I love optical illusions. First, they’re fun. Second they give us insight into how we think, which is pretty nifty. Pareidolia is a personal favorite, of course, but I also like simple illusions that fool us.
A new study* of how we perceive things shows that context is critical. We know that of course, but what these researchers found is that under some circumstances, our brains make things up as filler more often depending on context.
For example, a vague or faint background allows our brains to fill in the blanks more readily. If you see something, and the background is indistinct, your brain is more likely to add in details that simply weren’t there than if the background is very distinct.
I wonder if this might explain seeing ghosts or aliens in dark rooms (along with my own personal explanation). A vague shape is prime for this effect, and coupled with pareidolia I wonder if this doesn’t very well explain the vast majority of paranormal phenomena.
*Search for the title "Filling-In and Suppression of Visual Perception from Context: A Bayesian Account of Perceptual Biases by Contextual Influences".
Tip o’ the ectoplasmic beanie to BABLoggee Bret Hall.
February 25th, 2008 at 2:43 pm
It’s been my experience with ghosts that they are a combination of pareiodolia, the mind filling in the blank and at least two external stimuli being misinterpreted. this last bit is the important part. it’s easy to write off unexpected refractions of light but when they happen to coincide with a startling noise, the brain starts scrambling for a way to put the two together. In a darkened room, this gets tangled up with ghost stories and viola! A Haunting is born.
February 25th, 2008 at 3:28 pm
I recall many years ago staying overnight at a friends house & sleeping on a couch in the living room. When I awoke the first thing I saw was a door with painted on it a white sailing ship with all its sails set. For a few seconds I was at a complete loss as to where I was. Then I came fully awake, realised where I was and saw that the ‘white sailing ship’ was a featureless reflection in the door’s glossy paint of the light from the window behind me!
February 25th, 2008 at 3:34 pm
Aha! Another reason why eyewitness testimony is extremely unreliable.
February 25th, 2008 at 4:10 pm
Dark – I mean really really dark dark – has a tendency in me to start exploding in varying colors of dark purple, green, red, and other hues as my mind frantically tries to put something in my field of view. It tends to repeat, with a new “wash” of dark color expanding from the center, along with the occasional pattern. Nothing is there, of course, but my mind sure would like there to be something.
February 25th, 2008 at 5:34 pm
Read Pascal Boyer’s “Explaining Religion”. In it, he examines how evolutionary selection pressures in hominids caused us to have “hyperactive agency detectors” develop in our brains, i.e., we ascribe agency (beinghood) to the slightest environmental stimuli, e.g., wind making leaves rustles, strange noises, twigs cracking, and so forth. The reason is that in our evolutionary context as hunter-gatherers (99.9% of our time as a species) it was far, far more advantageous to survival if we mistakenly thought a twig snapping was a bear than if we thought a bear was a twig snapping. LOL. In other words, our brains are primed to see other living agents everywhere around us. Ghosts, demons, and other supernatural beings are but a logical extension of this agency detection module in the brain.
February 25th, 2008 at 5:35 pm
Didn’t someone try a study or experiment showing that sound in the 19 Hz range could produce odd illusions and sensations?
February 25th, 2008 at 6:22 pm
I have the same experience as Pat – I get the same effect if I close my eyes tightly. It is a cool effect and can be quite soothing sometimes
February 25th, 2008 at 6:25 pm
I love optical illusions as well. However, I think the name gives them a disservice–”illusion” implies that there’s something wrong with our visual system; that it exposes some “mistake” in our processing.
I don’t think this is true at all. Vision is a classic underconstrained problem–our brains have to reverse-engineer a 3D scene with only a pair of 2D inputs. These 2D inputs are not so great, either–there are only 3 (4, depending on how you count) frequency ranges, the resolution is limited, the light-gathering capacity is limited, there’s no true depth information, and so on.
Our brains *have* to guess, because there is simply not enough information. There are an infinite number of 3D configurations that map to the same visual stimuli. Further, there are an infinite number of configurations on top of that that represent irrelevant changes to the scene–differences in lighting, for instance. Our brains want apples to be apples, no matter whether it’s daytime or night.
