1. The chapter appeared first as a Technical Report of the Center for Human Information Processing (now emasculated thanks to unthinking administrators), and later as a chapter in the published proceedings of the conference (Mandler, 1975a), and shortly thereafter as Chapter 3 in my first book on cognition and emotion (Mandler, 1975b).

2. See Shallice (1991) for an account of the revival of consciousness.

3. The best presentation of my current position is a combination of this and the 1992 chapter.

4.There are several uses of "functionalism", the most prominent of which are the following four:

Functionalism1 - the use most frequently adopted by philosophers and cognitive scientists - argues that the study of the "functions" of the organism (and its mind, etc.) can be carried out without reference to the underlying (neuro-)physiological hardware, typically by the manipulation of symbol systems and by complex computation. It is how the brain/mind functions that is of concern, not what its functions are. Functionalism2 is the simplest sense, and used in sensory psychophysics when particular mathematical functions are used to describe variations in experience as a function of variations of the sensory stimulus. Functionalism3 is best represented by the "functionalism" of the Chicago School of psychology around the turn of the century (Carr, Dewey, et al.). It was concerned with observable behavior, its effects and its evolutionary "functions." Its secondary emphasis of describing behavior "as a function of" some external events (e.g., McGeoch, and also the psychophysics of functionalism2) can be seen as a forerunner of American behaviorism. Functionalism4 has been primarily used by linguists who wished to contrast their concern with interactive semantic, syntactic, pragmatic, social, psychological, etc. functions of language with formalist and modular views (e.g., Chomsky, Fodor).

My own approach comes closest to Functionalism4, stressing various functions of consciousness, i.e., asking what consciousness is for, but without wanting to neglect its physical (physiological) representation - when appropriate.

5. A similar interpretation of consciousness was advanced by E. Roy John (in Thatcher and John (1977, pp. 294-304). He noted that in consciousness "information about multiple individual modalities of sensation and perception is combined into a unified multidimensional representation," i.e., that "consciousness itself is a representational system."

6. This phenomenon applies primarily to established representation and not to vague, indeterminate, or novel stimulus situations.

7. Conscious memories ("remembering") are usually initiated by conscious "retrievals" of the relevant unconscious contents, but the latter may also appear without any preceding relevant conscious activity (see Mandler, 1994, for a discussion of such "mind-popping").

8. The "troubleshooting" function of consciousness is an important instance of such constructions. We become conscious of and focus on actions and events that are indicative of a breakdown in usual consequences and procedures (as for example when the brakes of our car fail or a familiar door will not open).

9. I shall not enter into a discussion of any differences between attention and consciousness. The distinction can and has been made (e.g., Kahneman & Treisman, 1984; Mandler, 1985), and for the present purposes it will suffice to agree that attentional processes (under some definitions) will produce conscious contents, but that a conscious content does not presuppose prior attention.

10. For a further pursuit of the analogy between dreaming and creativity see Mandler (in press).

11. Note that these activations are in addition to the usual flow of activation that takes place during unconscious processing (see, for example, McClelland and Rumelhart, 1981).

12. Posner and Snyder's (1975) hypothesis that conscious states preempt pathways by the inhibition of competing possibilities may in part be related to the assumption that such pathways are more available because of preferential additional activation.

13. Decisions among many alternatives (e.g., a restaurant menu) involve a restriction to some manageable subset that is accomplished by the feedback function, followed by (unconscious) decision processes that operate on the remaining choices.

14. This kind of "rehearsal" incorporates both maintenance and elaborative rehearsals (see Craik & Watkins, 1973; Woodward, Bjork & Jongeward, 1973).

15. I had originally included a contrast between slow conscious and fast unconscious processes, but this distinction may be no more than a reflection of the operation of serial and parallel processes.

16. With apologies to Susan Sontag's "The volcano lover."

17. Unfortunately some writers have used primary memory to refer to STM whether conscious or not. I use it as James defined it, i.e., as the "specious present and immediately-intuited past" (James, 1890, I, p.638).

18. The very useful concept of working memory (cf. Baddeley,1989) is more complex than either primary or short term memory, but some parts of it are be coextensive with conscious/primary memory.

19. Which, I believe, originated - like so many things - with D. O. Hebb.

20. My attempts in this direction can be seen as provisional, in the sense that other functions associated with consciousness may be discovered or postulated.

21. See footnote 4.