Several years ago, I was a juror on a trial in which Goeffrey Loftus was an expert witness for the defense. He testified regarding the fallible nature of eyewitness testimony. If his name sounds familiar, it’s because he’s the ex-husband of noted memory researcher Elizabeth Loftus. In an effort to discredit Loftus the prosecution asked whether Loftus had also investigated the moon illusion, which of course he had. It was an obvious appeal to the lowest common denominator intellectually, as there are always going to be those who have no idea what the moon illusion is, therefore the study of it must be “loony.”
I’ve read a number of essays regarding the moon illusion, some written prosaically and some highly technical. In all the work I’ve read on the subject, I’ve yet to come across what I believe is a rather simple possible explanation, and one whose fundamental principles were understood several hundred years ago!
At this point I need to make an admission. I have failed to do the serious bibliographic work required to get an essay like this taken seriously. I don’t have a degree in psychology or art history so some might dismiss my musings due to lack of credentials. I accept that, yet I’m convinced that my suggestion is at least plausible. Please take this essay for what it is, a suggestion, a preliminary sketch of an idea, not a rigorous argument.
For those not familiar with the moon illusion, it’s the psychological phenomenon whereby a full moon on the horizon seems unusually large; larger than when it’s high in the sky. Astronomers are quick to point out that it’s not an astronomical phenomenon, and defer to psychologists and those specialized in human optical and spatial perception.
For many people “perspective” in art means spatial perspective, i.e. how is three dimensional space depicted? But there is also “aerial perspective” which is (roughly) what effect the atmosphere has on perception of objects at a distance.
Let’s take a look at two paintings from the Renaissance, which will hopefully demonstrate what I’m talking about. The first is Giovanni Bellini’s Pieta’. Obviously the figure of Christ is the focus of the painting, but carefully examine the hillside behind Mary. It doesn’t seem quite “right,” does it? It almost feels composited, as if George Lucas had created it.
Now compare Bellini’s painting to Da Vinci’s Virgin of the Rocks. The rocks in the far distance are far more realistic, and part of the reason why is that Da Vinci’s work was more attentive to aerial perspective than Bellini’s.
When objects are seen at great distances in our atmosphere, they are seen through large masses of air. This tends to do several things. It reduces the contrast of the object as compared with objects at close range, and it reduces detail.
A weird counter-example is how astronauts walking on the moon have reported underestimating the length to big boulders seen at a distance. Da Vinci understood this hundreds of years ago, and astronauts of the 20th century discovered it for themselves: The mass of air between one and a distant object affects one’s spatial perception of the object’s distance.
So how does this factor into the moon illusion? When the moon is at the horizon, several factors are in effect. First off one is looking through a greater mass of air than when the moon is high in the sky. Just like the distant rocks in Da Vinci’s painting, the moon has less detail than when it’s high in the sky. Because the full moon rises not long after the sun sets, the moon also has less contrast against the still-illuminated sky in which it rises.
In my opinion, these two characteristics are sufficient to trigger the brain’s natural perception that the moon is at a great distance. From there our brains naturally adjudge the object to have great absolute size, thus the perception that the moon is larger at the horizon.
In my opinion, the moon illusion is nothing more than a rather unique example of the brain’s natural reaction to aerial perspective.