Believing Is Seeing

January 2001
by Robert Novella

“You are a creative genius. Your creative genius is so accomplished that it appears, to you and to others, as effortless. Yet it far outstrips the most valiant efforts of today’s fastest supercomputers. To invoke it, you only need to open your eyes.” (Hoffman ’98)

Vision is indeed the paragon of the human senses. With it we can distinguish millions of shades of colors, recognize thousands of faces instantly, and even detect the presence of a single photon of light. The vivid colors, the three-dimensional shapes, and the complicated motions we observe are all the more remarkable when we consider that the image on each retina is two-dimensional, upside down, distorted, and the size of a postage stamp. The vast majority of the information that sighted individuals acquire comes through the eyes and, if given a choice, most of us would rather lose any other sense than our precious vision. This feeling is likewise echoed by the brain’s architecture itself. Almost one third the brain is involved in processing visual information. Indeed, the eyes themselves are really little bits of our brain that are poking out of our heads and peering at the world.

The effortless ease and accuracy with which the eyes present the world to us tends to foster an attitude that they are near infallible arbiters of truth. The eyes are seen as cameras that unerringly bring the objects of the outside world to our minds. Our language and culture reflects this belief with expressions like, “I saw it with my own eyes” or “Seeing is believing.” Visual perception, however, is not such a passive process. Our eyes are not cameras. An exploration of what has been learned about human vision overturns many of these erroneous beliefs and helps illuminate the situations when we need to be skeptical of what we see.

The most common erroneous belief about our vision is that we passively take in light and our eyes, like cameras, send the “pictures” to our brains. Our brains, however, are not like movie theater screens. If that were so it would require another set of eyes to view the picture, which would require yet more eyes and so on in an infinite regression. The modern conception of visual perception involves a process called active construction in which images are assembled, processed, and interpreted starting in the retina itself (Scribner 94) (Hoffman 98) (Wolf 96). Visual signals are not processed as a gestalt; the brain deals with color, shape, and motion as separate problems to solve and then coalesces the finished product into a unified whole. It has been estimated that the retinal image processing alone is roughly equivalent to one thousand million instructions per second (Moravec 00). Vision is therefore not simply determined by a visual stimulus, it is a dynamic search of the available data for the best interpretation. When viewing a scene the brain makes what is essentially an educated guess, using simple assumptions and the information at hand. We think we see with our eyes, we actually see with our minds.

Pattern Recognition and Pareidolia

Our ability to construct and process visual images is clearly demonstrated in our pattern recognition abilities that are absolutely crucial for dealing with and understanding the world. We are pattern-seeking animals because recognizing patterns in nature allowed our ancestors to survive. No computer on the planet can match our ability in this regard. So strong is this innate skill that we easily find familiar patterns in meaningless images created by random and chaotic forces. How many times have you seen a face in the clouds or some other natural formation? We know that there is no real face in the cloud but still the perception persists. Once the interpretation is fixed in the mind it might be impossible to see something else. This is called pareidolia, which means misidentifying a vague, obscure stimulus as a distinct object or person. The image we perceive is not distilled from such stimuli; it is imposed on the stimuli as part of our particular visual interpretation and is therefore subject to error. The survival value of pareidolia for our distant ancestors was likely much higher than it is now. Recognizing a hidden predator or prey from indistinct sights and sounds was crucial to our survival. It’s still valuable today, but with our complex society and belief systems it is given many opportunities to produce believable but bizarre interpretations.

On December 17th, 1996, a woman exiting a building in Clearwater Florida noticed an unusual, colorful shape covering nine of its exterior glass windows. This vague outline (see picture on cover) was soon interpreted as the Virgin Mary making one of her many appearances to mankind. Word spread very quickly and in no time many of the faithful and curious journeyed to the site to see the image. Depending on your source, 500,000 to one million people from coast to coast eventually visited the building stopping traffic for miles around. All forms of media, local and national, also reported the phenomenon for days after this initial appearance. It’s not too difficult to see what appears to be the outline of a hooded robe draped over a head and upper body. As you can see, the image is very vague and many other interpretations could be visualized. This is a classic case of pareidolia.

Many have asked what could have created the image, and I was surprised that none of the accounts I read mentioned what I believe the most likely cause. We’ve all seen rainbow hued patterns in parking lots where oil has dripped and is floating on top of water. This phenomenon is called thin-film interference or iridescence. Oil has much less surface tension than water and therefore spreads out into extremely thin layers (comparable in size to the wavelength of visible light). Incident light can then easily reflect off both the top and bottom surfaces of the oil. The light from the bottom has traveled a little further than the light from the top and thus is slightly out of phase causing mutual interference as they propagate away from the oil. This interference cancels some wavelengths of light and reinforces others creating brilliant rainbow colors of the visible spectrum. I believe that the windows where the image appeared were probably recently cleaned with an oil based cleaner that left a residue on the glass. Thin-film interference then produced the resulting random pattern; pareidolia and active imaginations did the rest.