So it’s no surprise at all that our brains will sometimes make the wrong guess about a scene. It’s unavoidable. But the fact that we are fooled so infrequently in real-life is a testament to how good our visual system is. Further, when we are fooled it can often be attributed to making a conservative judgment (I don’t *know* that there’s a tiger in that cave, but it’s better to guess “yes” and be wrong than guess “no” and be wrong).
February 25th, 2008 at 6:48 pm
Dark – I mean really really dark dark – has a tendency in me to start exploding in varying colors of dark purple, green, red, and other hues as my mind frantically tries to put something in my field of view. It tends to repeat, with a new “wash” of dark color expanding from the center, along with the occasional pattern. Nothing is there, of course, but my mind sure would like there to be something.
It is funny you say that, because I’ve seen a very similar effect in dark, dark, dark places (and sometimes even at night in my house in a room where there is no outside light contamination.)
What is even stranger, is that any noise seems to trigger my brain to show different colors or shapes of color near where the sound came from — though this is probably entirely a figment of my imagination. I first saw this in a cave deep underground when someone turned off the lights. I heard (and saw, colorfully) drips of water falling in the cave, and saw other colors associated with other noises. Was really fun, but I wanted the lights back on, since I knew I couldn’t trust it and if I moved around, I’d likely be eaten by a grue or fall into a deep hole. It is quite mesmerizing, and I don’t know if anyone else has ever experienced it. It wasn’t echolocation, I couldn’t create a sound and then see everything, I could only see the “relative” locations of where other sounds were coming from, and I certainly wouldn’t trust it.
February 25th, 2008 at 7:06 pm
What is even stranger, is that any noise seems to trigger my brain to show different colors or shapes of color near where the sound came from — though this is probably entirely a figment of my imagination.
It sounds like synesthesia.
http://en.wikipedia.org/wiki/Synesthesia
February 25th, 2008 at 8:14 pm
Mighty Optical Illusions
February 25th, 2008 at 10:20 pm
I work retail, and I have a related problem with words. If a customer asks for something which I am not expecting, I have a habit of trying to fit the syllables into a cigarette brand name. Often I’m successful, but utterly fail at providing what the customer wants. This is all exacerbated by living in Florida, and having most regional American accents represented at least once a day, as well as the ubiquitous Southern Drawl, which can turn the trade name “Marlboro” into a two syllable word, i.e., “Mahbrugh”.
What is expected is often what is percieved
February 26th, 2008 at 1:43 am
“I wonder if this might explain seeing ghosts or aliens in dark rooms (along with my own personal explanation). A vague shape is prime for this effect, and coupled with pareidolia I wonder if this doesn’t very well explain the vast majority of paranormal phenomena.”
I’m not sure about the “vast majority”, but it is certainly yet another scientific explanation for seeing ghosts. No one explanation is ever going to cover everything, since “ghosts” is such a wide ranging topic. But it’s a bit like the “God of the Gaps” argument – science is finding more and more natural explanations for ghosts, and the world of the paranormal is constantly shrinking. Just as I thought it would.
February 26th, 2008 at 6:03 am
DavidHW writes:
[[Read Pascal Boyer’s “Explaining Religion”. In it, he examines how evolutionary selection pressures in hominids caused us to have “hyperactive agency detectors” develop in our brains...
Ghosts, demons, and other supernatural beings are but a logical extension of this agency detection module in the brain.]]
Seems like a non sequitur to me. We’re evolutionarily primed to see a tiger in the rustling bushes, therefore we conceive of a Creator who is the source of morality?
If you begin by assuming that religious experience is based on nothing real, then of course you can find all kinds of ways to explain it on a material basis. You can then use your material explanation to show that religious experience is based on nothing real. Of course, this is arguing in a circle.
The favored materialist explanation for religion in the ’50s was psychoanalytic — God is a father figure, a projection of the superego, etc. Since we’ve switched from Freud to cognitive theories, our materialist explanation of religion switches along with it. I wonder what materialists will come up with fifty years from now to explain religion?
February 26th, 2008 at 9:02 am
The link to Pareidolia is interesting: it takes you back to this blog entry. Was that deliberate? Maybe it’s just an illusion?
