y ago
  Ha! As I was reading that and pictured my pirate I was focusing on getting the drape of his pantaloons correct, so they hung over the top of the peg leg just so. When I went back to check color I realized it was like a coloring book image, thickly inked, but colorless.

by cortic 8y ago

wongarsu 8y ago

Interestingly, the clear image we think we see with our eyes is very much an illusion (or rather: best effort reconstruction). The area our eyes can see sharp in high resolution is tiny, and our peripheral vision is mostly greyscale. Yet we think we see everything sharp in full color. A similar thing may happen with mental images. You "see" the pirate, but unless you are looking at the trousers you can't tell their color (the same would work with a real life pirate in front of you)

cortic 8y ago
If vision was actually made up of conceptually 'tagged' zones, then it would stand to reason that the hierarchical structure would only be developed within our 'focus', so that makes sense. However, if i had a picture of a pirate, even a cursory glance would tell me if he had trousers on, even in the absence of hairy legs, censored, smudged, missing, i would notice it instantly and know it was missing before being asked.. but not if the image was in my mind, which means mental images don't act like actual images.
- feanaro 8y ago
  Elsewhere in the comments, a person suggests this exact test as a way of telling whether someone is aphantasic or not (https://news.ycombinator.com/item?id=18805666 https://news.ycombinator.com/item?id=18805666). I also remain unconvinced since I seem to construct details such as these automatically. In other words, I do believe I see the colour of the shirt even before the question is asked (though not always).

tbirrell 8y ago

Interestingly enough, as I followed along with your description of the demonstration, I too "painted in" the pirate. But when analyzing the rest of the scene, I realized that I had already created the background of the picture (ship, deck, sea, sky, rigging) and just had to look at those elements.

cortic 8y ago
I wonder if the things like ship, deck, sea, sky, rigging have sub-hierarchical tags that are missing in the same way the shirt or trousers colors were missing from the concept of a pirate.
culot 8y ago
Ha! As I was reading that and pictured my pirate I was focusing on getting the drape of his pantaloons correct, so they hung over the top of the peg leg just so. When I went back to check color I realized it was like a coloring book image, thickly inked, but colorless.