Pareidolia is not an obscure phenomenon that affects a minority of people with hyperactive imaginations. It affects everyone every day. It is also a viable alternative explanation for countless bizarre experiences based on sense perceptions such as UFOs, Loch Ness Monster sightings, ghost pictures and even the face on Mars. Pareidolia is the result of our extensive visual processing coupled with our hardwired drive to find recognizable patterns everywhere. Recognizing the ubiquity and potency of this phenomenon is invaluable for appreciating vision’s power and limitations.

Expectation and Suggestion

Another powerful example of the constructive, rather than passive, nature of human vision is the effect of expectation and suggestion on what we think we see. Our minds attempt to create not just a consistent image of the world around us, but a consistent and logical story of what is happening. For this reason, what we think we see may also be altered automatically by our brains so that it fits with our expectations of reality. Our brains will fill in missing pieces, and will make assumptions that should be true about what we are seeing. This effect is most powerful when a stimulus is vague or misleading. Magicians use this principle to great effect.

An investigator looking for Bigfoot, for example, when confronted with an unexpected and poorly perceived stimulus, might impose upon that stimulus the expected (and desired) pattern of a bigfoot. While at Loch Ness, the brain will construct a sea monster out of a floating log. Modern viewers seeing an indistinct form in the sky will construct a flying saucer. Meanwhile, viewers from the 1890s, seeing a similar form in the sky, reported seeing imaginary airships (Bartholomew, Howard 1998). After a panda was reported missing from a local zoo, there were dozens of panda sightings throughout the city, despite the fact that the panda was killed by a train a hundred feet from the zoo.

Illusions and Constancy

One of the hallmarks of human vision is the principle of constancy, which describes our stable perception of the world despite significant fluctuations in the images on our retina (Fineman, 98). As we approach objects, as light levels change, even as we tilt our heads, our retinal images change to a degree that is not manifested proportionally in our perception of the world. This is yet another indication that vision is an interpretation, that there is not a one-to-one relationship between what is on our retina and what is in our mind.

It is the failure of constancy, when what we see does not correspond to an objective measurement of an object, that perceptual artifacts or illusions occur. Of particular interest to skeptics interested in the UFO phenomenon should be the failure of size constancy. As an object approaches you, its image on your retina continually grows, but the object does not seem to expand. This is size constancy. Consider this event:

In the early nineties, three airline pilots with about 12,000 hours of combined flight experience were boarding a plane when they spotted what they could only describe as a classic UFO. It had a silvery color, was the size of a 747 and was silently moving in ways that would shame our best jet fighters. Accelerating to mach one (the speed of sound) and at jetliner altitude, it suddenly stopped and shot off at an oblique angle. Amazed and convinced they were witnessing a genuine extraterrestrial craft; the pilots were contemplating their future careers as talk-show guests when the illusion was abruptly shattered. As the craft approached the horizon and was about to disappear behind a line trees it “…unaccountably appeared a hundred feet away on our side of the trees. Instantly the illusion deconstructed into a milkweed seed drifting in the wind. Three airline pilots hoodwinked by a piece of fluff floating in the air.” (Wright, 95)

How could this happen? How could seasoned professionals misidentify a nearby seed as a distant UFO? It was because size constancy failed. The distance cues of the seed were not obvious so the pilots primarily relied on retinal image size to determine distance. Their brains then incorrectly assumed the object had to be miles away. If this assumption was correct then the object would have to be very large and moving very fast, hence their belief that the object had to be a UFO in order to explain its behavior. This brings into question the whole concept of expert witnesses. Many UFO reports are ostensibly bolstered by the experience and credentials of the observer. But we are all human and we all fall prey to impoverished viewing conditions that give rise to perceptual illusions.


It is not my intention to make you question your vision at every turn. Our vision has evolved and been refined over countless millennia. Clearly, it is exquisitely adapted to provide us with a surprisingly reliable and trustworthy representation of external reality. We must understand its imperfections and limitations, however, and that there are certain situations when we need to be just as skeptical about what we see, as we are about what we hear and read.

1) Donald D. Hoffman, Visual Intelligence: How We Create What We See, W.W. Norton & Company New York-London. 1998.
2) Francis Crick, The Astonishing Hypothesis: The Scientific Search for the Soul, New York: Simon & Schuster, 1995.
3) Hans Moravec, Will Computer Hardware Match the Human Brain,, 1997.
4) Mark B. Fineman Ph. D., Sightings: UFO’s and Visual Perception, The New England Journal of Skepticism, vol. 1, issue 3, page 3.
5) Don Wright, It Must Be Real- A Pilot Saw It, Skeptical Briefs vol.5 #4, 1995. Originally published in Phactum, the newsletter for the Philadelphia Association for Critical Thinking.
6) Rainer Wolf, Believing What We see, Hear, and Touch: The Delights and Dangers of Sensory Illusions. Skeptical Inquirer, May/June 1996, pg 24.
7) Bartholomew, Robert E., Howard, George S. UFOs & Alien Contact. Amherst, N.Y.: Prometheus Books, 1998.