Becoming an expert in medical image analysis may make you better at spotting illusions.
See full article...
See full article...
People in this career are better at seeing through optical illusions
- Thread starter JournalBot
- Start date
My dad was a radiologist, and he used to tell a story about reading a chest x-ray that was, somehow, just…‘wrong’. Turned out that the patient had the rare syndrome of left-right mirror flipping of his internal organs. Not seeing that, at least at first, was very much a matter of radiologists being trained to look at details of an image.
Upvote
43
(43
/
0)
Artem S. Tashkinov
Ars Scholae Palatinae
Lossy compression (JPEG especially) for these images makes my eyes hurt :-(
Found the first one in PNG, too lazy to find the rest:
Found the first one in PNG, too lazy to find the rest:
Upvote
1
(12
/
-11)
RickRoyLeonPrisZhoraRacha
Ars Praetorian
If you scroll up and down fast, the grey-blue circles look dimensional as though cut out of white paper above grey-blue material.
Upvote
-1
(2
/
-3)
This work seems to be conflating an optical illusion, as opposed to the well-studied effect that humans are much less accurate at comparing changes in area, which are rendered as circles vs. changes rendered as skinny bars.
Upvote
-3
(4
/
-7)
The natural world is fractal. The optical illusions work because they are super clear with the straight lines and smooth curves we don't usually see in our (normal) lives.
A better example we all know is looking at the moon - it appears huge to us, when objectively we know its (effectively) the same distance all the time...
A better example we all know is looking at the moon - it appears huge to us, when objectively we know its (effectively) the same distance all the time...
Upvote
-1
(3
/
-4)
hmmm in the first the left one indeed does look smaller
but for all the other 3 images I also see the left as smaller, most ppl apparently should only see the 3rd one
but for all the other 3 images I also see the left as smaller, most ppl apparently should only see the 3rd one
gulp!
Upvote
12
(12
/
0)
Most radiologists have a good intuition, at least based on my exposure from a couple of rotations in radiology at an academic medical center. I frequently saw radiologists set "weird" studies aside, read some other studies, and come back to the "weird" one. Many use a fallback technique to help them out: I saw one attending run through the CT with various portions of his monitor covered with a piece of paper (left, right, top, bottom). I also remember them going through any "abnormal" study (with an obvious abnormality like a lung mass) a second time with some feature of the software that lets you cover one part of the screen (the abnormality), to make sure you're not being distracted from something else.
Upvote
32
(32
/
0)
I was able to pick the differences out, but only because I did the “Magic Eye” squint and overlapped the circles. I’ve always found visual illusions fascinating but didn’t know about the age/sex/outlook influences, apart from depth perception developing in 3-6 month babies.
Upvote
14
(14
/
0)
Pathologists may be another group that could discern this, due to staring the stained slides of cells of various sized all day and having to make diagnosis.
My late wife and I were watching Life in Color on Netflix, and they had a segment showing how a tiger blends into its surroundings because its prey, some deer, don't see the orange, and mainly see in shades of green. They altered the video to show what the deer sees. My wife was able to instantly spot the tiger, because she was looking for the pattern, not the color.
My late wife and I were watching Life in Color on Netflix, and they had a segment showing how a tiger blends into its surroundings because its prey, some deer, don't see the orange, and mainly see in shades of green. They altered the video to show what the deer sees. My wife was able to instantly spot the tiger, because she was looking for the pattern, not the color.
Upvote
21
(21
/
0)
You'd think that he'd look for that since X-Rays used to be on transparent film and could be put on the lightboard backwards, as infamously done in the opening of Scrubs:
https://screenrant.com/scrubs-xray-gag-joke-deeper-meaning/
Upvote
12
(12
/
0)
Obviously the conclusion of the article was intended to be facetious, there are likely other - less intensive - occupations which may help peopl esee things as they are, rather than what they appear.
I wonder if still life artists have this skill, it's exactly the kind of thing you have to be able to switch off to draw or paint from life.
I wonder if still life artists have this skill, it's exactly the kind of thing you have to be able to switch off to draw or paint from life.
Upvote
11
(11
/
0)
I hope they repeat this study with people in other professions. For example, it might be interesting to see how tennis players, baseball players, golfers, skeet shooters, military fighter pilots, construction crane operators, and others do with these.
Upvote
13
(13
/
0)
As the moon rises/sets, we see it through more atmosphere causing it to visually distort.
Upvote
-13
(1
/
-14)
"Optical Illusions are the opposite of fun"
- Aircraft Pilots
Edited to include a link to the topic of optical illusions in aviation
Last edited:
Upvote
15
(15
/
0)
Upvote
23
(23
/
0)
When I am bored in meetings, I use this to overlap peoples faces and enjoy watching my brain make new faces out of the combined image.
