What has Longley got against common sense?
Bill Taylor
It's most unlikely that any reply from the bot itself is going to be
comprehensible, so maybe someone else might like to answer the question
for us?
Thanks.
-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
The "faith" meme discourages the excercise of the sort of critical
judgment that might decide that the idea of faith was a dangerous one.
-------------------------------------------------------------------------------
Seth Russell
Bill Taylor wrote:
> It's most unlikely that any reply from the bot itself is
> going to be
> comprehensible, so maybe someone else might like to answer
> the question
> for us?
I think the premises are something like:
1) The web of science consists of a) trusted procedures b)
actuarial databases c) inferences drawn on extensional
variables with 2nd order predicate calculus [I am not sure
where theories that match b and c are related to the
stance].
2) These are the only things that can be relied on to
predict events.
3) Effective intelligence is equated with the web of science
and its purpose is to predict events.
4) All advances in the web of science come from incremental
advances in a) b) and c) above and these advances come about
by following trusted procedures.
5) Computer science can (and has) implement trusted
procedures.
6) There is evidence from behavioral and neuroscience that
human common sense is fundamentally flawed.
7) There is evidence from logical analysis of natural
language that human common sense is fundamentally flawed.
8) There are examples where human common sense interferes
with trusted procedures, data, and extensional variables.
And the conclusion is:
9) Therefore to achieve effective intelligence the use of
human common sense is contra indicated. Not only should we
model artificial intelligence exclusively on a, b and c
above, but there is no effective intelligence outside of a,
b and c.
[Well somebody had to do it. Now where are the holes?]
--
Seth
See "Bozo's Conjecture" at
http://www.clickshop.com/ai/conjecture.htm
And then on to the AI Jump List ...
Lewis Jorgenson
mat...@math.canterbury.ac.nz wrote:
Taylor:
It's most unlikely that any reply from the bot itself is going to
be comprehensible, so maybe someone else might like to answer the
question for us?
What has Longley got against common sense?
Jorgenson:
I can't resist.
You obviously are unfamiliar with the primary literature and you
have failed to completely read the fragments collection edited
by Mssr. Longley.
Folks have imperfect brains and nervous systems where there
is plenty of documented proof that folks make illogical and
unsubstantiated assertions. This situation is made possible
by relying upon a personal subjective rating of analysis called
the intensional approach. Properness and correctness are
arbitrarily assigned by an inconsistent system instead of
using an objective, consistent logic namely the extensional
approach.
Instead of jumping to conclusions and going beyond the
data one should use such methods as actuarial analysis
upon data to reveal actual rather than imagined associations
and correlations.
Apparently some researchers in a branch of psychology called
"cognitive science" have posited states or entities or
attributes of cognition that are not amenable to actuarial
analysis. Mssr. Longley refers to this as folk psychology
where folks make up stories about folk's psychology and
then proceed as if those stories were documented fact.
The objective standard observation is that every single
word in a language is 'made up' as it were and so one
would have to proceed without using language or words
in order to be true to a pure extensional approach.
The data should and must speak for itself where a change
in the Mercury level of a thermometer does not depend
upon whether it is a Celsius or Fahrenheit thermometer.
It should be commonsense that one should not rely upon
commonsense and there lies the quandry. If we had
enough commonsense to not rely upon commonsense then
that commonsense would be adequate to rely upon.
Man is the measure of all things even though David
does not like that one bit. I vote for the pragmatic
approach where the proof is in the pudding. If the
theory allows useful understanding and predictions
that make the world a better place then it is OK.
When logicalists start asserting that everyone is
delusional and everyone is having contrived
experiences of reality then commonsense must
reassert itself.
Lewis Vance Jorgenson
Neil Rickert
Seth Russell <seth...@clickshop.com> writes:
>Bill Taylor wrote:
>> It's most unlikely that any reply from the bot itself is
>> going to be
>> comprehensible, so maybe someone else might like to answer
>> the question
>> for us?
>I think the premises are something like:
>1) The web of science consists of a) trusted procedures b)
>actuarial databases c) inferences drawn on extensional
>variables with 2nd order predicate calculus [I am not sure
>where theories that match b and c are related to the
>stance].
>2) ...
You are giving the type of answer we might see from the bot. But we
could also give a third person answer.
The bot doesn't have any common sense. Therefore the bot must try
to persuade ordinary folk to not use their common sense, because
otherwise the bot would be at a disadvantage.
David Longley
Perhaps those interested might like to try reading:
http://www.longley.demon.co.uk/Frag.htm
or downloading FRAGMENT.ZIP from the same website? I suspect I
*may* be able to express it all somewhat more concisely now than
I did when I first drafted it, but it was not long ago that John
McCarthy asked me to paraphrase what I meant by "the extensional
stance", and I believe he understood what I posted, albeit on the
second attempt. When I use the term "effective" I do so in the
sense that Church-Turing used the term. When I use the term
extensional, I do so as Quine (1954;1992;1995) does.
What I have put together in "Fragments" draws on a very large
body of research work across several disciplines. In the end, the
"extensional stance" has an explicit domain of application - and
the merits of that application can be assessed and evaluated in
terms of outcome (see the other papers on applying behavioural
technology in custodial settings at the above website). The
referents here are explict - interested readers will soon
come to appreciate that if they read the specified papers
The early paragraphs of the paper above provide enough of a
context for the genuinely interested reader - all I ask is that
people read the original and not just the personal points of view
which have appeared to date in this thread.
If all is not clear after readers have consulted the provided
papers, please contact me by e-mail.
--
David Longley (check end reply line #)
Longley Consulting London, UK
Behaviour Assessment & Profiling Technology,
Research, Data Analysis and Training Services,
Small IT Systems http://www.longley.demon.co.uk
Seth Russell
Stanley Friesen wrote:
> On Thu, 28 May 1998 16:25:05 -0700, Seth Russell
> <seth...@clickshop.com> wrote:
> >I think the premises are something like:
> >
> >1) The web of science consists of a) trusted procedures b)
> >actuarial databases c) inferences drawn on extensional
> >variables with 2nd order predicate calculus.
> >2) These are the only things that can be relied on to
> >predict events.
>
> I would say this is incorrect.
>
> These are the *most* *reliable* ways to predict outcomes, but
> certainly not the *sole* way to do so.
Actually I don't know of anyplace in the bot's writings where he has used
the word "only" in such a sweeping context. Perhaps that statement should
have been coded:
2) Use of the web of science is the only reliable way to predict outcomes
for the purposes of scientific advancement and professional practice.
> >3) Effective intelligence is equated with the web of science
> >and its purpose is to predict events.
> And, if this is what Longley really is trying to say, then this
> is where we part company.
Actually we could remove the redundancy with 2) and code 3) better:3) The
web of science is the only effective intelligence.
> Effective intelligence is being able to act effectively in a
> changing, uncertain world under conditions of incomplete data.
I know of no places where the bot has written about effective intelligence
outside of the context of *scientific advancement and professional
practice.* What is more when pressed about use of intelligence in child
rearing , art and commerce he has specifically stated that he had said
nothing about those domains. Perhaps we should record the actual premise
of the bot as follows:
3a) There is no effective *artificial* intelligence outside the web of
science, nor will there ever be.
And perhaps we should add the following to the premises.
3b) Modeling *any* AI after human common sense or emotional motivation is
"fundamentally flawed" .
> >6) There is evidence from behavioral and neuroscience
> > that human common sense is fundamentally flawed.
>
> I would also contend that Longley is wrong here.
>
> The actual results are more subtle than this. What is found is
> that human heuristic decision making processes are error-prone.
> This is *not* the same as saying they are flawed. To go from the
> former to the conclusion that human heuristics are flawed also
> requires evidence that there exist less error-prone procedures
> that can lead to effectively action under the conditions
> pertaining with living things.
>
> Under the time and informational constraints that living organism
> are subject to, there is actually considerable evidence that
> biological heuristics, error-prone as they are, are close to the
> optimal strategy.
I don't think the bot would contradict you point, though I believe it would
find it irrelevant to its stance which seems to concern itself with the web
of science and professional practice exclusively. I have no memory of any
instances of the bot responding in areas where human values (or common
sense) meet science/profesional-practice.
> >7) There is evidence from logical analysis of natural
> >language that human common sense is fundamentally flawed.
>
> And this, likewise, equates not following correct logic with
> being flawed. Again, this equation is dubious in the context
> of real life.
Well since the bot's domain web-of-science/profesional-practice does not
inclued "real life" its logic is sound.
> >8) There are examples where human common sense interferes
> >with trusted procedures, data, and extensional variables.
> >
> Accepted. This is why scientists have instituted such
> error-checking procedures as peer-review and replication.
Well we would all seem to be in agreement here, except that the bot would
probably want to restrict peer-review to extensional language.
> >And the conclusion is:
> >
> Science is distinct from effective intelligence, and for good
> reason.
I don't believe that the bot could accept that conclusion without altering
its premises.
[I was going to add that anyone interested in the exact (and apparently
complete) belief structure of the bot should consult Frag.htm at the UK
site, however the bot apparently has come back on-line. And not any too
soon, this was beginning to be a drag.]
Seth Russell
Neil Rickert wrote:
> You are giving the type of answer we might see from the bot. But
> we
> could also give a third person answer.
>
> The bot doesn't have any common sense. Therefore the bot must
> try
> to persuade ordinary folk to not use their common sense, because
>
> otherwise the bot would be at a disadvantage.
Well yes that is one of its strategies. But an even more successful
strategy is to make outrageous claims about "common sense" and
common folk so as to become notorious. The ad homonym behavior also
advances this same strategy. All in all this is one of the most
successful bots of record. After all, here we are discussing its
agenda long after it has gone off line. That the author has deemed
it more important to advertise his site and his stance, than to win
the Leobner prize is beyond my comprehension.
Stanley Friesen
On Thu, 28 May 1998 16:25:05 -0700, Seth Russell
<seth...@clickshop.com> wrote:
>I think the premises are something like:
>
>1) The web of science consists of a) trusted procedures b)
>actuarial databases c) inferences drawn on extensional
>variables with 2nd order predicate calculus.
>2) These are the only things that can be relied on to
>predict events.
I would say this is incorrect.
These are the *most* *reliable* ways to predict outcomes, but
certainly not the *sole* way to do so.
>3) Effective intelligence is equated with the web of science
>and its purpose is to predict events.
And, if this is what Longley really is trying to say, then this
is where we part company.
Effective intelligence is being able to act effectively in a
changing, uncertain world under conditions of incomplete data.
>...
>6) There is evidence from behavioral and neuroscience that
>human common sense is fundamentally flawed.
I would also contend that Longley is wrong here.
The actual results are more subtle than this. What is found is
that human heuristic decision making processes are error-prone.
This is *not* the same as saying they are flawed. To go from the
former to the conclusion that human heuristics are flawed also
requires evidence that there exist less error-prone procedures
that can lead to effectively action under the conditions
pertaining with living things.
Under the time and informational constraints that living organism
are subject to, there is actually considerable evidence that
biological heuristics, error-prone as they are, are close to the
optimal strategy.
>7) There is evidence from logical analysis of natural
>language that human common sense is fundamentally flawed.
And this, likewise, equates not following correct logic with
being flawed. Again, this equation is dubious in the context of
real life.
>8) There are examples where human common sense interferes
>with trusted procedures, data, and extensional variables.
>
Accepted. This is why scientists have instituted such
error-checking procedures as peer-review and replication.
>And the conclusion is:
>
Science is distinct from effective intelligence, and for good
reason.
>[Well somebody had to do it. Now where are the holes?]
See above.
Stanley Friesen
On 29 May 1998 02:21:36 GMT, LSV...@prodigy.com (Lewis
Jorgenson) wrote:
>Folks have imperfect brains and nervous systems where there
>is plenty of documented proof that folks make illogical and
>unsubstantiated assertions.
No disagreement here. This is obvious and not contested by
anybody I know of.
> This situation is made possible
>by relying upon a personal subjective rating of analysis called
>the intensional approach. Properness and correctness are
>arbitrarily assigned by an inconsistent system instead of
>using an objective, consistent logic namely the extensional
>approach.
This seems to be about right.
I maintain that, given the limitations of real life, this is the
best possible strategy for many (but not all) situations.
>
>Instead of jumping to conclusions and going beyond the
>data one should use such methods as actuarial analysis
>upon data to reveal actual rather than imagined associations
>and correlations.
