Binding Problem

I am still evolving, as I read all this NCC stuff, but in testing myself and my thinking, I find that I produce something very similar to what I have produced several times in the past year or so (also under the Vision tag):

My view has been that the phenomenal visual scene can be likened to a stack of qualia or phenomenal properties, all simultaneously experienced or bound together in such a way that it is often difficult to see the bound parts as distinct from one another, although they are distinguishable in principle. The root of this stack is the set of phenomenal properties that I believe are most often identified with ‘qualia’, i.e. properties that have scalar magnitudes or intensities. Brightness and darkness, color, contrast, and then at a slightly higher order, orientation, scale, direction, speed. These are familiar as physical objects of study either in the psychophysical field of spatial vision, or as determinants of sensitivity in the neurophysiology of the first few synapses of the initial retinocortical pathway for visual encoding. But they are not the only phenomenal properties of visual scenes, and in fact they are not the properties of scenes that we spend the most of our ordinary visual time analyzing. Instead, we spend most of our visual effort attending to more fuzzily inferred properties of the scene: identities, utilities, depths, valences, affordances. These are the properties of a scene that are immediately apparent to us, but they are the ones that require the most inference: the shape and meaning of a word; not so much its contrast or color, which we can easily adapt to and forget, although they remain in our phenomenal consciousness. I am reminded what Foucault said regarding the multiple layers of a calligram: “As a sign, the letter permits us to fix words; as line, it lets us give shape to things.” All these things are simultaneously present and part of the seen scene, but we tend to attend selectively to certain levels.

I think it is clear from this conception of the phenomenal scene that indicating the presence of phenomenal properties, i.e. that something is present in consciousness, requires the presence of the higher level inferences, but not necessarily of the lower level ‘root’. I can daydream or close my eyes and continue to experience visual phenomena, although they are indistinct and insubstantial, and I can tell you about what I experienced, and then we can argue over whether or not visual imagery constitute visual phenomena. However, if all I have is the spatial scene, but I am unable to make any inferences about it, then I cannot report anything about it – reporting presumes context, or cause, or object, and these all require higher level inferences. Or rather, perhaps I could report, but my reports would be nearly meaningless, not least because objective meaning is tied to subjective meaning, which is what we have removed in this example. My reports would, at best, maybe with some minimal inferences, allow me to transmit information about the perceptual magnitude of local, ‘low-level’ features. I would then be performing in a psychophysics experiment, and you would probably be using signal detection theory to interpret my responses. Norma Graham noted the strange convenience of this situation more than 20 years ago, when she noted, “It is (or we can hope it is) as if the simplicity of the experimental situation has made all the higher level stages practically transparent.”