Barton Paul: “I wonder what materialists will come up with fifty years from now to explain religion?”
Yet, surely there is a cause for it. If we take as a given that no evidence has been found for religion then what else could its foundation be except some driving inherent impulse? It’s not surprising that the perception of what that internal impulse might be varies with what we currently think constitutes the workings of the mind.
Sadly, psychology hasn’t yet achieved the level of sophistication needed to call it a proper science. It’s largely a collection of hypotheses that are difficult (if not outright impossible) to test. One possible exception: Cognitive Psych. I have always been deeply interested in the question of what it means to “think” so I took CP as a minor (I’m also a member of the AAAI). I was very disappointed that the focus seemed to be on Who, Where and When (in re experiments) rather than on What. There’s still hope for it though as it’s the only branch that is into objective experimentation at least with regard to perception.
BA: “I love optical illusions. First, they’re fun. Second they give us insight into how we think”
Perhaps but but you got it more right when you used “perceive.” Perception may indeed be the underlying basis of thought but that’s yet to be proven — partly because there is no clear definition of what thought and thinking are.
February 26th, 2008 at 9:10 am
It is true in a great many cases. Many optical illusions very much do expose flaws in our visual system. A classical example is the “disappearing dots” illusion, which appears to be due to a system in the retina (and most other senses) called “lateral inhibition”. This is useful for enhancing the edges of objects, but can lead to problems in specific cases like this.
Another is those illusions with 3D objects where a part in shadow looks to be a different darkness than a part not in shadow, even though their real color is identical. This is due to something called “simultaneous color contrast” (I know, I coached a talk on this). It allows a red apple to appear to be the same shade of read at midday and at dusk, under incandescent light and fluorescent, even though the actual spectral content is completely different under those conditions. But once again it can lead to problems in specific cases.
Yet another are those illusion where you stare at a colored object for a while then look at a solid black of white background and you see the colors reversed. This is due to adaptation in the retina which makes specific parts of the retina less sensitive to specific colors if they are stimulated with those colors for a long period of time. This is a standard part of many neurons and probably also helps filter out stimuli that is constant for long periods of time and thus probably not important.
Many illusions are due to the way our retina is laid out. Most people (I hope) know about the “blind spot” illusions, due to the really stupid placement of the optic nerve. But many other illusions are based on what is called the fovea. This is the center region of your eye and is the only part with even half-decent resolution. It is the only part you can use to read even large text for instance. Try looking at a piece of paper, focusing on one point, then see how far away from that point you can read without moving your eyes. Not very far, I bet. Many illusions require this, since they require you only being able to focus small part of the illusion at any one time. The fork illusion and its variants are good examples of this. This particular issue is not good under any circumstances, but is an inherent flaw in the vertebrate retina. Cephalopods don’t have this problem, nor do they have a blind spot, because their retina is laid out in a logical manner.
February 26th, 2008 at 9:16 am
Gack! The sentence that starts: “If we take as a given that no evidence has been found for religion …” should have read “If we take as a given that no evidence has been found for the foundations of religious belief …”
Obviously, there’s plenty of evidence for religion.
February 26th, 2008 at 9:28 am
“It is true in a great many cases. Many optical illusions very much do expose flaws in our visual system. A classical example is the “disappearing dots” illusion, which appears to be due to a system in the retina (and most other senses) called “lateral inhibition”. This is useful for enhancing the edges of objects, but can lead to problems in specific cases like this”
But to call it a “flaw” is judgemental. More likely, it’s the result of an evolutionary trade-off. If it is indeed a “flaw” then it will be eventually corrected but only if it provides evolutionary disadvantage.
It’s also not clear if the phenomenon has any purpose other than similar occurrences in other senses. The senses are (*ahem*) insensitive to steady-state input. It would seem this allows amplifying departures from normality however this is unproven conjecture.
February 26th, 2008 at 9:40 am
By that definition there is no such thing as an illusion.