This is where the contention begins.
There are many results, especially in the last 5 years, that
clearly indicate that such a strategy is impractical and
ineffective under many actual real world conditions.
The problems with this strategy come up when major costs or
benefits accrue to *rapid* decision making, and the current
situation of the agent does not permit *immediate* acquistion of
adequate data to apply "actuarial" methods.
Under these conditions, any procedure which allows immediate
action and at the same time increases the probability of
effective action by even a small amount is favored over slower
but more reliable procedures.
>
>Apparently some researchers in a branch of psychology called
>"cognitive science" have posited states or entities or
>attributes of cognition that are not amenable to actuarial
>analysis.
I have rarely seen any such thing an the primary literature.
What I *have* seen is cognitive scientists positing states and
entities that lead to intelligent agents not using actuarial
analysis.
> Mssr. Longley refers to this as folk psychology
>where folks make up stories about folk's psychology and
>then proceed as if those stories were documented fact.
Yet he seems to never actually give any concrete examples of
this.
Lewis Jorgenson
sfri...@americasttv.com wrote:
Jorgenson:
>Folks have imperfect brains and nervous systems where there
>is plenty of documented proof that folks make illogical and
>unsubstantiated assertions.
Friesen:
No disagreement here. This is obvious and not contested by
anybody I know of.
Jorgenson:
I was being silly. I certainly disagree where I see the
emphasis of that statement certainly misrepresents the
situation. The brains and nervous systems of most folks
are certainly perfect enough for most tasks presented by
nature. The tasks presented by other folks are a novel
arena of tasks. Also there is plenty of documented proof
that folks are capable of making logical and substantiated
assertions. The emphasis upon human cognitive flaws is
preposterous as it requires human cognition to make that
claim and human cognition to understand that claim and
human cognition to act in accordance with that claim.
Longley's approach is patent nonsense. His exaggerated
emphasis results in a cartoon.
>This situation is made possible
>by relying upon a personal subjective rating of analysis called
>the intensional approach. Properness and correctness are
>arbitrarily assigned by an inconsistent system instead of
>using an objective, consistent logic namely the extensional
>approach.
Friesen:
This seems to be about right.
I maintain that, given the limitations of real life, this is the
best possible strategy for many (but not all) situations.
Jorgenson:
The authors of technical papers are discouraged from using
the words 'I' and 'We' to cast a spell of objectivity over
the work. Thus one would say that the data suggests a
linear relationship rather than say that some of us think
the data shows there could be a linear relationship.
That objectivity serves to remind folks that essentially
anybody could get that data and that anybody would probably
see a possible linear relationship in that data. That
objectivity is contrived as data and observations and
conclusions can be lies or mistakes despite the appearance
of being rationally impossible of error.
>Instead of jumping to conclusions and going beyond the
>data one should use such methods as actuarial analysis
>upon data to reveal actual rather than imagined associations
>and correlations.
Friesen:
This is where the contention begins.
There are many results, especially in the last 5 years, that
clearly indicate that such a strategy is impractical and
ineffective under many actual real world conditions.
The problems with this strategy come up when major costs or
benefits accrue to *rapid* decision making, and the current
situation of the agent does not permit *immediate* acquistion of
adequate data to apply "actuarial" methods.
Under these conditions, any procedure which allows immediate
action and at the same time increases the probability of
effective action by even a small amount is favored over slower
but more reliable procedures.
Jorgenson:
You are quite correct and somehow one must also account for
individual events rather than statistical events. It
is all very well to say that A will occur significantly
many more times than B and yet humans still must deal with
A *and* B because both do occur. Certain individuals do
in fact win those lottery monies.
>Apparently some researchers in a branch of psychology called
>"cognitive science" have posited states or entities or
>attributes of cognition that are not amenable to actuarial
>analysis.
Friesen:
I have rarely seen any such thing an the primary literature.
Jorgenson:
You caught me. I was guessing. I did see one book on
cognitive science and the portion I read seemed to be
relying upon anecdotal wisdom and stereotypes. There
is some level of legitimacy to such approaches and
yet it is almost unscientific and certainly prone to
over-generalization. Anecdotal episodes usually
confound rather than support actuarial analysis.
Friesen:
What I *have* seen is cognitive scientists positing states and
entities that lead to intelligent agents not using actuarial
analysis.
Jorgenson:
Huh? Do you mean that actuarial analysis is contra-indicated
by the discovered states and entities or that actuarial
analysis in not required in light of the discovered states
and entities or what did you mean?
>Mssr. Longley refers to this as folk psychology
>where folks make up stories about folk's psychology and
>then proceed as if those stories were documented fact.
Friesen:
Yet he seems to never actually give any concrete examples of
this.
Jorgenson:
Are you incapable of recalling any such examples on your own?
How about demonic possession or Ids and egos and super egos
and other rationalizations explaining behavior?
Actually I think we agree on the major points and I am just
beating this poor dead horse some more for something to
occupy my mind.
Lewis Vance Jorgenson
lsa...@aol.com
In article <6kpmal$d2n6$1...@newssvr04-int.news.prodigy.com>,
LSV...@prodigy.com (Lewis Jorgenson) wrote:
>
> sfri...@americasttv.com wrote:
>
> Jorgenson:
> >Folks have imperfect brains and nervous systems where there
> >is plenty of documented proof that folks make illogical and
> >unsubstantiated assertions.
>
> Friesen:
> No disagreement here. This is obvious and not contested by
> anybody I know of.
>
> Jorgenson:
> I was being silly. I certainly disagree where I see the
> emphasis of that statement certainly misrepresents the
> situation. The brains and nervous systems of most folks
> are certainly perfect enough for most tasks presented by
> nature. The tasks presented by other folks are a novel
> arena of tasks. Also there is plenty of documented proof
> that folks are capable of making logical and substantiated
> assertions. The emphasis upon human cognitive flaws is
> preposterous as it requires human cognition to make that
> claim and human cognition to understand that claim and
> human cognition to act in accordance with that claim.
> Longley's approach is patent nonsense. His exaggerated
> emphasis results in a cartoon.
I concur wholeheartedly with this. Longley is wrestling with
a strawman of his own making. Biological nervous systems are
miracles of reliability. They are the only goal oriented systems
we know of that actually get better as the amount of information
and knowledge they process gets more complex. Our extensionally
engineered complex systems are so full of bugs that their
unreliability has become an international crisis. And the more
complex they get the more unreliable they become. This is the
exact opposite of natural learning systems. As an example, just
think of the amazingly reliable processing (not to mention the
extremely high number of minute decisions) that takes place when
we go up and down a flight of stairs of drive around an unfamiliar
town without getting into an accident. Longley's constant belitling
of natural nervous mechanisms is just a vehicle for showing off his
worthless and/or misapplied erudition. I hope our future robots
do not inherit his pomposity.
The goal of AI scientists is to emulate the mind-boggling
performance of biological systems so we can all go to the
beach and leave the work to our highly trustworthy robots, the
artificial nervous systems of which will be based on natural
neurons. We just need to figure out what neurons really do. And
we will do it very soon, IMO, the rants of a million Longleys
notwithstanding.
Regards,
Louis
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
Oliver Sparrow
Seth Russell <seth...@clickshop.com> wrote:
""
"" [Well somebody had to do it. Now where are the holes?]
As with a Beirut bus-shelter.
Your admirable summation is (dare I say it?) common sense, with which most
people who are not posting to alt.looney would agree. So where is the beef?
Fine. We do that. Now what?
My chief problem with the longley.bot is twofold (beyond the practical
issues of volume, of course.) The first is that an approach which divides
the conceptual universe in the light of truth and the darkness of unknowing
denies us the tools by which to bring the light unto the dark. That is, the
schemes which are advanced use old tools to examine new data, but they
never point to new tools. That which old tools (and capital T,
established-and-in-the-database Truth) can describe as True is wholly
defined and unchanging. This is a recipe for knowing ever-more about
ever-less. Further, it denies all that we know about knowledge
representation: few buy all the very mixed PoMo stew, but there is a
certain validity in the view that the ways in which we understand and
represent truth are socially constructed.
The second problem is with the linear model of knowledge (research
management, innovation, strategy) which is advanced, and its evident
inadequacy and actual rejection in practice. In practice, we iterate around
our attempts to clarify what we are thinking about or what we are setting
out to undertake. We tend to reconstruct a historical pathway to truth
during the process of explaining to outsiders why a given point of view is
true. (See, for example, the closing chapter of the Los Alamos Primer on
particle physics, in which this re-invention of the paths to the standard
model is discussed by those who took part in its formulation.)
Formulated truth is much more than social convention. Nevertheless, it is
subject to being ripped up as often as a major highway, and its pipes and
conduits re-layed. The linear model of engineering: specify the
Eurofighter, define a development process, prove the concept, learn how to
make it, fly it.... will have taken from about 1966 to 2002. The relatively
simple Cruise was specified in the early Sixties, flew in the Eighties with
a 1 kilobyte memory & a 4-bit processor. See also the Hubble telescope.) It
becomes too difficult to change things and less-than-optimal equipment is
included, despite its being known to be pathetic when compared to the
latest child's toy on the market. The head of the UK defence research
establishment recently showed a new model of research, in which concepts,
procurement, tasks and engineering modules co-develop, aiming to achieve
less-than decadal cycles.
Reality in organisms and programs (and almost certainly, AIs) is concerned
with iteration, with homing in on useful ways of seeing and using these as
oblique lighting by which to throw the whole into relief. Knowing about an
issue is about deploying many tool kits, not about Templates of Truth.
______________________________________
Oliver Sparrow
David Longley
The problem with all that I have seen in this thread so far is
that the authors just don't know how to differentiate their own
beliefs/desires from the facts which make up the knowledge base
of contemporary empirical psychology and philosophy of mind/AI.
The sine qua non for such discrimination is a sound knowledge of
the literature - a point I have made many times and one which
most rational folk would regard as uncontroversial. Nevertheless,
I have gone to some lengths to provide sign-posts to this
literature, along with quite representative extracts in an effort
to focus discussion on the conclusions which I have drawn from
the evidence.
The references and extracts are, I submit, a helpful and
essential guide to a literature which few except professionals
are likely to have encountered.
Some issues covered in:
http://www.longley.demon.co.uk/Frag.htm
and other papers at the same web-site include:
a) the nature of enlightened (Quinean) empiricism.
b) the significance of the clinical vs actuarial literature for AI.
c) the nature of folk psychology and descriptive empirical psychology.
d) the fragmentary nature of behaviour & merits of the extensional stance.
e) indeterminacy, intensionalism and other problems of cognitivism.
f) the application of AI as the extensional stance.
Neil Rickert
sfri...@americasttv.com (Stanley Friesen) writes:
>On Sun, 31 May 1998 20:23:11 GMT, lsa...@aol.com wrote:
>>... As an example, just
>>think of the amazingly reliable processing (not to mention the
>>extremely high number of minute decisions) that takes place when
>>we go up and down a flight of stairs of drive around an unfamiliar
>>town without getting into an accident.
>Great examples. These are exactly the sorts of things that I
>maintain are *impossible* for Longley's type of "extensional"
>systems.
>And I would suggest that it is these sorts of problems that are
>the most interesting in AI work. (I really *would* love a robot
>that could actually drive a car as well as, or better than, I
>can, under all driving conditions).
When we drive a car, the amount of information we pick up and use in
our judgements vastly exceeds what is picked up by any automated
system we have ever designed.
The illusion of traditional AI (both symbolic and connectionist), is
that there somewhere exist some truly marvellous algorithms, if only
we could find them, which could get by with the meager amount of
information we collect with automated systems, yet still outperform
the human in spite of the huge information advantage available to
that human.
Stanley Friesen
On Fri, 29 May 1998 16:43:38 -0700, in Seth Russell
<seth...@clickshop.com> wrote:
>Stanley Friesen wrote:
>Actually I don't know of anyplace in the bot's writings where he has used
>the word "only" in such a sweeping context. Perhaps that statement should
>have been coded:
>
>2) Use of the web of science is the only reliable way to predict outcomes
>for the purposes of scientific advancement and professional practice.
>
Which is practically a tautology. (I might quibble slightly
about the "professional" part).
>> And, if this is what Longley really is trying to say, then this
>> is where we part company.
>
>Actually we could remove the redundancy with 2) and code 3) better:3) The
>web of science is the only effective intelligence.
>
Which doesn't really change my response.
