A combined pocket size Hyperscope & Psudoscope,
and more.

Discover the Eyebenda
here...

NTERNATIONAL JOURNAL OF CONTEMPORARY VISUAL ARTISTS
Leonardo Vol.15 N0 3.pp 205-207 1982

David W. Brisson

Use the following links to quickly jump to to further information within this document.

The Moon Illusion - The Relativity of Size Constancy

I have been interested in stereoscopy in its various forms for a long time. As a child I remember very clearly the old stereopticon at my grandfathers beach cottage with the slides, or rather 'display cards', of the Grand Canyon, groups of miners in front of a general store, etc. in miniature three-dimensions. Its the miniature quality of such scenes which interests me now, and which is the subject of this paper.
While visiting the British artist Terry Pope at the University of Reading, I was fortunate enough to have him demonstrate some if his work with stereoscopy to me. One such demonstration involved the use of a pair of mirrored glasses which effectively widened the distance between the two eyes . In other words, by the use of mirrors, very cleverly and beautifully constructed, Pope made it possible to view the world as though the two eyes were situated an inch or so farther apart than they actually are. Pope had not had clear responses from the many persons who had tried looking through these glasses and for quite a long period while I looked through them, I was not sure exactly what was wrong, but I knew that there was something strange about the way things looked. Suddenly, I understood what it was.
The world looked in every way normal, but everything looked, very tiny. Everything was in correct relation to everything else, but everything was small. The coffee cup and the box of matches on the table were minute: likewise a pencil was the size of a tooth-pick. In every other aspect everything seemed right: there was no increase or decrease in the field of vision. No distortion of perspective or anything except a sense of uneasiness when I watched my hands as I picked things up or when things moved.
This seemed a logical result by virtue of the effect of simply changing the amount of binocular disparity. Binoculars, of course are constructed in the same way, but they are always constructed in conjunction with telescopic lenses. In this case there was no magnification of the images. It might be presumed that the mirrors increased the effective distance from the objects and thus reduced the retinal image, which in fact they did, but not by a significant amount, at least not sufficient to produce the enormous effect that occurred.
I was reminded of the effect of the stereopticon which was very similar. Images through the stereopticon had always looked tiny to me in the same way, although the effect was hardly as powerful as in this case. The experience was also connected to the idea of size constancy which is clearly a fundamental aspect of visual perception. This experience seemed to deny the principle of size constancy in a very serious way. By size constancy I mean the well-known perceptual principle that a change in the size of the retinal image of an object does not usually result in a sense of any change in the perceived size of the object. Here was a clear case where the object was perceived as altered in its perceived size without having changed its size on the retina. In fact, as these 'miniature' objects were moved about, they retained their perceived miniature' size. Thus, size constancy was preserved after all! Apparently, however, size constancy is not simply a function of the 'known' size of the object. I should add here, that although the world 'looked' little, at no time did I have any feeling that it actually was the size that it appeared to be. There was no anxiety associated with it at all, but on the contrary a sense of sheer delight with a fascinating experience. The whole experience led me to consider the problem of size constancy from a fresh viewpoint, for clearly it is a much more complex problem than it is usually understood to be.

THE MOON ILLUSION
(return to top of page)

One of the classic problems concerned with size constancy is that of the so-called moon illusion. Many people appear to perceive the moon as looking larger at the horizon than it does when it is overhead. M. Luckiesh demonstrated in his book Visual) Illusions (I) that there is no significant difference in the optical image projected upon the retina to account for this perceived effect. He describes how this was done by a direct process of photographing the moon in its trajectory and carefully measuring the photograph. Various explanations were given, which Luckiesh notes. Among them are James. who suggests that it is basically an effect of atmospheric perspective. Zoth, who suggests it is an effect of the elevation of the eyes, for which Guttman's experiments give some credibility, but perhaps the most interesting explanation is a two-hundred year old theory by the Englishman Robert Smith. His explanation is that our experience with the sky, by way of clouds, etc., leads us to perceive, by way of simple perspective, the sky as a flattened vault. The idea is simply that since the distance to the clouds is usually beyond the range of binocular clues, that the clouds directly overhead are seen as being closer to us than those at the horizon because of perspective. Common observation of cloud layers, even of cirrus clouds, quite obviously supports this argument. Since there is nothing to demonstrate a factor of distance with a clear sky, except perhaps the haze of atmospheric perspective, which is quite a different effect, the suggestion is that the subjective distance of the vault of the sky is learned from the experience of looking through our lifetime at cloud layers in perspective. In short, 'the distance to the sky 'vertically, is subjectively less than the distance in the direction of the horizon. In conjunction with the fact that the projected size of the moon is the same in both cases, on the retina, it is seen resting on the vault of the sky ,and thus is understood to be farther away from us when at the horizon than at the zenith, and therefore bigger at the horizon .

The Relativity of Size Constancy
(return to top of page)

The above analysis of the moon illusion as a variant on the Ponzo illusion, in particular in relation to the effect upon myself of Pope's remarkable glasses, suggests some interesting things with respect to size constancy. Contrary to general consensus, it would appear that is not the size that we know a thing to be that is significant; although there are many clear instances where this is obviously a factor. but rather one's subjective sense of the spatial dimensions of the perceived space.
A similar effect to Pope's glasses may be experienced by those middle-aged wearers of 'reading glasses' with presbyopia. Alter becoming adjusted to wearing the glasses, when looking around without them, things look suddenly small, although not as dramatically so as with Pope's glasses. This is the result of the magnifying effect of the reading glasses, and demonstrates the relation of the two effects.
Having experienced both of these effects. I have observed as well that in neither case is there any loss of the sense of the continuity of the spatiality of the total space. It is not a sense of suddenly perceiving 'small objects in a large space' but rather a shrinking of the whole perceived space. This is a peculiar thing in a sense, for in the case of Pope's glasses, there was no change in the perimeter of the visual field. As a consequence. what must happen is a change in one's absolute sense of space! It is clear from the Ponzo illusion and from the moon illusion that the principle of size constancy can be violated if constancy of size violates the framework of perceived subjective metric space. It is further clear from the effect of Pope's glasses that our sense of metric space is learned. This is not generally evident because we are not usually subjected to a change of perceptual experience such as Pope's glasses affords. The truly startling quality of the Lilliputian space created by wearing Pope's glasses is that size constancy holds for the minuscule objects of that space! Size constancy is thus relative to the perceived metric of perceived space.