February 26th, 2008 at 10:01 am
“By that definition there is no such thing as an illusion”
Not so. Dictionary (M-w) 2a: “a misleading image presented to the vision ”
I can fool the thermostat in my house by holding a hairdryer to it or packing it in ice. IOW: forcing the system outside of what it was designed to do (i.e., outside of its normal operating conditions). The thermostat will be misled but is that a “flaw” in the thermostat? I don’t think so.
February 26th, 2008 at 10:13 am
I’ve never heard the word “illusion” applied to inanimate objects, only sensory stimuli. And where does the word “flaw” appear in that definition? I don’t see it. A more complete definition:
I do not see the word “flaw”, or anything synonymous with flaw, in that definition. In fact the word “flaw” does not appear in any of these definitions:
http://dictionary.reference.com/search?q=illusion
So I am not sure where you are getting this idea that an illusion must expose a flaw.
But it definitely does expose “some ‘mistake’ in our processing”. The processing is optimized for certain things, but certain sorts of stimuli can make it screw up and give erroneous results. That is a mistake (I can give a definition for that as well).
And the last example I mentioned most definitely is a flaw in any meaningful sense of the word. It is an evolutionary accident. It may have been beneficial at some point but for us it causes nothing but problems.
February 26th, 2008 at 10:28 am
Thanks for that link to the Bayesian study!
Here are some more examples for your collection of pareidolia from the world of diagnostic radiology:
Radiology Valentines
Pareido-leo
February 26th, 2008 at 11:17 am
TBC: “So I am not sure where you are getting this idea that an illusion must expose a flaw”
I’m not exactly sure what you’re getting at. either. You said: “Many optical illusions very much do expose flaws in our visual system.” Note your use of the word flaw. My point was that ILLUSION does not equal FLAW.
February 26th, 2008 at 11:42 am
You wrote:
That is synonymous with “a flaw in our visual system”, at least in my understanding of “something wrong with” and “flaw”. Neither appears in any definition of “illusion” I can find.
February 26th, 2008 at 12:12 pm
TheBlackCat, are you one of those posters who says “A” and when another posts “not A; B” you then reply “I said ‘B’ why are you saying ‘A’?” I hope not. It’s quite childish.
BTW, I never said “‘illusion’ implies that there’s something wrong with our visual system” however you (TheBlackCat on 26 Feb 2008 at 9:10 am) said “Many optical illusions very much do expose flaws in our visual system.” It’s there for all to see. Are you really denying having said that? If so, grow up.
February 26th, 2008 at 12:29 pm
Phil: your comment “…a vague or faint background allows our brains to fill in the blanks more readily…” Sometimes the background need not be faint or vague: just random. A fellow graduate student encountered this problem during his first attempt at selecting a topic for his master’s thesis (some forty years ago). A recent (then) paper had suggested that the expanding shock wave from supernovas (and perhaps even novas) caused interstellar plasma to collapse, thus leading to star formation in spherical shells about the supernova event. These shells, it was hypothesized should be visible as circular patterns of stars in which all of the stars were of the same age. My colleague decided to test the hypothesis by locating all circular patterns of stars within a given sector of the sky from available star maps made from photographic plates, and statistically evaluating the number of circular star patterns that contained stars of similar age and then comparing them to the number with differing ages – a not unreasonable test.
Supposedly, if the mechanism proposed was real, then there should be a surplus of stars in circles that had similar formation times. He occasionally elicited the help of his fellow grad students in identifying circles of stars.
It is truly amazing how many circles one sees in an essentially random background of stars in any arbitrary area of the sky: it is also amazing how seldom one maps the same pattern of circles on the same plate (unknowingly) when it is mapped on different days! Even though committed to the original hypothesis, my colleague had to eventually admit defeat and select another thesis topic.
I suppose this is an example where skepticism should be practiced with the same zeal when considering things scientific as it should be with things metaphysical. “That upon which you dwell with the most satisfaction, hold most in suspicion.” (Needs attribution but I can’t remember the original source!)