>> Effective intelligence is being able to act effectively in a
>> changing, uncertain world under conditions of incomplete data.
>
>I know of no places where the bot has written about effective intelligence
>outside of the context of *scientific advancement and professional
>practice.*
Then what relevence is this to designing an AI? An AI needs to
be intelligent in a much wider context than just scientific
advancement. An "AI" that can not do anything except science
will have a hard time dealing with people!
> What is more when pressed about use of intelligence in child
>rearing , art and commerce he has specifically stated that he had said
>nothing about those domains. Perhaps we should record the actual premise
>of the bot as follows:
>
>3a) There is no effective *artificial* intelligence outside the web of
>science, nor will there ever be.
Such a limited AI is of little interest or power. It would be
unable to even understand the contents of a photograph, as doing
so requires heuristic procedures to recognize partially obscured
objects.
>
>And perhaps we should add the following to the premises.
>
>3b) Modeling *any* AI after human common sense or emotional motivation is
>"fundamentally flawed" .
>
Only if one accepts that humans are "funamentally flawed" in this
sense. I would rather say that humans are the result of a design
compromise, and are optimized for their original environment.
>
>> I would also contend that Longley is wrong here.
>>
>> The actual results are more subtle than this. What is found is
>> that human heuristic decision making processes are error-prone.
>> This is *not* the same as saying they are flawed. To go from the
>> former to the conclusion that human heuristics are flawed also
>> requires evidence that there exist less error-prone procedures
>> that can lead to effectively action under the conditions
>> pertaining with living things.
>>
>> Under the time and informational constraints that living organism
>> are subject to, there is actually considerable evidence that
>> biological heuristics, error-prone as they are, are close to the
>> optimal strategy.
>
>I don't think the bot would contradict you point, though I believe it would
>find it irrelevant to its stance which seems to concern itself with the web
>of science and professional practice exclusively. I have no memory of any
>instances of the bot responding in areas where human values (or common
>sense) meet science/profesional-practice.
I do not merely mean *values*, or even common sense, here.
Merely recognizing a horse partially hidden behind a tree
requires inexact, error-prone precedures. (In that context we
call the errors "optical illusions"). Deciding if that bit of
tawny color is a lion or a clump of grass, and what to do in
regard to it, is of considerable significance, and cannot be
determined by the sorts of "effective procedures" the Longley bot
talks about.
>
>> And this, likewise, equates not following correct logic with
>> being flawed. Again, this equation is dubious in the context
>> of real life.
>
>Well since the bot's domain web-of-science/profesional-practice does not
>inclued "real life" its logic is sound.
Which, to my mind, again takes his point out of the realm of
interest in *creating* an AI, since any real AI, to be
sufficiently flexible in relating to humans, will need to deal
with "real life" at least some of the time.
And for such interesting areas of intelligence as scene
interpretation, spoken word comprehension, and navigating through
a complex environment without constantly asking a human where to
go next, dealing with "real life" is necessary to even accomplish
the activity.
The Mars rover may have been a technolgical marvel, but a true AI
could do the job even better, and without all the standing around
waiting for instructions. For that sort of thing, there is no
substitute for dealing with "real life".
>
>> Accepted. This is why scientists have instituted such
>> error-checking procedures as peer-review and replication.
>
>Well we would all seem to be in agreement here, except that the bot would
>probably want to restrict peer-review to extensional language.
>
Gee, I suppose that means most of linguistics is not scientific?
:-)
>> >And the conclusion is:
>> >
>> Science is distinct from effective intelligence, and for good
>> reason.
>
>I don't believe that the bot could accept that conclusion without altering
>its premises.
Well, it seems to have a different definition of "effective
intelligence" than I do.
I suppose I could reword this to:
Science is distinct from real-world intelligence, and for good
reason.
Stanley Friesen
On 30 May 1998 19:23:01 GMT, LSV...@prodigy.com (Lewis
Jorgenson) wrote:
>sfri...@americasttv.com wrote:
>I was being silly. I certainly disagree where I see the
>emphasis of that statement certainly misrepresents the
>situation. The brains and nervous systems of most folks
>are certainly perfect enough for most tasks presented by
>nature.
This is mostly a matter of preferred usage. I prefer to say that
most folk's brains are "sufficient" or "adequate", or even
"efficient" for most real tasks. I dislike the use of the word
"perfect" in the context of engineeering trade-offs.
[Much that I see no reason to disagree with deleted].
>>What I *have* seen is cognitive scientists positing states and
>>entities that lead to intelligent agents not using actuarial
>>analysis.
>
>Huh? Do you mean that actuarial analysis is contra-indicated
>by the discovered states and entities or that actuarial
>analysis in not required in light of the discovered states
>and entities or what did you mean?
Sorry. I made that too convoluted.
I meant that in day-to-day activities real people do not use
Longley's "actuarial analysis" to make decisions, and that this
fact has been confirmed by cognitive science.
[I *hope* that is clearer].
>>Yet he seems to never actually give any concrete examples of
>>this.
>
>Are you incapable of recalling any such examples on your own?
>How about demonic possession or Ids and egos and super egos
>and other rationalizations explaining behavior?
I was talking about in the context of cognitive science and
*modern* psychology. Certainly there have been such folk myths
in the past, but they are irrelevent to the current state of
cognitive science.
Stanley Friesen
David Longley
In "Fragments" and in a number of extended discussions related to
the themes developed in the above, I have made the point that
since the early 1970s the old practice of modelling natural human
decision making has shifted from using normative models (which
was typical in the Information Processing phases of Cognitive
Psychology from the early 50s on), to descriptive models which
accept that the former do not provide very good fits.
I have also pointed out that this subtle change in research has
implications for naive GOFAI which uncritically continues to
attempt to incorporate "common-sense" into its models in an
attempt to increase what is sometimes called their "ecological
and face validity".
Furthermore, I have made the case that normative models are
inseparable from the evolving web of knowledge which we refer to
as science, and as such, any attempt to claim these procedures,
(almost always regimentable within the effective language of the
predicate calculus) for AI is likely to be ephemeral is not
naive.
The "extensional stance" draws on the W.V.O Quine's external or
enlightened (post 1951) radical empiricism. It eschews "common
sense" heuristics simply because we know from empirical research
that these heuristics result in well circumscribed biases
(documented at length by Tversky, Kahneman and others).
"AI", from the extensional stance, just is the application of
technology derived from our empirical web of unified scientific
knowledge. Whilst some might prefer to restrict the range of
applications somewhat, I see no logical reason why this must be
done other than that of arbitrary administrative convenience.
I illustrate how this quite conservative view works out in
practice in a field where application of natural human judgement
can be empirically evaluated against the implementation of the
extensional stance.
http://www.longley.demon.co.uk/Frag.htm
and other papers at the same webite.
Those requiring consultancy services should make contact through
the above website.
Bill Taylor
sfri...@americasttv.com (Stanley Friesen) writes:
|> > Mssr. Longley refers to this as folk psychology
|> >where folks make up stories about folk's psychology and
|> >then proceed as if those stories were documented fact.
|>
|> Yet he seems to never actually give any concrete examples of this.
HEAR HEAR!!
I too, would like to see this glaring lacuna filled!
As would many others here, it seems likely.
So, if the Longley-bot is still scanning this thread,
please accede to this request.
NOTE:- We would like to see a BRIEF post with such examples...
...in your OWN words;
...with no excerpts from or references to any other literature;
...taking less than 2 pages, but more than a paragraph;
...if you would be so kind.
Until there is money there, the mouth is hollow.
-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
The first European words spoken in the new world...
"They don't look very Chinese, Christopher."
-------------------------------------------------------------------------------
David Longley
In article <6l59js$lpr$3...@cantuc.canterbury.ac.nz>
mat...@math.canterbury.ac.nz "Bill Taylor" writes:
>
> HEAR HEAR!!
>
> I too, would like to see this glaring lacuna filled!
The lacuna is in your knowledge - *you* fill it yourself. I'm not
prepared to run crash courses in psychology and AI for folk who
can't read references.
Extract from an Open University Third Level Course:
Professional Judgment and Decision Making:
'There is no controversy in social science that shows such a large
body of qualitatively diverse studies coming out so uniformly in the
same direction as this one. When you are pushing 90 investigations,
predicting everything from the outcome of football games to the
diagnosis of liver disease and when you can hardly come up with a half
dozen studies showing even a weak tendency in favour of the clinician,
it is time to draw a practical conclusion, whatever theoretical
differences may still be disputed.'
Meehl 1986, pp373-4
COURSE TEXT:
JACK: Meehl's study set off, or at least much inflamed, the
'statistical versus clinical judgment' controversy, which has rumbled
on ever since, though it's somewhat less fashionable than it was.
PENELOPE: Why?
JACK: Cynically, because the human judges didn't like the results and
made sure that they or their authors didn't get the funding,
circulation or promotion they deserved. Closed shops (as most
professions are to some extent) are not likely to vote for what they
see as de-skilling, and alternative approaches that showed more
respect for the human judge became fashionable and fund worthy
(especially the expert systems we shall meet in the session after
next). Uncynically, the methodological problems in policy-capturing
research are real: it IS difficult to establish the external validity
of the results.'
Page 63 Volume 1 Introductory Text 2
And finally:
'Kelley's (1973) 'covariation principle', which is the
most fundamental assumption of contemporary attribution
theory and of that theory's characterization of people
as generally adequate intuitive scientists, is
essentially an assertion that the layperson can
recognize and make appropriate inferential use of
covariation between events. Common sense also suggests
that the ability to detect covariation would seem
necessary for understanding, predicting and controlling
social experience. The assessment of covariation between
early symptoms and later manifestations of problems, of
covariation between particular behavioural strategies
and subsequent environmental outcomes, and of
covariation between potential causes and observed
effects seems critical to people's success in responding
adaptively to the opportunities and dilemmas of social
life. Indeed, one might well be tempted to regard
harmonious social interchange and the general
effectiveness of personal functioning as EVIDENCE of
people's capacity to recognize covariation between
events.
Most research that has dealt with people's abilities to
recognize and estimate covariation has not been
flattering to the layperson's abilities. Even Peterson
and Beach (1967), in their generally positive evaluation
of 'man as an intuitive statistician' were not very
optimistic about people's capacity to appreciate
relationships between variables. In this chapter we
first review evidence of these shortcomings and then
attempt to reconcile the apparent contradiction between
peoples' failures in the laboratory and their successes
in meeting the demands of everyday social judgement and
interaction.
Judging Covariation from Fourfold Contingency Tables
Most of the research available at the time of the
Peterson and Beach review was on people's ability to
estimate association correctly from fourfold, presence-
absence tables of the type presented below, in which the
task is simply to determine whether symptom X is
associated with disease A.
DISEASE A
Present Absent
Present 20 10
SYMPTOM X
Absent 80 40
This task would seem superficially to be the least
demanding covariation-detection problem that one could
pose. The data are dichotomous rather than continuous.
There are no problems of prior data collection,
estimation, or recall; there are no prior, potentially
misleading notions of the relationship; and the data are
conveniently arrayed in summary form that should promote
accurate assessments of covariation. Nevertheless, the
evidence (for example, Smedslund 1963; Ward and Jenkins
1965) shows that people generally perform such tasks
quite poorly.
Almost exclusive reliance upon the 'present/present'
cells seems to be a particularly common failing. Many
subjects say that symptom X is associated with disease A
simply because many of the people with the disease do in
fact have the symptom. Other subjects pay attention only
to two cells. Some of these will conclude that the
relationship is positive because more people who have
the disease have the symptom than do people who do not
have the disease. Others conclude that the relationship
is negative because more people with the disease do not
have symptom A than have it.
Without formal statistical training, very few people
intuitively understand that no judgment of association
can be made legitimately without simultaneously
considering ALL FOUR cells. The appropriate method
compares the ratio of the two cells in the 'present'
column to that of the two cells in the 'absent' column.
One might be tempted to dismiss this research as simply
a demonstration that lay people cannot 'read'
contingency tables and that the errors and biases shown
are artifacts of the unusual format of the judgmental
task. The incapacity, however, resembles shortcomings
that have been observed in circumstances that do not
require 'table reading'. In chapter 3 we reviewed the
literature on the ability of people (and animals) to
learn from negative or null instances. It should be
recalled that learning is greatly retarded or prevented
altogether when the instances are conceptually negative
('blue things are NOT gleeps'). The finding that
subjects are preoccupied with the present/present cell
in contingency tables is reminiscent of people's
inability to learn from negative instances.'
