GEMINILOGGBOOKOBERDADAISTICUS

Gibson's theory of ecological perception

In a reductionistic scheme perception would be studied in the laboratory with artificial tasks, quite dissimilar to our everyday experience. Studies of pitch perception might involve pure sine tones, but absolutely no real musical excerpts. James J. Gibson's ecological theory, which I do not know too much about, points out the necessity of considering the senses as they operate in context in complex environments. Gibson's often cited books were long out of print, now some of them have reappeared, but unfortunately not affordably priced. However, a few papers can be found on the internet.

Gibson's ecological theory of perception has had some influence on certain musicologists and researchers in auditory perception (Eric Clarke, Stephen Handel, among others). This is perhaps not what one would expect, given that Gibson developed his theory mainly in the visual domain. I'm not aware that Gibson would have had nearly as big an impact on more recent research on visual perception, but I could have missed something.

In 1971 there was a lively letter exchange in Leonardo between Gibson, Gombrich, and Arnheim. Although they seem to agree on many points and they all share an interest in Gestalt psychology, there are some subtle differences as to their subject of discussion: what is an image and how do we see it? Gibson presents two prevailing theories, neither of which he finds adequate. According to the first theory, the set of light rays arriving to a viewer from a realistic picture is somehow equivalent to the set of light rays arriving to a stationary viewer positioned where the picture is made; simply put, looking at a picture is similar to looking out through a window. The second theory proposes that a picture consists of symbols, and hence is something one can learn to read. The symbol theory, which also likens representational pictures to descriptions, seems indefensible except for images consisting entirely of symbols, such as pages of text or idiograms. Gibson's argument against the perspective theory is somewhat hard to follow; he argues that we actually see invariants of objects, not just a single perspectival projection of them. Crucially, we are free to move around and see things from multiple angles and distances, which inform how we see them as physical objects as well as the way we see them in flat images.

Ambiguous drawings such as the communicating vases, or a folded sheet of paper which can be seen with the fold pointing either towards or away from the viewer indicate that perception is not deterimined by external stimuli alone; the way we interpret figure and ground or perspectival orientation seems to depend strongly on the viewer. Gibson explains the shifting perceived perspectival arrangement of such figures as the cause of two incompatible kinds of pictorial information present simultaneously in the image.

The conceptual artist Jan Dibbets made a series of photos of "perspective corrections" (1967-69) that show scenes such as a grass field with a carefully cut out square — that is, on the surface of the photo the shape is a square, but out in the field it had to expand outwards from the camera's position. This produces an unsettling uncertainty in the viewer, who has to make sense of two conflicting cues. Yes, the shape is a perfect square imbricated in the picture's frame, but it must be seen also as an outwards expanding trapezoid on the ground.

As Gibson points out, we constantly shift between seeing pictures as that which is depicted, with perspective and depth, and as a flat surface. In an experiment he mounted a large photo of some trees on a wall. Viewers at a fixed position were easily able to estimate the distance to the trees as if they were walking in an alley. But then, when asked about the distance to the photo on the wall, they had no problem estimating that distance either.

For Clement Greenberg, flatness was an essential trait of modernist painting. The more the painting did away with any illusory depth the better. And today, when it is popular to paint from mobile phone photos, many painters experiment with a style that looks photo realistic at sufficient distance, but as you approach the painting disintegrates into raw brush strokes and patches of colour.

glog index

Main page