February 26th, 2008 at 12:51 pm
Hmmm … sounds like a decent hypothesis but I think he went about it the wrong way. 1) It presupposes that enough young stars would form to have a recognizable circle and 2) the stars would all form at the same time. #2 might not be all that unreasonable but #1 is pushing it. I can see eyeballing it to try to cut down the search time but he might just have tried looking at random stars surrounding an event to see if stars of similar age could be found lying close to a circle around it. He would have needed more than one supernova event to prove anything, perhaps a lot. I have no idea how time consuming that would be.
February 26th, 2008 at 6:30 pm
Quiet_Desperation wrote:
It sounds like synesthesia.
http://en.wikipedia.org/wiki/Synesthesia
Yes, it does, but only in pitch black. During normal light conditions, I do not see colors or see the location from where a sound comes from depicted through colors. When a loud noise happens, I hear it and turn towards it to see it, but do not see any colors associated with the noise. I certainly do not experience other elements of synesthesia — I have met someone before who had it and said that he would see colors and shapes often when certain words were uttered, but I certainly don’t have anything like that.
It appears though, at least from this wikipedia entry, that there are folks researching how the brain mixes the senses, and maybe this is just an artifact of the brain compensating for the loss of one of its senses (sight) by teaming it with another one (sound). I’d be interested to know if others have had the same experience. I probably should start by talking to a few blind people.
February 26th, 2008 at 8:20 pm
This might be painfully obvious but noone else has said it; isn’t this the exact thing that led Percival Lowell to believe that there were canals on Mars? His mind assuming there to be specific shapes where there really were none? Or am I totaly messed up here?
February 26th, 2008 at 9:06 pm
One day I was visiting my family I was in the barn with my mother, we hear a noise coming from the eves of the barn on the side of the chicken house. The eves of the barn are open and overhang the roof of the chicken house next door. We both look up towards the top of the wall and there appears to be a snake peering over the wall into the barn, flicking it’s tongue in the air. We both go outside around the barn to the side of the chicken house. On the roof of the chicken house is one of the cats. We both agreed that we saw a snake, down to the detail of the flicking tongue, However since the cat wouldn’t be laying so peacefully near a snake our only recourse is to believe that there never was a snake no matter what we saw. (And that is what science is about, we hypothesized that we saw a snake, we tested our snake theory by moving outside and around to get a better look, and then based on evidence we hypothesized there there never was a snake. So far this theory stands.)
February 26th, 2008 at 11:48 pm
I think people have missed the core of my argument, which is that of *ambiguity*.
TheBlackCat–you mentioned the illusion where a patch of color in shadow appears different than one in light, even if they are actually the same color. You’re probably thinking of that illusion with a cylinder over a checkerboard, and it appears that a dark square is darker than a light square in shadow. So does that represent a mistake or not?
Imagine that we’ve constructed that scene in real life, but in two different ways. In the first way, it’s as you might expect–you have a light source off to one side, and the cylinder therefore casts a shadow. The dark square is legitimately darker than the light square in terms of pigment.
In the other case, though, the actual lighting environment is completely even (like an overcast day). What appears to be a shadow has actually been manually colored in where the shadow would have been. The two squares have been painted the exact same color value.
Perceptually–that is, given the limitations of our eyes–these two scenes are completely identical. If it’s been constructed accurately, there is simply no way for us to distinguish between the two cases, even in principle, unless you add additional sensory inputs.
So our brains need to make a judgment call based on what is more likely–that the shadow comes from an offset light source, or that the shadow was inserted manually and has nothing to do with the light.
Clearly, the former is going to be more frequent. So that’s the prediction our brains make. And it’s the right decision, even with the existence of constructed illusions, because there is no negative survival value with misinterpreting these illusions, while there is indeed negative survival value with misinterpreting real-world scenarios.
Now, I’ll certainly grant that there are illusions that don’t fall into this category. The optic nerve thing is certainly a fault in our visual system (although our brains still deal with it in an impressive way). But I’d argue that most things that are called illusions are really instances of our brains making the correct (and by “correct” I mean the most likely) prediction in the face of ambiguity, and that this ambiguity is unavoidable.
PS: It’s true that nowhere in the definition of “illusion” does it say “flaw”, which is why I said “implies”. The problem is that it’s not our vision system that’s fooling us; it’s reality itself (a constructed reality in the case of many illusions).