I used to try to “see” through the speakers skin and hallucinate watching a skinless, muscled skull talking. Had to stop doing that one when it started happening on its own.
Brains are cool.
Upvote
14
(14
/
0)
The moon has pretty much the same angular size in the sky regardless of if it is near the horizon or top of the sky (it does change by about 10% at different points in its orbit, but this does not correlate to sky position).
The effect is psychological. Close to the horizon, you compare it to stuff on the earth and it looks large. At the top of the sky you have no reference except the whole sky, so it looks small in comparison.
https://wtamu.edu/~cbaird/sq/2012/12/15/why-does-the-moon-get-bigger-when-its-closer-to-the-horizon/
Edit for lack of coffee.
Upvote
28
(28
/
0)
Not true (or at least that doesn't account for the illusion). Take a cardboard tube (like what a toilet paper roll comes on, or a paper towel roll) and look at the Moon near the horizon. It suddenly looks smaller.
Upvote
9
(9
/
0)
Always thought more about human perception should be taught early in education. We would not have so many people fooled by...everything, basically, if they learned more about illusions, perception, cognitive bias, and all the ways that "thinking" can be flawed..
Upvote
13
(13
/
0)
Question for those more knowledgeable than I: How empirically solid are the broad differences between all these groups, especially when you remove environmental variables?
The lack of illusion-sensitivity among young children I find most plausible, because their brains are actually developing quite rapidly. However, differences among cultural groups seem to fit my pop-media informed perceptions of these cultures all too conveniently. I'm sure the Japanese are less individualistic than Americans are, but why should the way I treat my grandparents noticeably affect the way I see circles on a piece of paper?
The article later states that environmental forces may influence this (such as the likelihood of living in a much denser urban environment), but given that this is the simplest explanation -- daily visual stimuli inform habits of visual perception -- shouldn't we lead with that instead? It also aligns much better with the finding that radiologists are much better at seeing details for what they are, which wasn't terribly surprising. Browsing through the linked papers revealed a lot of talk about how people can be primed for/against a particular illusion.
Perhaps there's a confusion of terms here. Being "sensitive to context" and preferring "holistic" approaches over "analytic" ones probably have more specialized meanings in the field of perception science than they do in (Western) casual conversation about women or folks from East Asia. Most people reading this article likely hail from outside of this specific field of perception science, so they probably use the casual connotations of these terms instead.
Maybe my radar is too sensitive, but my spidey sense starts tingling whenever I hear broad statements about people groups that tempt me to assign social values, instead of front-loading environmental causes that explain these differences far more directly (see the example about the Himba people in Namibia). So, removing environmental forces, how much of our visual perception "depends on who you are?"
The lack of illusion-sensitivity among young children I find most plausible, because their brains are actually developing quite rapidly. However, differences among cultural groups seem to fit my pop-media informed perceptions of these cultures all too conveniently. I'm sure the Japanese are less individualistic than Americans are, but why should the way I treat my grandparents noticeably affect the way I see circles on a piece of paper?
The article later states that environmental forces may influence this (such as the likelihood of living in a much denser urban environment), but given that this is the simplest explanation -- daily visual stimuli inform habits of visual perception -- shouldn't we lead with that instead? It also aligns much better with the finding that radiologists are much better at seeing details for what they are, which wasn't terribly surprising. Browsing through the linked papers revealed a lot of talk about how people can be primed for/against a particular illusion.
Perhaps there's a confusion of terms here. Being "sensitive to context" and preferring "holistic" approaches over "analytic" ones probably have more specialized meanings in the field of perception science than they do in (Western) casual conversation about women or folks from East Asia. Most people reading this article likely hail from outside of this specific field of perception science, so they probably use the casual connotations of these terms instead.
Maybe my radar is too sensitive, but my spidey sense starts tingling whenever I hear broad statements about people groups that tempt me to assign social values, instead of front-loading environmental causes that explain these differences far more directly (see the example about the Himba people in Namibia). So, removing environmental forces, how much of our visual perception "depends on who you are?"
Upvote
7
(7
/
0)
Good an interesting article.
Could we NOT descend down this path:
"People in this career are better at seeing through optical illusions"
Please?
I almost decided not to read it at all, just to rob the title of my one click. ... :-/
Could we NOT descend down this path:
"People in this career are better at seeing through optical illusions"
Please?
I almost decided not to read it at all, just to rob the title of my one click. ... :-/
Upvote
44
(45
/
-1)
Upvote
4
(4
/
0)
My mom was an artist and she'd take a photograph and then make a grid over the photo to base her painting on so each part would be accurately reflected.