R. Nisbett and L. Ross (1980)
Ch.5 Assessment of Covariation p.90-92
Not only does this sit well with Popperian falsificationism and the
Quine-Duhem thesis (Quine 1951) as the LOGIC of scientific discovery,
but in eschewing intuitive strategies, in not going beyond deduction
(not 'going beyond the data given'), a co-operative use of technology
can in the end only pay dividends in the quest to more effectively
manage the Prison Service:
'Humans did not "make it to the moon" (or unravel the
mysteries of the double helix or deduce the existence of
quarks) by trusting the availability and
representativeness heuristics or by relying on the
vagaries of informal data collection and interpretation.
On the contrary, these triumphs were achieved by the use
of formal research methodology and normative principles
of scientific inference. Furthermore, as Dawes (1976)
pointed out, no single person could have solved all the
problems involved in such necessarily collective efforts
as space exploration. Getting to the moon was a joint
project, if not of 'idiots savants', at least of savants
whose individual areas of expertise were extremely
limited - one savant who knew a great deal about the
propellant properties of solid fuels but little about
the guidance capabilities of small computers, another
savant who knew a great deal about the guidance
capabilities of small computers but virtually nothing
about gravitational effects on moving objects, and so
forth. Finally, those savants included people who
believed that redheads are hot-tempered, who bought
their last car on the cocktail-party advice of an
acquaintance's brother-in-law, and whose mastery of the
formal rules of scientific inference did not notably
spare them from the social conflicts and personal
disappointments experienced by their fellow humans. The
very impressive results of organised intellectual
endeavour, in short, provide no basis for contradicting
our generalizations about human inferential
shortcomings. Those accomplishments are collective, at
least in the sense that we all stand on the shoulders of
those who have gone before; and most of them have been
achieved by using normative principles of inference
often conspicuously absent from everyday life. Most
importantly, there is no logical contradiction between
the assertion that people can be very impressively
intelligent on some occasions or in some domains and the
assertion that they can make howling inferential errors
on other occasions or in other domains.'
R. Nisbett and L. Ross (1980)
Human Inference: Strategies and Shortcomings of Social
Judgment
REFERENCES:
Inductive Reasoning as Heuristics
Agnoli F & Krantz D. H. Suppressing Natural Heuristics by Formal
Instruction: The Case of the Conjunction Fallacy Cognitive Psychology
21, 515-550, 1989
Cooke R M Experts In Uncertainty Opinion and Subjective
Probability in Science Oxford University Press 1991
Derthick M Mundane Reasoning by Settling on a Plausible Model
Artificial Intelligence 46,1990,107-157
Eddy D M Probabilistic Reasoning in Clinical Medicine: Problems
and Opportunities In Kahneman, Tversky and Solvic (Eds) Judgment Under
Uncertainty: Heuristics and Biases Cambridge University Press 1982
Fong G T, Lurigio A J & Stalans L J Improving Probation Decisions
Through Statistical Training Criminal Justice and
Behavior,17,3,1990,370-388
Fong G T & Nisbett R E Immediate and delayed transfer of training
effects in Statistical reasoning. Journal of Experimental Psychology
General; 1991,120(1) 34-45
Gardner H The Mind's New Science: A History of the Cognitive
Revolution Basica Books 1987
Gladstone R Teaching for transfer versus formal discipline
American-Psychologist; 1989 Aug Vol 44(8) 1159
Gluck M A & Bower G. H. From conditioning to category learning: An
adaptive network model. Journal of Experimental Psychology General
(1988) Sep Vol 117(3) 227-247
Gluck M A & Bower G H Component and pattern information in
adaptive networks. Journal of Experimental Psychology General; (1990)
Mar Vol 119 (1) 105-9
Hecht-Nielsen R NEUROCOMPUTING Addison Wesley 1990
Holland J H, Holyoak K J, Nisbett R E & Thagard P R Induction:
Processes of Inference, Learning, and Discovery Bradford Books:MIT
Press 1986
Lehman D R & & Nisbett R E A Longitudinal Study of the Effects of
Undergraduate Training on Reasoning. Developmental Psychology,1990,26,
6,952-960
Nisbett R E & Wilson T D Telling more than we can know: Verbal
reports on mental processes Psychological-Review; 1977 Mar Vol 84(3)
231-259
Nisbett R E & Ross L Human Inference: Strategies and Shortcomings
of Social Judgment Century Psychology Series, Prentice-Hall (1980)
Nisbett R E & Krantz D H The Use of Statistical Heuristics in
Everyday Inductive Reasoning Psychological Review, 1983, 90, 4, 339-
363
Nisbett R E, Fong G T, Lehman D R & Cheng P W Teaching Reasoning
Science v238, 1987 pp.625-631
Ploger D & Wilson M Statistical reasoning: What is the role of
inferential rule training? Comment on Fong and Nisbett. Journal of
Experimental Psychology General; 1991 Jun Vol 120(2) 213-214
Reeves L M & Weisberg R W Abstract versus concrete information as
the basis for transfer in problem solving: Comment on Fong and Nisbett
(1991). Journal of Experimental Psychology General; 1993 Mar Vol
122(1) 125-128
Rescorla R A & Wagner A R A Theory of Classical Conditioning:
variations in the effectiveness of reinforcement and Nonreinforcement.
In Classical Conditioning II: Current Theory and Research (Black &
Prokasy) pp. 64-99. Appleton Century Crofts 1971.
Rescorla R A Pavlovian Conditioning: It's Not What You Think It
Is. American Psychologist, March 1988.
Ross L & Nisbett R E The Person and The Situation: Perspectives of
Social Psychology McGraw Hill 1991
Shafir E & Tversky A Thinking Through Uncertainty:
Nonconsequential Reasoning and Choice Cognitive Psychology 24,449-474,
1992
Smith E E, Langston C & Nisbett R E The case for rules in
reasoning. Cognitive Science; 1992 Jan-Mar Vol 16(1) 1-40
Sutherland S IRRATIONALITY: The Enemy Within Constable: London 1992
Tversky A & Kahneman D Extensional Versus Intuitive Reasoning: The
Conjunction Fallacy in Probability Judgment Psychological Review
v90(4) 1983
Wason W C Reasoning in New Horizons in Psychology, Penguin Books
1966
Wason W C & Johnson-Laird P Psychology of Reasoning
London:Batsford 1972
C. On Clinical vs. Actuarial Judgment
Arkes H R & Hammond K R Judgment and Decision Making: An
interdisciplinary reader Cambridge University Press 1986
Dawes R.M The Robust beauty of improper linear models in decision
making American Psychologist, 1979 34,571-582
Dawes R M Rational Choice in an Uncertain World Orlando: Harcourt,
Brace, Jovanovich 1988
Dawes R M, Faust D & Meehl P E Clinical Versus Actuarial Judgement
Science v243, pp 1668-1674 1989
Elstein A S Clinical judgment: Psychological research and medical
practice. Science; 1976 Nov Vol 194(4266) 696-700
Einhorn H J & Hogarth R M Behavioral decision theory: Processes of
judgment and choice Annual Review of Psychology (1981), 32, 53-88
Faust D Data integration in legal evaluations: Can clinicians
deliver on their premises? Behavioral Sciences and the Law; 1989 Fal
Vol 7(4) 469-483
Gigerenzer G How to Make Cognitive Illusions Disappear: Beyond
"Heuristics and Biases" in European Review of Social Psychology,
Volume 2 eds W Stroebe & M Hewstone 1991, Ch 4 pp. 83-115
Goldberg L R Simple models or simple processes? Some research on
clinical judgments American Psychologist,1968,23(7) p.483-496
Kahneman D, Slovic P & Tversky A Judgment Under Uncertainty:
Heuristics and Biases Cambridge University Press 1982
Lundberg G A Case Studies vs. Statistical Methods - An Issue
Based on Misunderstanding. Sociometry v4 pp379-83 1941
Meehl P E Clinical vs. Statistical Prediction: A Theoretical
Analysis and a Review of the Evidence University of Minnesota Press,
Minneapolis. 1954
Meehl P E When Shall We Use Our Heads Instead of the Formula?
PSYCHODIAGNOSIS: Collected Papers 1971
Meehl P E Causes and effects of my disturbing little book J
Person. Assess. 50,370-5,1986
Oskamp S Overconfidence in case-study judgments J. Consult.
Psychol. (1965), 29, 261-265
Sarbin T R Clinical Psychology - Art or Science? Psychometrica
v6 pp391-400 (1941)
David Longley
In article <6l59js$lpr$3...@cantuc.canterbury.ac.nz>
mat...@math.canterbury.ac.nz "Bill Taylor" writes:
> sfri...@americasttv.com (Stanley Friesen) writes:
>
> |> > Mssr. Longley refers to this as folk psychology
> |> >where folks make up stories about folk's psychology and
> |> >then proceed as if those stories were documented fact.
> |>
> |> Yet he seems to never actually give any concrete examples of this.
>
>
> HEAR HEAR!!
>
> I too, would like to see this glaring lacuna filled!
> As would many others here, it seems likely.
>
> So, if the Longley-bot is still scanning this thread,
> please accede to this request.
>
>
> NOTE:- We would like to see a BRIEF post with such examples...
>
> ...in your OWN words;
But many of them are *not* my words and I have written
extensively on the problem of intensional contexts, namely
indirect quotation. Your frequent demands for paraphrase reveal
that you have totally failed to grasp what's being said.
> ...with no excerpts from or references to any other literature;
Why?
> ...taking less than 2 pages, but more than a paragraph;
Ridiculous - the thesis is much too substantial.
> ...if you would be so kind.
>
>
> Until there is money there, the mouth is hollow.
>
You should stop trying to *translate* the material into your
already familiar folk psychology, and silly platitudes such as
the above should be the first to go.
David Longley
> In article <6l59js$lpr$3...@cantuc.canterbury.ac.nz>
> mat...@math.canterbury.ac.nz "Bill Taylor" writes:
>
> > sfri...@americasttv.com (Stanley Friesen) writes:
> >
> > |> > Mssr. Longley refers to this as folk psychology
> > |> >where folks make up stories about folk's psychology and
> > |> >then proceed as if those stories were documented fact.
> > |>
> > |> Yet he seems to never actually give any concrete examples of this.
I have forwarded material by Tversky & Kahneman, and Ross &
Anderson (converted to HTML) to Friesen, Taylor and Russell.
These are from the reference list provided in my earlier post.
Attribution Theory and Cognitive Psychology are vast areas of
research in psychology, and I believe I have given a sound
account of their place within the discipline, *and* most
importantly, provided a new perspective on that place -
particularly for those with interests in AI and applied work in
general.
The whole point of posting to this newsgroup from spring 1995
onwards has been because I judged the perspective I have been
formulating, both important and original. Conclusions derived
from having taught graduate psychologists an MSc course over 10
years during their early years in an applied field which
crucially needs to put this work into practice.
see http://www.longley.demon.co.uk/Frag.htm and other papers at
that site.
I am looking for contributions/comments from major researchers in
the field of AI, and am grateful for those who have already done
so, both publicly and privately over the past three years.
For those who find this material unfamiliar, I'm not surprised -
and can only ask that they consult the literature I have
provided - I did not have them primarily as my intended audience.
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>> In article <6l59js$lpr$3...@cantuc.canterbury.ac.nz>
>> mat...@math.canterbury.ac.nz "Bill Taylor" writes:
>> > sfri...@americasttv.com (Stanley Friesen) writes:
>> > |> > Mssr. Longley refers to this as folk psychology
>> > |> >where folks make up stories about folk's psychology and
>> > |> >then proceed as if those stories were documented fact.
>> > |>
>> > |> Yet he seems to never actually give any concrete examples of this.
>I have forwarded material by Tversky & Kahneman, and Ross &
>Anderson (converted to HTML) to Friesen, Taylor and Russell.
>These are from the reference list provided in my earlier post.
I don't know what will be the reaction of Friesen and Russell. But I
would almost bet that Bill Taylor is laughing his head off. Bill has
already made the tentative assessment that Longley is bankrupt of
intelligent ideas. And now Longley, trying to prove otherwise, has
actually sent further evidence confirming Bill's assessment.
