Christopher D. Green (2000) Let's not Confuse Artificial Intelligence. Psycoloquy: 11(073) Ai Cognitive Science (13)

Volume: 11 (next, prev) Issue: 073 (next, prev) Article: 13 (next prev first) Alternate versions: ASCII Summary

Topic:

Article:

PSYCOLOQUY (ISSN 1055-0143) is sponsored by the American Psychological Association (APA).

Psycoloquy 11(073): Let's not Confuse Artificial Intelligence

LET'S NOT CONFUSE ARTIFICIAL INTELLIGENCE
WITH COMPUTATIONAL FUNCTIONALISM
Reply to Shanker on Green on AI-Cognitive-Science

Christopher D. Green
Department of Psychology,
York University
Toronto, Ontario M3J 1P3
Canada
http://www.yorku.ca/faculty/academic/christo/

christo@yorku.ca

Abstract

Shanker (2000) argues that "the demise of AI as the paradigm of Cognitive Science has been remarkably swift", and that my target article is an example of this. I disagree. Although I, like Fodor, hold that there is a difference between AI and CF (computational functionalism), and that not all AI provides good evidence for CF, I do not think that CF's day has come yet simply because we have come to question the value of AI-based evidence for CF.

Keywords

artificial intelligence, behaviorism, cognitive science, computationalism, Fodor, functionalism, Searle, Turing Machine, Turing Test.

1. Shanker's (2000) is a particularly interesting counterpoint to Plate's (2000). I'm not at all sure, however, that Fodor's (1992) comments warrant Shanker's (2000) claim that "the demise of AI as the paradigm of Cognitive Science has been remarkably swift" (para. 1). The belief that they do, I think, stems from a failure to grasp the distinction Fodor was trying to forge between CF (computational functionalism) and AI. Although it is true, as Shanker says, that both Bruner and Neisser have moved away from computationalism, as has Putnam, Fodor does not seem to have moved away from it one iota. He has only rejected AI as the primary method. Moreover, "movers and shakers" such as Chomsky, Dennett, Block, Stich, and the Churchlands (not to mention Rumelhart, McClelland, Smolensky, and Hinton) still seem to be strong advocates of computationalism, even if some of them have moved from the symbolic version to the connectionist version. (For all its power, this shift has done nothing to solve the foundational problems of psychology.)

2. I assume that it is my own remarks--the ontological questions that go well beyond what Fodor intended--that lead Shanker to tout my article as one more example of the alleged collapse of AI. I don't think I would go so far--not just yet anyway. Although I believe that psychology still labours under at least some of the "deep conceptual problems" that have afflicted psychological thought for centuries, if not millennia, I still harbour some hope that computation will be able to shed some light on these issues. What computation promised to bring to psychology, inter alia, was a mechanism by which chains of Humean associations could be forged more closely to resemble chains of rational thought rather than chains of free association. It also promised to explicate the crucial notions of mental representation (i.e., in virtue of what properties does a certain thought represent, say, a dog rather than, say, a cat?) and mental causation (i.e., how does the mind make the body move?).

3. The ability of CF to make good on these promises is, to my mind, still up in the air. Lest some people, on this basis, think me to be a wild optimist, let me say also that I think the probability of computation being able to explain other traditional problems--such as consciousness and qualia--to be practically nil. Unlike many devoted CF-ists, I do not regard these as dead issues in psychology; just ones that no one knows how to approach with any measurable degree of success. Fodor (1992) seems to agree. For all CF's problems, however, I see no other even potentially plausible account of any of these issues, and so, on balance, CF still holds my attention, if not my complete allegiance.

REFERENCES

Fodor, J. A. (July, 1992). The big idea: Can there be a science of mind? Times Literary Supplement. pp. 5-7.

Green, C.D. (2000) Is AI the Right Method for Cognitive Science? PSYCOLOQUY 11(061) ftp://ftp.princeton.edu/pub/harnad/Psycoloquy/2000.volume.11/ psyc.00.11.061.ai-cognitive-science.1.green http://www.cogsci.soton.ac.uk/cgi/psyc/newpsy?11.061

Plate, T. (2000) Caution: Philosophers at work. PSYCOLOQUY 11(70) ftp://ftp.princeton.edu/pub/harnad/Psycoloquy/2000.volume.11/ psyc.00.11.070.ai-cognitive-science.10.plate http://www.cogsci.soton.ac.uk/cgi/psyc/newpsy?11.070

Shanker, S. (2000) The demise of AI. PSYCOLOQUY 11(072) ftp://ftp.princeton.edu/pub/harnad/Psycoloquy/2000.volume.11/ psyc.00.11.072.ai-cognitive-science.12.shanker http://www.cogsci.soton.ac.uk/cgi/psyc/newpsy?11.072