October 1973
Note on The Conceptual Muddle Underlying the Optical Inversion Experiments
J. J. Gibson, Cornell University
The World Wide Web distribution of James Gibson’s “Purple Perils” is for scholarly use with the understanding that Gibson did not intend them for publication. References to these essays must cite them explicitly as unpublished manuscripts. Copies may be circulated if this statement is included on each copy.
For at least 80 years, since Stratton’s experiment on “inverted vision”, there has been confusion about the supposed puzzle of how we see the world as upright when the retinal image is inverted. Although some thinkers have dimly suspected that it was a false problem, nevertheless efforts to resolve it have continued. In particular, the formula that we somehow learn to perceive the world as upright has persisted. I suggest that ecological optics can show how it is merely a result of conceptual confusion, nothing more.
1. The terrestrial environment (having an intrinsic up-and-down due to gravity) is the standard with reference to which every object of the environment is properly said to be inverted or upright,. Only the “furniture” of the environment is upright, tilted, or inverted. To think that the environment is inverted would be nonsense, a contradiction in terms, and to perceive that it is inverted would be impossible.
2. Similarly, the ambient optic array for a terrestrial environment, consisting of two hemispheres separated by a horizon, is a standard with reference to which any “form” of the array (component solid angle) can be said to be upright or inverted. It would thus be nonsense to suppose that an ambient optic array could be inverted —or even to assert that it is upright. The fallacy comes from confusing the environment with the concept ofspace, which has arbitrary axes of reference.
3. An observer who stands on his head is truly inverted (“upside down”) relative to the permanent terrestrial environment. So is a pot resting on its mouth. When an observer bends over and looks through his legs, his head is inverted. (His visual perception is then somewhat impaired, as experiments demonstrate.)
4. The field of view of an observer with an inverted head is inverted, that is, the eye sockets in the skull. But the content of the field of view is not, that is, the structure of the sample of the ambient optic array registered by the ocular system with that posture of the head. To be sure, the content of the field of view is inverted relative to the head and its eye sockets, but that is a different matter. It is sometimes reported that the “scene” appears to be inverted in that situation but the report is ambiguous: the structure is inverted with reference to the head but not with reference to the ambient array.
5. A picture is an object that can be either upright or inverted, like a pot. That is, it can be hung on a wall or projected on a screen in either position. (When it is inverted the viewer’s perception of what is depicted is impaired, e.g., the facial expression of a portrayed head.)
6. When the retinal image of a chambered (vertebrate) eye is said to be inverted the saying suggests a picture of the terrestrial scene that is projected on the retinal wall upside down. If it were a picture (but it is not) it would be inverted.
7. If spectacles are worn on the head that optically invert or reverse the content of the field of view (by means of a lens system such as Stratton’s or a right-angle prism such as Ivo Kohler’s) the structure of this sample will be inverted relative to the remaining structure of the ambient array. In this case the observer can say without ambiguity that the “scene” is inverted or reversed, for it is true with reference to both the total array and the head. (He might say, of course, that his head was upside down relative to the scene.)
To report that the phenomenal world is upside down in this situation is a far cry from the report that the scene is upside down. Only if an observer confuses the two is he tempted to do so. It would be impossible to perceive that the terrestrial environment was upside down, and it cannot “appear” to be upside down. The popular descriptions of the results of the optical inversion experiments are quite mistaken. The actual results are extremely complicated and have no simple or obvious theoretical implications.
The adaptation that occurs in the course of a spectacle-wearing experiment is not the becoming upright of the phenomenal world, or even the righting of the scene. What is it? Dolezal’s recent study, based on ecological optics, suggests to me that the main phenomenal change is the reduction in what Stratton called the “swinging of the scene” during head-movement, and that this results from a readjustment of the compensatory eye-movements during head-turning. The optical information for the perception of the self (the edges of the field of view) is radically altered in these experiments. The observer has to learn to extract the new invariants that specify the environment, separating them from the new optical motions that specify the movements of the head.
The puzzle of how we see the world as upright when the retinal image is inverted presupposed that first the retinal image is seen (sensed) and then the world is seen (perceived). If this presupposition is rejected the puzzle disappears.
Ecological Psychology and Homeokinetics 