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>The whole point of posting to this newsgroup from spring 1995
>onwards has been because I judged the perspective I have been
>formulating, both important and original. Conclusions derived
>from having taught graduate psychologists an MSc course over 10
>years during their early years in an applied field which
>crucially needs to put this work into practice.
>see http://www.longley.demon.co.uk/Frag.htm and other papers at
>that site.
>I am looking for contributions/comments from major researchers in
>the field of AI, and am grateful for those who have already done
>so, both publicly and privately over the past three years.
Most of the contributions in this forum have been in the form of
telling you that you are nuts. However, you have not seemed at all
grateful to that kind of contribution. I guess that means you have a
selective filter -- you only consider evidence which agrees with your
preconceived doctrines.
David Longley
Let's be clear on one thing here. I have gone to extreme lengths
to try to explain the nature of enlightened empiricism and some
empirical research evidence which, as a professional
psychologist, I consider to be relevant to what I refer to as
"the extensional stance". Our knowledge is public, acquired as
acquisition of a specialised language and what I have been doing
over recent years is drawing attention to scientific evidence -
applying that to what I consider to be an important area of
application.
Rickert is, (who in my view hasn't much of a grasp of anything
which has a bearing on any of this) like all too many in this
newsgroup, is basically an unenlightened solipsist or "folk
psychologist" who lacks the formal education in the relevant
disciplines to recognise what this means. The fact that being a
folk psychologist seems socially undesirable, understandably
prompts him to respond the way that he does, but in doing so, he
just confirms exactly what I have said of him - he shows no grasp
at all of what is being said. That he is inclined to behave this
way is covered by the very literature I keep referring him, and
others to.
There is no corrective to this except long and apposite study,
something Rickert shows little inclination towards or aptitude
for. He knows it all already .... somehow .... and it is this
"arrogance" which really prevents him from learning anything
worthwhile.
David Longley
This really is very stupid - as anyone who knows any of the
relevant history will tell you. You, and others like you should
make the effort to listen .... it may all seem "nuts", but that's
only because you are thinking as a "folk psychologist".
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>In article <6l96ue$g...@ux.cs.niu.edu> ric...@cs.niu.edu "Neil Rickert" writes:
>Let's be clear on one thing here. I have gone to extreme lengths
>to try to explain the nature of enlightened empiricism and some
>empirical research evidence which, as a professional
>psychologist, I consider to be relevant to what I refer to as
>"the extensional stance".
You have never explained anything. But you have gone to great
lengths to demonstrate your complete inability to explain. Thus, in
place of explanation you provide regurgitated quotations.
>Rickert is, (who in my view hasn't much of a grasp of anything
>which has a bearing on any of this) like all too many in this
>newsgroup, is basically an unenlightened solipsist or "folk
>psychologist" who lacks the formal education in the relevant
>disciplines to recognise what this means.
Ah, yes. I do not have the certification in Longley's theology.
Therefore I am to be criticized for my doctrinal heresy. But what
Longley mistakes for heresy is actually atheism. I disbelieve the
gospel according to Longley.
The general view among scientists and mathematicians, is that science
divides itself into the hard sciences and the soft sciences.
Psychology is one of the soft sciences. The generally held view is
that soft scientists understand how to do science barely, if at all.
Thus what comes out of the soft sciences is viewed with great
skepticism.
When a particular soft scientist issues proclaimations claiming that
the soft scientists have convincingly demonstrated that the methods
used effectively by the hard scientists are bogus and irrational, we
know that said soft scientist is at the lunatic edge of the softness
spectrum.
David Longley
> Da...@longley.demon.co.uk (David Longley) writes:
> >In article <6l96ue$g...@ux.cs.niu.edu> ric...@cs.niu.edu "Neil Rickert" writes:>
> >Let's be clear on one thing here. I have gone to extreme lengths
> >to try to explain the nature of enlightened empiricism and some
> >empirical research evidence which, as a professional
> >psychologist, I consider to be relevant to what I refer to as
> >"the extensional stance".
>
> You have never explained anything. But you have gone to great
> lengths to demonstrate your complete inability to explain. Thus, in
> place of explanation you provide regurgitated quotations.
No..I have gone to no great lengths to TEACH.. there's a
difference, and there's a reason why one expects some
demonstration of competence in a field before expecting students
to be able to benefit. I'm just not interesed in teaching in this
forum.
>
> >Rickert is, (who in my view hasn't much of a grasp of anything
> >which has a bearing on any of this) like all too many in this
> >newsgroup, is basically an unenlightened solipsist or "folk
> >psychologist" who lacks the formal education in the relevant
> >disciplines to recognise what this means.
>
> Ah, yes. I do not have the certification in Longley's theology.
> Therefore I am to be criticized for my doctrinal heresy. But what
> Longley mistakes for heresy is actually atheism. I disbelieve the
> gospel according to Longley.
No - form what I have read here over the years you just don't
have the requisite backgroup to either contribute either way.
Since you insist on claiming otherwise I have gone to some
lengths, fruitlesslessly it seems to point you in a direction
which might help you. In response I just get the sort of rubbish
you have just posted above and in previous posts.
You do not have an adeuate grasp of the relevant fields to
comment - it's time you grasped that.
>
> The general view among scientists and mathematicians, is that science
> divides itself into the hard sciences and the soft sciences.
I'm really not interested in turning this into a discussion about
soft vs. hard science - I have made it quite clear what I take to
be important in this regard, and have made it central to all I
have had to say.
Instead of writing silly remarks about work you clearly do not
understand, you would do better to spend some time reading
through the research and learning to appreciate how well it
describe what people actually get up to. As it is you have a very
unrepresentative conception of natural human ability and an even
more unrepresentative conception of what is "nuts"...
I've seen too many undergraduates express views which you hold
and express here on the net (to take what you say as anything but
a sign of ignorance). Invariably such views fade as students
become more knowledgeable.
I've been good enough to refer those who don't know it to the
primary literature, and I've also been kind enough to quote it
verbatim to facilitate communication. Your arrogance is an
impediment to your correcting your ignorance in this field - but
as I say above - I am *not* fundamentally interested in educating
you or others of the same ilk. My target audience is elsewhere,
and I really do expect people like yourself to behave the way you
do.....It's predicatble..
Nikolaus Strater
bla bla testing bla bla
Bill Taylor
Da...@longley.demon.co.uk (David Longley) writes:
|> > I too, would like to see this glaring lacuna filled!
|>
|> The lacuna is in your knowledge - *you* fill it yourself. I'm not
|> prepared to run crash courses in psychology and AI
In other words, Longley hasn't got any examples, and can't think of any.
We thought as much. All piss and wind.
He farts good, but hasn't got the shit to back it up!
-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
You are what you remember.
-------------------------------------------------------------------------------
Oliver Sparrow
Da...@longley.demon.co.uk (David Longley) wrote:
"" No..I have gone to no great lengths to TEACH.. there's a
"" difference, and there's a reason why one expects some
"" demonstration of competence in a field before expecting students
"" to be able to benefit. I'm just not interesed in teaching in this
"" forum.
Breathtaking arrogance.
______________________________________
Oliver Sparrow
David Longley
In article <35809479...@news.demon.co.uk>
oh...@chatham.demon.co.uk "Oliver Sparrow" writes:
"Breathtaking" - maybe (but make sure you respond to this emotion
with care). Arrogance - no. There is a substantial point being
made here, and it's one which will not be properly grasped unless
readers make a serious effort to cover the literature I have
referred to. The conclusions can not be readily paraphrased and
protestations from *within* folk psychology on grounds of what
amounts to little more than political correctness are basically
naive and misguided.
There are assumptions shaping attitudes here, and you will not
become aware of these assumptions until you learn how they
operate.
Read the papers.
David Longley
In article <6lg388$m2h$1...@cantuc.canterbury.ac.nz>
mat...@math.canterbury.ac.nz "Bill Taylor" writes:
> Da...@longley.demon.co.uk (David Longley) writes:
>
> |> > I too, would like to see this glaring lacuna filled!
> |>
> |> The lacuna is in your knowledge - *you* fill it yourself. I'm not
> |> prepared to run crash courses in psychology and AI
>
> In other words, Longley hasn't got any examples, and can't think of any.
>
> We thought as much. All piss and wind.
>
> He farts good, but hasn't got the shit to back it up!
>
> -------------------------------------------------------------------------------
> Bill Taylor W.Ta...@math.canterbury.ac.nz
> -------------------------------------------------------------------------------
> You are what you remember.
> -------------------------------------------------------------------------------
>
In other words, Taylor can't or won't read what's provided. What he
thinks is a reflection of the little he knows - he should study his
own sig line more carefully.
David Longley
In article <357ce79b....@pop.americasttv.com>
sfri...@americasttv.com "Stanley Friesen" writes:
> Surely you could at least post a short *summary*, or abstract, as
> is standard practice most refereed journals. So some of the
> details are left out - big deal. Almost anything can be
> summarized in two pages.
I've already done that - but you apparently can't understand it.
>
> [P.S. thanks for the e-mail, I will try to read the material when
> I get the time].
>
Well, hopefully *after* you have read the material I sent you'll
be in a better position to understand the summaries I *have*
provided.
Whether or not you feel inspired to read my web site material
matters little to me. I was hoping to hear the comments of those
already familiar with the cited literature.
--
David Longley
Seth Russell
David Longley wrote:
> I have forwarded material by Tversky & Kahneman, and Ross &
> Anderson (converted to HTML) to Friesen, Taylor and Russell.
Ok I have reviewed the materials and I find that these articles
certainly make a impelling case showing where human "common sense"
falls into error. What are not represented in these studies is the
other half of the issue: where "common sense" out performs normative
procedures and where "common sense" is the only strategy available.
It is as if we are looking at a study of a new drug showing the side
effects, but the part of the study showing its positive effects is
missing. My question to Longley remains: If you say that we should
eschew the use of common sense for AI, are there *no* qualifications
to that statement?
I have found a articulate voice for Longley's view in Keith Devlin's
book Good bye, Descartes: The end of Logic and the search for a new
Cosmology of the Mind.
http://www.amazon.com/exec/obidos/ISBN=0471142166 . Devlin paints the
same picture but provides an understandable path for the future of AI.
--
Seth
See "Bozo's Conjecture" at http://www.clickshop.com/ai/conjecture.htm
And then on to the AI Jump List ...
Stanley Friesen
On Fri, 05 Jun 98 08:05:45 GMT, Da...@longley.demon.co.uk# wrote:
>In article <6l59js$lpr$3...@cantuc.canterbury.ac.nz>
> mat...@math.canterbury.ac.nz "Bill Taylor" writes:
>> I too, would like to see this glaring lacuna filled!
>> As would many others here, it seems likely.
>>
>> So, if the Longley-bot is still scanning this thread,
>> please accede to this request.
>>
>>
>> NOTE:- We would like to see a BRIEF post with such examples...
>>
>> ...in your OWN words;
>
>But many of them are *not* my words and I have written
>extensively on the problem of intensional contexts, namely
>indirect quotation. Your frequent demands for paraphrase reveal
>that you have totally failed to grasp what's being said.
Well, actually, we do find what you say somewhat
incomprehensible. We are asking you to try to help clarify what
you mean for us.
>
>> ...with no excerpts from or references to any other literature;
>
>Why?
So that the summary is all present in the posting.
Until we have a better udea of what you are trying to say, we
really do not have any reason to use our time looking at your
references, or even your Web page.
Convince us that you have something worthwhile to say. and we may
go further.
>
>> ...taking less than 2 pages, but more than a paragraph;
>
>Ridiculous - the thesis is much too substantial.
>
Surely you could at least post a short *summary*, or abstract, as
is standard practice most refereed journals. So some of the
details are left out - big deal. Almost anything can be
summarized in two pages.
[P.S. thanks for the e-mail, I will try to read the material when
Stanley Friesen
On Mon, 08 Jun 1998 14:50:23 -0700, Seth Russell
<seth...@clickshop.com> wrote:
>David Longley wrote:
>
>> I have forwarded material by Tversky & Kahneman, and Ross &
>> Anderson (converted to HTML) to Friesen, Taylor and Russell.
>
>Ok I have reviewed the materials and I find that these articles
>certainly make a impelling case showing where human "common sense"
>falls into error.
I have just made a quick review of these articles. I find
nothing in them that I consider particularly controversial, or
even surprising. I have never in this forum denied the fact that
heuristics are often prone to error.
[In fact, I believe I am familiar with some of the studies
mentioned in the Tversky & Kahneman paper - they certainly are
very similar to some studies I have previously been exposed to].