Upvote
2
(2
/
0)
"You won't believe what people in this career can do!"
It's reposted from another site, which may not have the same standards as Ars.
Upvote
8
(9
/
-1)
Upvote
2
(2
/
0)
Yes, for the love of all that you hold dear, DON'T start going for this sort of title. It literally made me not read the article and come here just to post on how much it annoys me. Apparently the story is actually interesting, and I might even read it, but for all it's worth (probably not much) I skip ahead 100% of the time when I come across these shenanigans. I realize this is a cross-post from another publication, but if you can't touch the original title, maybe give it a pass?
Upvote
21
(22
/
-1)
But what are the control set of people? You have to pit smart, educated people against the radiologists.
The average 100 IQ person is less likely to be ready for a trick question and they might just answer on gut while the smarter radiologist is on the look out for a trick.
The average 100 IQ person is less likely to be ready for a trick question and they might just answer on gut while the smarter radiologist is on the look out for a trick.
Upvote
1
(2
/
-1)
For all I know, the radiologists have rulers on them and could measure the 5% smaller image. (i.e. 38 pixels instead of 40)
Upvote
2
(2
/
0)
There is also a cultural element at work; I know a number of people who regard being incorrect or admitting mistake or fault to be a show of weakness, and who will just double down if shown to be wrong because their pride and ego demands it.
But some basic media and internet literacy would sure help. It absolutely breaks my brain how many people see an image of a six-fingered Trump rescuing a dog in a flood with a tail that blends into the water and just immediately believe with their entire hearts that such a thing actually happened.
Upvote
5
(5
/
0)
Upvote
17
(17
/
0)
It seems to me that the illusion of a difference in size is connected to an illusion of depth. I tend to see the larger group of circles as closer than the smaller group, even though I know that the image is flat. If one orange circle were actually more distant than the other, then it would be true that it's larger when they cover the same solid angle.
Upvote
4
(4
/
0)
Yeah, frequency transform codecs like JPEG do continuous tone well, but not pixel-sharp edges. PNG can often losslessly encode this sort of image smaller than JPEG would with visible artifacts.
Modern formats like HEIC and AVIF support internal lossless and non-frequency transform modes than can be applied block by block, and can handily outperform both PNG and JPEG at the kind of image those are respectively best at. And even more dramatically for mixed images (like a picture with text block below).
Getting back to the article, traditional image and all video compression themselves are based on optical illusions of sorts. They’re all tuned for the peculiarities of the human visual system, like us seeing how things change more than how they are, our color detail being a lot lower than for luminance, our lesser ability to see distortions in diagonal lines than straight horizontal/vertical, and so very on.
Upvote
10
(10
/
0)
Yes, precisely! While we think of depth perception as being primarily about binocular vision (the differences between what the left and right eyes see), that really only goes out to 20-30 feet before the eyes stop seeing a difference. All of our depth perception beyond that is based on other visual cues. Big ones are knowing how big things normally are, what is in front or behind of something else we may know the distance to, haze, moving our head left and right to give us better angles (in effect giving us eyes much farther apart).
Lots of optical illusions work because they are 2D synthetic images we can only look at straight on, with a visual system tuned for a moving 3D world with tons of other visual cues.
Upvote
7
(7
/
0)
Yeah, our visual system compensates for so many things so we can tell what things are in the real world that can easily be faked out in 2D synthetic images.
It’s not “wrong” we see the same pixel values as being different brightness in the shade, because IRL we’d actually be able to figure out what the grid pattern would look like in different lighting conditions.
Effects like that are a big part of how better graphics rendering looks more and more realistic. Good ray tracing brings a verisimilitude traditional rendering can’t really match. I notice hardcore gamers, who spend so much time looking at synthetic images, seem less awed by good RT than more casuals who spend less time looking at games and more at reality in all its visual complexity.
Upvote
5
(5
/
0)
They also have tools that can measure the exact size of something by drawing a line through it, which would easily reveal that the center circle in the above examples is identical size. Quantification can do a lot to dispel illusions.
Upvote
4
(4
/
0)
Humans actually have somewhat parallel systems in the brain for brightness and color. We see contrast in sharp detail and can capture very fast motion (like tiger stripes against a tree line). While our color perception is lower in detail and speed, but very precise in terms of hue and saturation.
On net, that means we’re good at hunting and not being hunted, but aren’t going to notice if the orange in a tiger’s fur is off until it’s dead. And we can tell if fruit is ripe or poisonous, when it is right in front of us, but not while it is falling from a tree.
Upvote
4
(4
/
0)