You are currently at an obsolete version of this web site. You should be automatically redirected to the current version of
If you are not automatically redirected, please click on the link below:
The site will continue to be updated, maintained, and expanded at the new location. This version of the site may not be. |
This debate, principally between myself (Nigel
Thomas) and Patrick Hayes, the well known computer scientist and
Artificial Intelligence researcher, took place through the internet mailing
list for the discussion of the scientific study of
consciousness, PSYCHE-D
(moderated by Patrick Wilken), which is associated with the on-line journal
PSYCHE. The
discussion touches on the various different senses in which the expression
"mental image" may be used, the underlying cognitive mechanisms of imagery,
and the relevance of an understanding of imagery to the understanding
of conscious thought, and thought in general. As the debate became rather
'unthreaded' on the list, following it through this page may help the reader
to better understand what was going on.
The first three links below do not deal with imagery explicitly, but provide a necessary context for following the subsequent debate, and are of considerable interest in their own right. Please note that, due to the way the PSYCHE-D archives are now laid out, you will have to scroll down a frame each time a page loads to get past the header material and see the actual message (not my fault!).
Things really began with the following account by Hayes of the bouts of temporary aphasia (loss of the ability to speak or understand language) from which he occasionally suffers. Hayes argued that, as he does not lose his ability to think during these episodes, natural language cannot be essential to thought:
Hayes on his aphasic episodes.
This provoked Gilberto Lourenco-Gomes to start a thread questioning Hayes about the precise nature of his conscious experience of thinking during his aphasic episodes, and at other times:
Gilberto Lourenco-Gomes initiates the thread "Re: Thought during Aphasia episodes".
Hayes appeared to be saying that he never used conscious inner speech in his thinking. Several contributors to the list found this hard to believe. (This part of the discussion remained fairly well threaded, and may be followed from the above link. However, some relevant exchanges would seem to have taken place privately, off the list). The discussion eventually provoked Hayes into making the following important clarification (in most immediate reply to Rick Prescott):
Hayes on his conscious experience of 'inner speech', and its role in thought processes.
This led to my first intervention, for the first time explicitly raising the issue of the role of mental imagery in general in both ordinary conscious thought processes and in Hayes' aphasic episodes:
Nigel Thomas on imagery and conscious thought.
This brought a reply from Jeff Dalton:
Jeff Dalton on conscious thought.
At this point things became somewhat confusing. Hayes posted a long reply to me, which he quickly asked to be withdrawn from the list (I am not quite sure why). He then, in essence, split this withdrawn posting into two separate replies. The first of these dealt with my points about his views on thinking and about his aphasia experiences.
Hayes on Thomas on conscious thought.
Unfortunately I did not see the above posting at the time (I probably confused it with the withdrawn post in my mailbox). I only discovered it whilst preparing this page. No doubt it would be inappropriate for me to reply in detail at this stage. However, I do feel I should defend myself from the rhetorical accusation that I (Thomas) display "so many prejudices that I [i.e. Hayes] despair of having the time to explain them all". No examples of these alleged prejudices are given. In fact it seems to me that most of the disagreement here stems not from any unwarranted prejudices of mine, but from Hayes' failure to take on board the distinction that I am trying to draw between conscious thought and whatever non-conscious neural and (perhaps) computational process underlie or embody it, and perhaps supplement it. There may well be many representational media for thought in this latter sense (as Hayes suggests), and it may be that some or all of them are properly conceived of as computational. If so, there is a sense in which conscious thought is computational. However, there is also a sense in which it clearly is not. To say flatly that conscious thought is computational would be a bit like saying that cookery is quantum mechanical: true in some sense, perhaps, but for many purposes irrelevant and unhelpful. I have certainly never been aware of my brain's computations as such, neither have I ever been actually conscious of any of the postulated non-imaginal representational media that my brain might be employing in its computations. Judging from his evasiveness when challenged to describe what his thought processes are subjectively like, neither has Hayes (see this post for his clearest response). Incidentally, I never said, and do not believe, that only speaking things can be conscious, and I never intended to deny that Hayes is conscious during his aphasic episodes. I think that there can be conscious, non-verbal imagery!
In any case, Hayes' other reply to my original post was the following insightful posting about mental images:
Patrick Hayes on Mental Images.
Further discussion between Hayes and myself, concerning the nature, proper definition, and functional significance of imagery, then ensued. Although we probably did not reach much agreement, I believe that our exchange was quite useful in clarifying some significant issues:
Thomas replies to Hayes on Mental Images;
Thomas' second reply to Hayes on Imagery.
The discussion did not continue after this posting.
 Click here for the Discussion Board |
or you can email me.