> What are not represented in these studies is the
>other half of the issue: where "common sense" out performs normative
>procedures and where "common sense" is the only strategy available.
Indeed. In fact the studies cited appear to have been
specifically *designed* to bring out the limitations of heuristic
estimation procedures. They are in a sense "unfair", as they
load the deck against the heuristics.
I find it interesting that, even with this approach, Tversky &
Kahneman end their paper with the following:
"These heuristics are highly economical and usually effective,
but they lead to systematic and predictable errors. A better
understanding of these heuristics and of the biases towhich they
lead could improve judgments and decisions in situations of
uncertainty."
I particularly draw attention to the phrase "highly economical
and usually effective". Hmm, "usually effective". That sounds
rather like what I have been trying to say! I get the impression
that these two authors actually agree with you and me, not with
Longley.
This reminds me of a review article in _Science News_ a year or
two ago. It covered the changes in the research on heuristics in
recent years. It pointed out that in the past research has
concentrated on the limitations of heuristics, and only recently
has much research been done on their capabilities. It covered
quite a number of studies, on both sides. A very interesting
article indeed. It presented a number of studies showing
circumstances in which heuristics outperform "normative
procedures".
[P.S. The HTML documents look like they were scanned in using an
OCR system without any proofreading: they were full of
run-together words, strange spellings, and odd combinations of
characters - all of which I see when I scan documents, prior to
proofreading. Stuff such as "juagmenl ullu~r ull~l~alII~y" and
"InsufJicient" look like what comes out of OmniPage Pro raw].
David Longley
In article <357d1662....@pop.americasttv.com>
sfri...@americasttv.com "Stanley Friesen" writes:
> I particularly draw attention to the phrase "highly economical
> and usually effective". Hmm, "usually effective". That sounds
> rather like what I have been trying to say! I get the impression
> that these two authors actually agree with you and me, not with
> Longley.
>
This is silly - by posting the above you are just confirming that
you don't understand 1) what the work is all about and 2) the
implications *I* have drawn for AI.
The biases which the representativeness and availability
heuristics are violations of normative principles such as the law
of large numbers and the conjunction rule. Even Kahneman and
Tversky title the paper on the conjunction fallacy "extensional
reasoning and the conjunction fallacy"
Spend some more time studying this work. If you want something
which drives it home like a sledge-hammer - read Sutherland's
book "Irrationality - The Enemy Within".
Effective procedures and the extensional stance are acquired
skills (we can leave aside the often serendipidous contexts of
discovery) - and these are only achieved through formal, in
context training. They are relatively rare in terms of actual
practice, but some of the basic effective principles have become
almost mandatory as a function of cultural development (in most
parts of the world).
The implicit function approximation and constraint satisfaction
enabled by natural neural nets just do not suffice for what we
refer to as intelligent behaviour - what's required is a social
system of rules and relations, and a set of contingencies which
operate to ensure that the nets operate within acceptable
parameters. This is often such a demanding set of contingencies
that we turn wherever possible to direct implementation of truly
effective procedures - ie ones which have known parameters of
operation and fault estimation.
We depend on the extensional stance, and try to make people
operate *according to it* - with only a modicum of success. For
these reasons the early aspirations of GOFAI which naively took
human cognitive performance (actually couched at the time in
information theoretic terms), is and continues to be misguided.
I have suggested that the empirical results require us to accept
two important conclusions. 1) that empirical psychology provides
a descriptive account of how natural behaviour operates (the fact
that formal training does not transfer well outside its formal
training context and domain of application is worth noting here)
and 2) that there is therefore a demand for the application of
science and technology in all areas where natural judgement is
normally executed. Because this is usually conceived in terms of
judgement, rather than straightforward technology per se, it is
inevitable perhaps that this is seen my some as an application of
AI (cf. Knowledge Engineering). But all applications of Risk
Assessment and actuarial analysis in all its guises really brings
home the wider conclusion that AI just *is* science and
technology.
Only administrative convenience accounts for it having it's
current limited range of application. For practical implications
of what I have been saying, see the context of application at:
http://www.longley.demon.co.uk/Frag.htm and other papers there.
The focus is on natural vs extensionally driven decision making.
PS thanks for the OCR tip - I'll check the settings.
--
David Longley (check end reply line #)
Longley Consulting London, UK
JFZeigler
You know, I think I may have figured out what Longley's
difficulty is.
Da...@longley.demon.co.uk writes:
>This is silly - by posting the above you are just confirming that
>you don't understand 1) what the work is all about and 2) the
>implications *I* have drawn for AI.
Now, I'm not equipped to evaluate Longley's work. When
it comes to psychology and the philosophy of AI I'm just a
delittante. Most of what he's posted is unsurprising to me, but
perhaps that's just because I'm unfamiliar with the literature
he quotes.
On the other hand, I can see why he isn't getting the response
he appears to expect from Usenet.
Usenet is chock-full of folks who think they've discovered
The Secret Truth behind some discipline. A certain self-
absorption ("only *I* have these insights") and a desire to be
noticed by the experts in the field ("I want to hear from the
people who are equipped to understand me") are common
features of this species. So is a readiness to dismiss all criticism
as driven by ignorance.
Anyone who roams through Usenet for a while will encounter
this phenomenon many times. He will develop a heuristic to
filter out further examples, so as to avoid wasting effort. To the
extent that Longley appears to fit this description, he isn't going
to get the kind of attention he wants. . .
----------
Jon F. Zeigler: Mathematician, amateur historian, science fiction fan,
freelance writer, occasional scribbler of bad poetry
JFZe...@aol.com
"Never speak for others. You can get in enough trouble speaking for yourself."
Seth Russell
David Longley [in his own words] wrote:
> The implicit function approximation and constraint satisfaction
> enabled by natural neural nets just do not suffice for what we
> refer to as intelligent behaviour - what's required is a social
> system of rules and relations, and a set of contingencies which
> operate to ensure that the nets operate within acceptable
> parameters. This is often such a demanding set of contingencies
> that we turn wherever possible to direct implementation of truly
> effective procedures - ie ones which have known parameters of
> operation and fault estimation.
Yes, the shock value is from a narrow perspective that is totally
true. And from my long career of applying computers to solve
problems for my clients, I can personally vouch for the fact that
"we turn wherever possible to direct implementation of truly
effective procedures" i.e. the primary use of the computer is to
control a situation. I would like people on this news group to
understand the paragraph above very carefully. This is an exact
prescription for enslaving the human race.
I am glad that we finally got the longley.bot to clearly state its
agenda. Now I suggest we ignore it. Perhaps we can relegate the
"Extensional Stance" and what it would imply for our future to a
footnote on failed 20th century elitist fascism.
David Longley
In article <357D3959...@clickshop.com>
seth...@clickshop.com "Seth Russell" writes:
A typically naive response to behaviourism - with predictable
consequences.
David Longley
In article <199806091327...@ladder01.news.aol.com>
jfze...@aol.com "JFZeigler" writes:
> You know, I think I may have figured out what Longley's
> difficulty is.
*I* don't have the difficulty - *you* and a few others seem to be
the ones having the difficulty by your own statements!
The fact that it is not a problem of exposition can be seen from
the fact that colleagues reviewing the work in "Fragment.zip" and
in the editor's introduction of:
http://www.longley.demon.co.uk/Frag.htm
have no such problem - nor do others having read the related
papers at the above website.
I expect to have the same sort of difficulties in areas in which
I have little expertise in, so why don't you?
As it is, you've just provided another example of the fallibility
fo folk psychological attribution - failing to *fully* appreciate
the implications of your own acknowledged ignorance.
JFZeigler
Da...@longley.demon.co.uk writes:
>As it is, you've just provided another example of the fallibility
>fo folk psychological attribution - failing to *fully* appreciate
>the implications of your own acknowledged ignorance.
<* chuckle *>. I admit to being quite ignorant of the arcana
you and your colleagues work with. I have no interest in
trying to debate the merits of your treatment of those arcana,
given my ignorance of the subject. For all I know, you really
have discovered a new and interesting insight.
I am *not* ignorant of the behavior of people on the Internet,
however -- having spent many years observing such behavior.
I'll restate, since I was apparently too polite to get my point
across last time:
There are many crackpots on the Internet. Many of them
behave in ways that are strongly reminiscent of your behavior.
Hence a casual observer might well conclude that you are a
crackpot and not bother to study your work more closely.
Anyone who wishes to survive on the Internet will develop
heuristics that help them ignore or dismiss crackpots.
If you want people to study your work more closely, cease
behaving like a crackpot. And, I might add, don't bother
trying to advertise your results on the Internet. Refereed
journals are probably going to be more effective in reaching
your intended audience. Of course, that audience will also
be capable of informed criticism of your ideas -- but that
shouldn't worry you.
David Longley
"In the 1970's, Cognitive Science was formulated in
terms of Putnam's doctrine of functionalism, in which
mental states are the functional states of an abstract
digital computer, thinking is abstract symbol
manipulation as in the operation of a computer program,
and the mind's symbols get their meaning by denoting
things in the world. Functionalism has since become the
mainstream doctrine within the philosophy of mind. But
in the mid-1970's, overwhelming empirical evidence
against functionalism began to pile up, and in the late
1970's, Putnam himself found functionalism to be
logically incoherent. In this volume, Putnam brilliantly
reveals the philosophical fallacies in the doctrine he
founded, showing why functionalism must fail as a
philosophy of mind. The fall of functionalism has major
consequences for generative linguistics, artificial
intelligence, and cognitive and developmental
psychology".
Lakoff (1988).
Which is a vindication of what Skinner long argued.
'Cognitive psychology is frequently presented as a
revolt against behaviorism, but it is not a revolt, it
is a retreat. Everyday English is full of terms
derived from ancient explanations of human behavior.
We spoke that language when we were young. When we
went out into the world and became psychologists, we
learned to speak in other ways but made mistakes for
which we were punished. But now we can relax.
Cognitive psychology is Old Home Week. We are back
among friends speaking the language we spoke when we
were growing up. We can talk about love and will and
ideas and memories and feelings and states of mind, and
no one will ask us what we mean; no one will raise an
eyebrow.'
('The Shame of American Education')
B.F. Skinner 1987
'Regardless of how much we stand to gain from supposing
that human behavior is the proper subject matter of a
science, no one who is a product of Western civilization
can do so without a struggle. We simply do not want such
a science.'
B F Skinner (1953)
Can Science Help? - The Threat to Freedom
(in Science and Human Behavior p.7)
David Longley
In article <357e1a93....@pop.americasttv.com>
sfri...@americasttv.com "Stanley Friesen" writes:
> On Tue, 09 Jun 98 08:00:22 GMT, Da...@longley.demon.co.uk# wrote:
>
> >In article <357d1662....@pop.americasttv.com>
> > sfri...@americasttv.com "Stanley Friesen" writes:
> >
> >> I particularly draw attention to the phrase "highly economical
> >> and usually effective". Hmm, "usually effective". That sounds
> >> rather like what I have been trying to say! I get the impression
> >> that these two authors actually agree with you and me, not with
> >> Longley.
> >>
> >This is silly - by posting the above you are just confirming that
> >you don't understand 1) what the work is all about
>
> I will admit that we *disagree* about the significance of this
> work.
>
> > and 2) the
> >implications *I* have drawn for AI.
> >
> >The biases which the representativeness and availability
> >heuristics are violations of normative principles such as the law
> >of large numbers and the conjunction rule. Even Kahneman and
> >Tversky title the paper on the conjunction fallacy "extensional
> >reasoning and the conjunction fallacy"
>
> Yep. I fully concur and agree. Heuristics are not statistically
> valid procedures, and can often be fallable.
>
> So what? Representativeness and availability are also very
> useful, even "usually effective" is producing *quick*, *raw*
> estimates of various important facets of life. This is
> especially true in real-world situations, as opposed to the
> artificial situations set-up in the laboratory to test their
> limiations.
> >
> >Spend some more time studying this work. If you want something
> >which drives it home like a sledge-hammer - read Sutherland's
> >book "Irrationality - The Enemy Within".
>
> I doubt there is anything in this book I really disagree with!
> Yes, heuristics can produce erroneous conclusions, and, under
> some circumstances, can even lead to harmful biases.
>
> BFD!
>
> This is *not* my point. I in no way contend that these results
> are incorrect. What I disagree with is the *conclusions* which
> *you* draw from these results.
