That’s interesting. As I said it the yellow dots faded pretty dramatically when I did it although they didnt disappear entirely until I threw my eyes out of focus just a little.
Even if that is not true for everyone the explanation still does not fit entirely since the flashing green dot does not appear to be necessary. You can replace it with your finger or the point of a pencil and the illusion still works and in addition you can focus on one of the non-flashing yellow dots and the flashing green dots and the other yellow dots will all disappear. I suspect that part of the rotating background lattice is disappearing at times too although I can;t be sure of it since the individual sections are not distinct.
If someone here knows how to create animations it would be interesting to invert the experiment to make the lattice stationary and rotate the triad of yellow dots around the center. If they still disappear and the lattice does not it would disprove the theory.
OK this is a bit sloppy but here’s my attempt at making an animated gif where the dots rotate instead of the grid. They don’t disappear as I look at them so the fact they they stand still does seem to have something to do with why we don’t see them.
Our eyes detect changes in light intensity, and if the scene does not change, our eyes jiggle a bit to stimulate the light sensitive cells (look up saccadic). The rotating background and flashing green dot supply the movement, so our eyes don’t need to jiggle. Without movement of the yellow dots, they seem to fade.
As far as I know this is the exact explanation. When the eyes do not move at all, one can see nothing in a static scene. In experiments where the eye muscles were anaesthetised, and the head was in a fixed position, only moving things can be seen. The blue grid is moving, so it suppresses the need for the eyes to move a little, and so the yellow dots disappear.
Try this: on a rainy day, stare at one point of a bush, or a pile of sand, until your view gets vague (takes only a few seconds). Then you will be able to notice every drop falling on the leaves or the sand, because those places are the only ones where there is change. It’s a fascinating view!
It makes sense and it seems to fit until you mix things up a bit. If you take the image as is then the theory fits. The blue grid is moving and does not disappear. The green dot is flashing and it does not disappear either so the theory seems to fit what we observe, but try this. If instead of focusing on the green dot, focus on one of the yellow dots. It is not moving and not flashing and yet it does not disappear. The green dot which is still flashing will disappear. This contradicts the theory.
Also in the admittedly crude variation of the animation I posted above the grid is sitting still and it does not disappear.
It makes sense and it seems to fit until you mix things up a bit. If you take the image as is then the theory fits. The blue grid is moving and does not disappear. The green dot is flashing and it does not disappear either so the theory seems to fit what we observe, but try this. If instead of focusing on the green dot, focus on one of the yellow dots. It is not moving and not flashing and yet it does not disappear. The green dot which is still flashing will disappear. This contradicts the theory.
No, that is not what I am seeing. For me the grid and the the flashing green dot are visible, the other 2 yellow dots disappear. I assume that in the optical centre of the retina there are most visual cells, i.e. this is the small area where vision is sharp. That would also mean that only the slightest movements would suffice to let you see the static yellow dot.
In the current consensus, fixational eye movement contributes to maintaining visibility, by continuously stimulating neurons in the early visual areas of the brain, which mostly respond to transient stimuli. In the absence of retinal jitter (a laboratory condition called retinal stabilization), stabilized images as a visual percept rapidly fade out and completely disappear (provided the stabilization is good enough)
So my suggestion to explain it is that we are still moving a little (saccades, micro-saccades, or even your head). But that does not explain that for you even the green dot disappears.
I think we need more observers to make any conclusions. In the original image if I focus on one of the yellow dots the green dot easiy disappears and again, if I freeze the entire image as a screen shot I can focus on the green dot and get the yellow ones to disappear so it seems there is some observer variability here. Its hard to make any solid conclusions when different people are seeing different things.
There is a point for the circling yellow dots (in every rotation), that they all 3, simultaneously, jump forward a small distance. IOW they seem to disappear from one spot and instantly appear a slight distance ahead.
I don’t know if that has anything to do with anything.
There is a point for the circling yellow dots (in every rotation), that they all 3, simultaneously, jump forward a small distance. IOW they seem to disappear from one spot and instantly appear a slight distance ahead.
I don’t know if that has anything to do with anything.
Tim the animation with the moving dots is my sloppy attempt at an animated gif. I never made one before and I was trying to alter the parameters of the experiment to see how it would change the results. It jumps becauseI tried to keep everything about the original the same except I wanted to keep the grid stable and have the dots move. In an attempt to diminish the number of frames I had to create I only rotated the triad of dot 120 degrees at which point the animation should have been able to recycle and look like one continuous rotation but it turns out the dots are not all the same distance from the center nor are the exactly 120 degrees apart so the shortcut resulted in a jumpy animation. Sorry for the confusion.
