Color Phi and Dennett's Theory

Perhaps the cornerstone of Daniel Dennett's case for his ``multiple drafts" view of consciousness in his well-known Consciousness Explained [5] is a set of inferences he draws from the phi phenomenon. Phi was first introduced by the great gestalt psychologist Max Wertheimer [12], and a number of fascinating variations have been studied by, among others, Paul Kolers and Michael von Grünau [7]. In the simplest version of phi, two or more small dots are briefly lit in rapid succession, but it seems to the subject that a single spot moves back and forth. In the color phi phenomenon (the study of which was prompted by questions from the philosopher Nelson Goodman [6]), the two illuminated spots are different colors (red and green, say). Though temporal parameters are imperfect, we have replicated color phi using JAVA, and have put the scheme on the Web for people across the globe to experience. (A snapshot of our system is shown in Figure 1, with the dots here configured as triangles.)

 
Figure 1: Snapshot of Possible Configuration of our Web-based Phi Scheme

Remarkably, if the two spots are lit for 150msec each (50msec gap between them)

the first spot seem[s] to begin moving and then change color abruptly in the middle of its illusory passage toward the second location. Goodman wondered: ``How are we able to fill in the spot at the intervening place-times along a path running from the first to the second flash before that second flash occurs?" ([5], p.114, emphasis his)

Dennett holds that the only way to provide an answer to Goodman's question, the only way to explain color phi, is to invoke his (Dennett's) ``multiple drafts" theory (MDT) of consciousness, according to which information entering the nervous system is under continuous parallel ``editorial revision." Here is Dennett introducing both MDT and color phi:

According to the Multiple Drafts model [MDT], all varieties of perception -- indeed, all varieties of thought or mental activity -- are accomplished in the brain by parallel, multitrack processes of interpretation and elaboration of sensory inputs. Information entering the nervous system is under continuous ``editorial revision" A good way of coming to understand [MDT] is to see how it handles a relatively simple phenomenon that defies explanation by the old theory. Exhibit A is a discovery about apparent motion [phi] that was provoked, I am happy to say, by a philosopher's [Goodman's] question. ([5], p.111, 114, emphasis ours)

MDT is intended by Dennett to supplant traditional accounts of cognition seen, for example, in cognitive psychology -- accounts which include subsystems such as long-term memory, short-term memory, etc., as well as the notion of an ``executive controller" (cf. Anderson's ACT* [2], [1]). Accounts which posit an executive controller are particularly irksome to Dennett: he calls the mental space wherein the controller ``watches" and works the `Cartesian Theater.' Armed with color phi, Dennett also means to overthrow views of the mind which distinguish between some stimulus s seeming to be F to a subject, and the subject's judging that s is in fact F. Here Dennett appeals to what is disclosed when subjects introspect about their experience during phi: he claims that such subjects cannot say, in a principled manner, whether they judged the spots to move because of what they seemed to see, or whether they seemed to see movement because they judged there to be movement.