No..the conclusins are *yours* - attributed to me through lack of
carefully reading of what I have said. You should study carefully
what I have said about the nature of intensional contexts.
>
> You keep sending me mere evidence that heuristics are fallable.
> Since I do not disagree with that, this is unlikely to influence
> me particularly. It is the step fraom fallability to uselessness
> that I object to, especially in the light of good *laboratory*
> evidence that heuristics actually *are* useful in many
> circumstances. You seem to be ignoring this evidence (it is too
> bad that I cannot keep old issues of _Science News_, or I could
> give you some good citations on that).
>
>
The alternatives and conclusions you keep drawing are your own,
not mine. You should look carefully into what I have actually
written and not confabulate it with your own incorrect
inferences and paraphrases.
Of course they're useful as a modus vivendi - but they are *not*
something which AI systems should seek to incporporate - the
point is that in all of the professions we invest considerable
efforts in trying to replace such heuristics with actuarially or
logically based systems - ie procedures developed from the
extensional stance.
There is more to what I have written than you are giving credit
to.
David Longley
In article <199806091633...@ladder01.news.aol.com>
jfze...@aol.com "JFZeigler" writes:
> Da...@longley.demon.co.uk writes:
> >As it is, you've just provided another example of the fallibility
> >fo folk psychological attribution - failing to *fully* appreciate
> >the implications of your own acknowledged ignorance.
>
> <* chuckle *>. I admit to being quite ignorant of the arcana
> you and your colleagues work with. I have no interest in
> trying to debate the merits of your treatment of those arcana,
> given my ignorance of the subject. For all I know, you really
> have discovered a new and interesting insight.
>
> I am *not* ignorant of the behavior of people on the Internet,
> however -- having spent many years observing such behavior.
>
> I'll restate, since I was apparently too polite to get my point
> across last time:
If you (and others of your ilk) bothered to read what's provided
rather than jumping to conclusions about intentions and other
matters (particularly when you begin by acknowledging your
ignorance of the relevant material), you might learn something.
David Longley
From the preface -"Irrationality: The Enemy Within"
S Sutherland 1992
"Pace Aristotle, it can be argued that irrational
behaviour is the norm not the exception. In order to
demonstrate this, I have provided many startling
examples of irrationality in everyday life and in the
activities of the pro- fessions. It turns out that the
decisions of doctors, generals, engineers, judges,
business men and others are no more rational than those
made by you or me though their effects are often more
calamitous.
However, the real proof of the prevalence of
irrationality comes from the massive amount of research
on the topic undertaken over the last thirty years by
psychologists. Their discoveries- unlike those of
cosmologists are as yet scarcely known to the general
public. Although I have not myself worked directly on
the topic, I became fascinated by the ingenuity of their
experiments and by the light they throw on the workings
of the mind. This book integrates the many factors that
have been shown to cause irrational behaviour, including
social and emotional biases as well as the many quirks
of thought pro- duced by such failings as not taking
account of negative cases or being too swayed by what
first comes to mind. Many of the experimental findings
are so surprising that the reader's credulity may well
be strained: almost all of them have, however, been
replicated many times. To stave off the sceptical
reader, there is a rather daunting list of sources,
which needs to be consulted only by those lacking faith
in my veracity or desiring to pursue specific issues in
more detail.
I have tried to make clear to the layman work that is
often hard to follow in the technical journals; for the
most part I have avoided mathematical and statistical
concepts, but of necessity a few elementary ones are
introduced and explained towards the end of the book."
Bill Taylor
sfri...@americasttv.com (Stanley Friesen) writes:
|> Almost anything can be summarized in two pages.
EXACTLY!
As Ernest Rutherford was reputed to have said:-
"Anyone who can't explain his work to the cleaning lady is a charlatan."
A bit extreme, perhaps, but to the point. No doubt being a classic
piece of folk psychology, this will cut no ice with Longwind.
But yes, pretty much anything can be reasonably summarized in two pages.
But David Longquotes isn't going to do it because...
(a) he's too lazy (by his own admission)
(b) he can't (by general conclusion)
(c) he's been asked many times already and hasn't (by observation)
(d) he's just too damn bloody-minded (by reputation)
(e) he's too scared to use his own words and be shown up as an idiot, (by
deduction), when he can hide behind excessively long, boring quotes; so:-
(f)-off to the rest of us!
-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
Faith is what you fall back on when you've lost the argument.
-------------------------------------------------------------------------------
David Longley
In article <6likt9$3p7$6...@cantuc.canterbury.ac.nz>
mat...@math.canterbury.ac.nz "Bill Taylor" writes:
On the other hand, it could just be that you don't know what
you're talking about, haven't read any of the relevant literature
and have far too short an attention span to grasp anything
outside your immediate area of ability.
It makes one wonder why anyone bothers to write so many books and
articles - and why it usually takes several years of study at
degree level to grasp what it all means.
Changed sig line - but again - you would be wise to meditate upon
it.
Jim Balter
David Longley wrote:
> 'Kelley's (1973) 'covariation principle', which is the
> most fundamental assumption of contemporary attribution
> theory and of that theory's characterization of people
> as generally adequate intuitive scientists, is
> essentially an assertion that the layperson can
> recognize and make appropriate inferential use of
> covariation between events. Common sense also suggests
^^^^^^^^^^^^
> that the ability to detect covariation would seem
> necessary for understanding, predicting and controlling
> social experience. The assessment of covariation between
> early symptoms and later manifestations of problems, of
> covariation between particular behavioural strategies
> and subsequent environmental outcomes, and of
> covariation between potential causes and observed
> effects seems critical to people's success in responding
> adaptively to the opportunities and dilemmas of social
> life. Indeed, one might well be tempted to regard
> harmonious social interchange and the general
> effectiveness of personal functioning as EVIDENCE of
> people's capacity to recognize covariation between
> events.
>
> Most research that has dealt with people's abilities to
> recognize and estimate covariation has not been
> flattering to the layperson's abilities. Even Peterson
> and Beach (1967), in their generally positive evaluation
> of 'man as an intuitive statistician' were not very
> optimistic about people's capacity to appreciate
> relationships between variables. In this chapter we
> first review evidence of these shortcomings and then
> attempt to reconcile the apparent contradiction between
> peoples' failures in the laboratory and their successes
> in meeting the demands of everyday social judgement and
> interaction.
It would seem from this that the *common sense* assessment of the
importance of the detection of covariance to the ability to achieve
"harmonious social interchange" is flawed. Yet Longley's prescription
is to stick to covariance and other "actuarial" methods *to the
exclusion* of "layperson" methods.
Of course, as someone without Longley's precise professional background,
it's unlikely that I could reach an understanding merely by reading
a bunch of material, even if it's precisely the material Longley
quotes.
But perhaps Longley could humor me and quote the attempted
reconciliation referred to above, even if that part of the text does not support his thesis.
--
<J Q B>
Jim Balter
David Longley wrote:
>
> In article <6l96ue$g...@ux.cs.niu.edu> ric...@cs.niu.edu "Neil Rickert" writes:
>
> > Da...@longley.demon.co.uk (David Longley) writes:
> > >> mat...@math.canterbury.ac.nz "Bill Taylor" writes:
> >
> > >> > sfri...@americasttv.com (Stanley Friesen) writes:
> >
> > >> > |> >where folks make up stories about folk's psychology and
> > >> > |> >then proceed as if those stories were documented fact.
> > >> > |>
> > >> > |> Yet he seems to never actually give any concrete examples of this.
> >
> > >I have forwarded material by Tversky & Kahneman, and Ross &
> > >Anderson (converted to HTML) to Friesen, Taylor and Russell.
> >
> > I don't know what will be the reaction of Friesen and Russell. But I
> > would almost bet that Bill Taylor is laughing his head off. Bill has
> > already made the tentative assessment that Longley is bankrupt of
> > intelligent ideas. And now Longley, trying to prove otherwise, has
> > actually sent further evidence confirming Bill's assessment.
> >
> Let's be clear on one thing here. I have gone to extreme lengths
> to try to explain the nature of enlightened empiricism and some
> empirical research evidence which, as a professional
> psychologist, I consider to be relevant to what I refer to as
> acquisition of a specialised language and what I have been doing
> over recent years is drawing attention to scientific evidence -
> applying that to what I consider to be an important area of
> application.
>
> Rickert is, (who in my view hasn't much of a grasp of anything
> which has a bearing on any of this) like all too many in this
> newsgroup, is basically an unenlightened solipsist or "folk
> psychologist" who lacks the formal education in the relevant
> folk psychologist seems socially undesirable, understandably
> prompts him to respond the way that he does, but in doing so, he
> just confirms exactly what I have said of him - he shows no grasp
> at all of what is being said. That he is inclined to behave this
> way is covered by the very literature I keep referring him, and
> others to.
>
> There is no corrective to this except long and apposite study,
> something Rickert shows little inclination towards or aptitude
> for. He knows it all already .... somehow .... and it is this
> "arrogance" which really prevents him from learning anything
> worthwhile.
Like the vast majority of Longley's own contributions (in contradistinction to the material he quotes), this is just a string
of unsupported ad hominem claims and folk psychological judgements,
lacking even the relatively low level of intellectual criteria
appropriate to this forum.
--
<J Q B>
Jim Balter
David Longley wrote:
>
> In article <6l975i$g...@ux.cs.niu.edu> ric...@cs.niu.edu "Neil Rickert" writes:
>
> > Da...@longley.demon.co.uk (David Longley) writes:
> >
> > >onwards has been because I judged the perspective I have been
> > >formulating, both important and original. Conclusions derived
> > >from having taught graduate psychologists an MSc course over 10
> > >years during their early years in an applied field which
> > >crucially needs to put this work into practice.
> >
> > >see http://www.longley.demon.co.uk/Frag.htm and other papers at
> > >that site.
> >
> > >I am looking for contributions/comments from major researchers in
> > >the field of AI, and am grateful for those who have already done
> > >so, both publicly and privately over the past three years.
> >
> > Most of the contributions in this forum have been in the form of
> > telling you that you are nuts. However, you have not seemed at all
> > grateful to that kind of contribution. I guess that means you have a
> > selective filter -- you only consider evidence which agrees with your
> > preconceived doctrines.
> >
> This really is very stupid - as anyone who knows any of the
> relevant history will tell you.
Let's see if any of those people come forth.
> You, and others like you should
> make the effort to listen .... it may all seem "nuts", but that's
> only because you are thinking as a "folk psychologist".
The question here isn't whether it seems nuts, but whether you have
been told that you are nuts; and indeed those who know the relevant
history (namely, three years of postings to c.a.p) can readily assess
that. One gentleman even went so far as to provide a diagnosis -- autism.
--
<J Q B>
Jim Balter
Stanley Friesen wrote:
> I find it interesting that, even with this approach, Tversky &
> Kahneman end their paper with the following:
>
> "These heuristics are highly economical and usually effective,
> but they lead to systematic and predictable errors. A better
> understanding of these heuristics and of the biases towhich they
> lead could improve judgments and decisions in situations of
> uncertainty."
Having errors be systematic and predictable is of course a *good*
characteristic when employing a heuristic (it of course would be bad if
exposed to a hostile agent).
> I particularly draw attention to the phrase "highly economical
> and usually effective". Hmm, "usually effective". That sounds
> rather like what I have been trying to say! I get the impression
> that these two authors actually agree with you and me, not with
> Longley.
This is something that Rickert, Weinstein, Sloman, and others have
pointed out repeatedly over the last three years.
> This reminds me of a review article in _Science News_ a year or
> two ago. It covered the changes in the research on heuristics in
> recent years. It pointed out that in the past research has
> concentrated on the limitations of heuristics, and only recently
> has much research been done on their capabilities. It covered
> quite a number of studies, on both sides. A very interesting
> article indeed. It presented a number of studies showing
> circumstances in which heuristics outperform "normative
> procedures".
Indeed the performance characteristic of *economy* is something that
Longley has systematically and characteristically avoided, though
it has been pointed out many times.
> [P.S. The HTML documents look like they were scanned in using an
> OCR system without any proofreading: they were full of
> run-together words, strange spellings, and odd combinations of
> characters - all of which I see when I scan documents, prior to
> proofreading. Stuff such as "juagmenl ullu~r ull~l~alII~y" and
> "InsufJicient" look like what comes out of OmniPage Pro raw].
