bigmoon_000105 Anyone attempting to photograph a fat orange full moon rising on the horizon will undoubtedly be disappointed upon viewing the resulting image.
Rather than seeing the enormous orb the photographer remembers, the moon appears as a small circular spot, as interesting as an aspirin hanging against a twilight background.
This is lunar illusion at work: A trick of perception seems to exaggerate the size of the moon when it is near the horizon. While it is no closer or farther away from the viewer on Earth, the low-hanging moon appears a good deal larger than it seems when elevated higher in the nighttime sky.
The popular explanation is that the moon looks bigger when you have something to compare it to. Seeing the moon compared to other objects -- trees, buildings, mountains -- makes it appear larger. The moon alone in the middle of the sky has no surrounding frame of reference, so it appears smaller.
Certainly this is the case, but it doesn't explain what happens in the human brain to change the viewer's perception. Why should a comparison to features on the ground change the way we perceive the moon? What trick is the brain playing that makes the moon appear to change size?
Thinkers throughout history have struggled to explain the illusion, and today the little-understood phenomenon is still a subject of controversy.
Two opposing theories vie for acceptance, but they are contradictory: One suggests that the horizon moon appears large because visual cues in the intervening landscape make the moon seem far away. The other idea says those same cues make the same moon appear closer.
Now a father-son research team's experiments may settle the dispute. Lloyd Kaufman, a professor emeritus of psychology and neural science at New York University, and James Kaufman, a physicist at IBM's Almaden Research Center in San Jose, California, explain their work in the January 4 issue of the Proceedings of the National Academy of Sciences.
The Kaufmans argue that the explanation of the illusion lies in the fact that viewers judge a horizon moon to be much farther away than the overhead moon. The brain then exaggerates the perceived size of the moon as if to drive home this conclusion: the moon is so far away, therefore it must really be huge if it takes up so much space in the sky.
When the moon is at its zenith, there are fewer visible cues the brain can use to estimate the moon's distance from Earth. The brain concludes that the moon is nearer to the viewer and thus offers a perception that plays down its size, according to the theory, which was first proposed by Lloyd Kaufman and James Rock in the early 1960s.
Regardless of the moon's elevation, the distance between an observer (at the center of the horizontal line) and the moon remains constant (unfilled circles). However, a moon perceived as growing closer as its elevation increases (filled circles), must appear to grow smaller. Credit: PNAS, Jan. 4, 2000 issue.
The competing idea, developed a few years after Rock and Kaufman's work, reasons that the moon appears larger over the horizon because the brain judges it to be closer than an elevated moon. Just as a softball 10 feet from the viewer appears smaller than one at arm's length, a closer moon would appear larger, the argument goes.
This is the account that most people will offer when pressed to explain why the horizon moon appears so large, said James Kaufman.
"If you ask people what they see -- whether it's farther away or whether it's near when they're viewing a large moon -- they'll say, 'Oh, it looks close,'" Kaufman said. "They're not saying it looks close relative to the elevated moon. They're just saying 'It's big, it looks close.' "
So which is it? Does the moon appear larger because it is perceived as close, or because it seems far away?
"In no case was there an experimental measurement of how far the brain is treating the horizon moon versus the elevated moon," James Kaufman said. Kaufman and his father discussed the illusion over many years.
"Finally I said, 'why don't we just measure it?' There must be a way we can make an experimental measurement of what the brain is doing as opposed to what people are concluding," he said.
The Kaufmans devised an experiment that used a stereoscopic display on the sky itself. They used a computer display and a set of lenses to project two moons onto a large semi-reflective piece of glass. When a person looked through the glass at the sky, the Kaufmans could reflect the projections of the artificial moons into the viewer's eyes. Thus, a subject would see two realistic-looking moons in the sky.
One of these moons was fixed, and the other could be moved by the subject to appear closer or farther away. The researchers asked subjects to place the variable moon at a point exactly halfway between themselves and the fixed moon. They did this with the moons projected at the horizon (which appeared large, consistent with the lunar illusion), and with elevated moons.
"The task was to move it halfway between you and the horizon moon or between you and the elevated moon," James Kaufman said
In each of five cases, the subjects placed the halfway point between themselves and the horizon moon at a farther distance than the point they picked as halfway to the elevated moon. On average, the subjects placed the horizon moon more than four times further away than the elevated moon.
The Kaufman's argue that these measurements demonstrate that the brain perceives the horizon moon as farther away and adjusts the viewer's perception to emphasize the fact that it must be an enormous object to be so far away and still take up so much of the sky.
This happens even though when asked, subjects said the larger horizon moon must be closer than the elevated moon.
"This is a conclusive demonstration that the elevated moon is registered by the perceptual system as closer than the horizon moon -- even if subjects say otherwise," Lloyd Kaufman said. "In this particular case, what we're learning is that the perceptual system operates in ways that do not entail our conscious awareness."
This same perceptual magnification of objects that seem far away is illustrated by the simple
Ponzo Illusion. Consider two converging lines drawn as if they were railroad tracks stretching toward into the distance. If two horizontal bars of exactly the same length are drawn between the "tracks," they appear to be different sizes. The line that looks to be farther away down the tracks will actually seem larger. The illusion distorts perception of any same-size objects placed between the converging lines."Giving size meaning, if you will, depends upon the perceptual system responding to the information about distance to the object," Lloyd Kaufman said. "A pussy cat close by does look smaller than a tiger far away, even if the Tiger's image is very tiny."
The brain uses a variety of different cues to give meaning to the size of objects -- to keep small objects looking small even if they are nearby, and large objects staying large even if they are far away.
The perceptual system encounters difficulty with celestial objects, though.
"At those vast distances, you can't get an accurate perception of distance," Kaufman continued. "The moon is 240,000 miles away and it's 4,000 miles across, and it doesn't look like its 4,000 miles across."
At the horizon, details in the foreground allow the brain to determine -- even if it can't tell exactly -- that the moon is, at least, very far away. "If you were to look toward an elevated moon where there are less salient cues to the distance to the moon, you locate it at a closer distance and therefore it is perceived as being smaller," he said.
advertisement