Most of us have been oddly polite to Longley about this matter over the
years. I guess it's just a corollary of the netiquette concerning typos.
--
<J Q B>
Jim Balter
David Longley wrote:
>
> In article <199806091327...@ladder01.news.aol.com>
> jfze...@aol.com "JFZeigler" writes:
>
> > You know, I think I may have figured out what Longley's
> > difficulty is.
>
> *I* don't have the difficulty - *you* and a few others seem to be
> the ones having the difficulty by your own statements!
Jon Zeigler points out that Longley's behavior triggers our error-prone
heuristics, and Longley responds with a point-missing confirmation,
rather than an adaptive change of behavior. It seems the .bot could use
a few more heuristics of its own.
--
<J Q B>
Stanley Friesen
On Tue, 09 Jun 98 08:00:22 GMT, Da...@longley.demon.co.uk# wrote:
>In article <357d1662....@pop.americasttv.com>
> sfri...@americasttv.com "Stanley Friesen" writes:
>
>> I particularly draw attention to the phrase "highly economical
>> and usually effective". Hmm, "usually effective". That sounds
>> rather like what I have been trying to say! I get the impression
>> that these two authors actually agree with you and me, not with
>> Longley.
>>
BFD!
You keep sending me mere evidence that heuristics are fallable.
David Longley
In article <357E105D...@sandpiper.net>
j...@sandpiper.net "Jim Balter" writes:
> Stanley Friesen wrote:
>
> > I find it interesting that, even with this approach, Tversky &
> > Kahneman end their paper with the following:
> >
> > "These heuristics are highly economical and usually effective,
> > but they lead to systematic and predictable errors. A better
> > understanding of these heuristics and of the biases towhich they
> > lead could improve judgments and decisions in situations of
> > uncertainty."
>
> Having errors be systematic and predictable is of course a *good*
> characteristic when employing a heuristic (it of course would be bad if
> exposed to a hostile agent).
>
> > I particularly draw attention to the phrase "highly economical
> > and usually effective". Hmm, "usually effective". That sounds
> > rather like what I have been trying to say! I get the impression
> > that these two authors actually agree with you and me, not with
> > Longley.
>
> This is something that Rickert, Weinstein, Sloman, and others have
> pointed out repeatedly over the last three years.
>
> > This reminds me of a review article in _Science News_ a year or
> > two ago. It covered the changes in the research on heuristics in
> > recent years. It pointed out that in the past research has
> > concentrated on the limitations of heuristics, and only recently
> > has much research been done on their capabilities. It covered
> > quite a number of studies, on both sides. A very interesting
> > article indeed. It presented a number of studies showing
> > circumstances in which heuristics outperform "normative
> > procedures".
>
> Indeed the performance characteristic of *economy* is something that
> Longley has systematically and characteristically avoided, though
> it has been pointed out many times.
>
Pointed out - I think not. What we have here (as all too often,
is a few folk, finally grasping *some* of the material, but
having no idea how to respond except to try to identify flaws.
The fact is that there is a very practical technology of decsion
theory to drawn upon in place of the heuristics.
ON LOGICAL VS. INTUTIVE JUDGMENT
I thought the following may be of some interest (even if only as a
source of a few useful quotes). It is an extract from a piece of work
I put together last year as part of a large project entitled 'A System
Specification for PROfiling BEhaviour (PROBE)'. This extract, (taken
from Section C of Volume 1) is largely a review of the 'Clinical vs.
Actuarial issue in decision making, written primarily with applied
Criminological Psychologists working within the British Prison Service
in mind. It was also written to provide an infrastructure for
practising applied statistics - ie within the context of a large
relational database management system which focused on records of
behaviour.
EXTRACT
It will help if an idea of what we mean by 'clinical' and 'actuarial'
judgement is provided. The following is taken from a an early (Meehl
1954), and a relatively recent review of the status 'Clinical vs.
Actuarial Judgement' by Dawes, Faust and Meehl (1989):
'One of the major methodological problems of clinical
psychology concerns the relation between the "clinical"
and "statistical" (or "actuarial") methods of
prediction. Without prejudging the question as to
whether these methods are fundamentally different, we
can at least set forth the main difference between them
as it appears superficially. The problem is to predict
how a person is going to behave. In what manner should
we go about this prediction?
We may order the individual to a class or set of classes
on the basis of objective facts concerning his life
history, his scores on psychometric tests, behavior
ratings or check lists, or subjective judgements gained
from interviews. The combination of all these data
enables us to CLASSIFY the subject; and once having made
such a classification, we enter a statistical or
actuarial table which gives the statistical frequencies
of behaviors of various sorts for persons belonging to
the class. The mechanical combining of information for
classification purposes, and the resultant probability
figure which is an empirically determined relative
frequency, are the characteristics that define the
actuarial or statistical type of prediction.
Alternatively, we may proceed on what seems, at least,
to be a very different path. On the basis of interview
impressions, other data from the history, and possibly
also psychometric information of the same type as in the
first sort of prediction, we formulate, as a psychiatric
staff conference, some psychological hypothesis
regarding the structure and the dynamics of this
particular individual. On the basis of this hypothesis
and certain reasonable expectations as to the course of
other events, we arrive at a prediction of what is going
to happen. This type of procedure has been loosely
called the clinical or case-study method of prediction'.
P. E. Meehl (1954)
The Problem: Clinical vs. Statistical Prediction
'In the clinical method the decision-maker combines or
processes information in his or her head. In the
actuarial or statistical method the human judge is
eliminated and conclusions rest solely on empirically
established relations between data and the condition or
event of interest. A life insurance agent uses the
clinical method if data on risk factors are combined
through personal judgement. The agent uses the actuarial
method if data are entered into a formula, or tables and
charts that contain empirical information relating these
background data to life expectancy.
Clinical judgement should not be equated with a clinical
setting or a clinical practitioner. A clinician in
psychiatry or medicine may use the clinical or actuarial
method. Conversely, the actuarial method should not be
equated with automated decision rules alone. For
example, computers can automate clinical judgements. The
computer can be programmed to yield the description
"dependency traits", just as the clinical judge would,
whenever a certain response appears on a psychological
test. To be truly actuarial, interpretations must be
both automatic (that is, prespecified or routinized) and
based on empirically established relations.'
R. Dawes, D. Faust & P. Meehl (1989)
Clinical Versus Actuarial Judgement Science v243, pp
1668-1674 (1989)
As long ago as 1941, Lundberg made it clear that any argument between
those committed to the 'clinical' (intuitive) stance and those arguing
for the 'actuarial' (statistical) was a pseudo-argument, since all the
clinician could possibly be making his or her decision on was his or
her limited experience (database) of past cases and outcomes.
'I have no objection to Stouffer's statement that "if
the case-method were not effective, life insurance
companies hardly would use it as they do in
supplementing their actuarial tables by a medical
examination of the applicant in order to narrow their
risks." I do not see, however, that this constitutes a
"supplementing" of actuarial tables. It is rather the
essential task of creating specific actuarial tables. To
be sure, we usually think of actuarial tables as being
based on age alone. But on the basis of what except
actuarial study has it been decided to charge a higher
premium (and how much) for a "case" twenty pounds
overweight, alcoholic, with a certain family history,
etc.? These case-studies have been classified and the
experience for each class noted until we have arrived at
a body of actuarial knowledge on the basis of which we
"predict" for each new case. The examination of the new
case is for the purpose of classifying him as one of a
certain class for which prediction is possible.'
G. Lundberg (1941)
Case Studies vs. Statistical Methods - An Issue Based
on Misunderstanding. Sociometry v4 pp379-83 (1941)
A few years later, Meehl (1954), drawing on the work of Lundberg
(1941) and Sarbin (1941) in reviewing the relative merits of clinical
vs. statistical prediction (judgement) reiterated the point that all
judgements about an individual are always referenced to a class, they
are always therefore, probability judgements.
'No predictions made about a single case in clinical
work are ever certain, but are always probable. The
notion of probability is inherently a frequency notion,
hence statements about the probability of a given event
are statements about frequencies, although they may not
seem to be so. Frequencies refer to the occurrence of
events in a class; therefore all predictions; even those
that from their appearance seem to be predictions about
individual concrete events or persons, have actually an
implicit reference to a class....it is only if we have a
reference class to which the event in question can be
ordered that the possibility of determining or
estimating a relative frequency exists.. the clinician,
if he is doing anything that is empirically meaningful,
is doing a second-rate job of actuarial prediction.
There is fundamentally no logical difference between the
clinical or case-study method and the actuarial method.
The only difference is on two quantitative continua,
namely that the actuarial method is more EXPLICIT and
more PRECISE.'
P. Meehl (1954)
Clinical vs. Statistical Prediction:
A Theoretical Analysis and a Review of the Evidence
There has, unfortunately, over the years, been a strong degree of
resistance to the actuarial approach. It must be appreciated however,
that the technology to support comprehensive actuarial analysis and
judgment has only been physically available since the 1940s with the
invention of the computer. Practically speaking, it has only been
available on the scale we are now discussing since the late 1970s with
the development of sophisticated DBMS's (databases with query
languages based on the Predicate Calculus; Codd 1970; Gray 1984;
Gardarin and Valduriez 1989, Date 1992), and the development and mass
production of powerful and cheap microcomputers. Minsky and Papert
(1988) in their expanded edition of 'Perceptrons' (basic pattern
recognition systems) in fact wrote:
'The goal of this study is to reach a deeper
understanding of some concepts we believe are crucial to
the general theory of computation. We will study in
great detail a class of computations that make decisions
by weighting evidence.....The people we want most to
speak to are interested in that general theory of
computation.'
M. L. Minsky & S. A. Papert (1969,1990)
Perceptrons p.1
The 'general theory of computation' is, as elaborated elsewhere,
'Recursive Function Theory' (Church 1936, Kleene 1936, Turing 1937),
and is essentially the approach being advocated here as evidential
behaviourism, or eliminative materialism which eschews psychologism
and intensionalism. Nevertheless, as late as 1972, Meehl still found
he had to say:
'I think it is time for those who resist drawing any
generalisation from the published research, by
fantasising about what WOULD happen if studies of a
different sort WERE conducted, to do them. I claim that
this crude, pragmatic box score IS important, and that
those who deny its importance do so because they just
don't like the way it comes out. There are few issues in
clinical, personality, or social psychology (or, for
that matter, even in such fields as animal learning) in
which the research trends are as uniform as this one.
Amazingly, this strong trend seems to exert almost no
influence upon clinical practice, even, you may be
surprised to learn, in Minnesota!...
It would be ironic indeed (but not in the least
surprising to one acquainted with the sociology of our
profession) if physicians in nonpsychiatric medicine
should learn the actuarial lesson from biometricians and
engineers, whilst the psychiatrist continues to muddle
through with inefficient combinations of unreliable
judgements because he has not been properly instructed
by his colleagues in clinical psychology, who might have
been expected to take the lead in this development.
I understand (anecdotally) that there are two other
domains, unrelated to either personality assessment or
the healing arts, in which actuarial methods of data
combination seem to do at least as good a job as the
traditional impressionistic methods: namely, meteorology
and the forecasting of security prices. From my limited
experience I have the impression that in these fields
also there is a strong emotional resistance to
substituting formalised techniques for human judgement.
Personally, I look upon the "formal-versus-judgmental"
issue as one of great generality, not confined to the
clinical context. I do not see why clinical
psychologists should persist in using inefficient means
of combining data just because investment brokers,
physicians, and weathermen do so. Meanwhile, I urge
those who find the box score "35:0" distasteful to
publish empirical studies filling in the score board
with numbers more to their liking.'
P. E. Meehl (1972)
When Shall We Use Our Heads Instead of the Formula?
PSYCHODIAGNOSIS: Collected Papers (1971)
In 1982, Kahneman, Slovic and Tversky, in their collection of papers
on (clinical) judgement under conditions of uncertainty, prefaced the
book with the following:
'Meehl's classic book, published in 1954, summarised
evidence for the conclusion that simple linear
combinations of cues outdo the intuitive judgements of
experts in predicting significant behavioural criteria.
The lasting intellectual legacy of this work, and of the
furious controversy that followed it, was probably not
the demonstration that clinicians performed poorly in
tasks that, as Meehl noted, they should not have
undertaken. Rather, it was the demonstration of a
substantial discrepancy between the objective record of
people's success in prediction tasks and the sincere
beliefs of these people about the quality of their
performance. This conclusion was not restricted to
clinicians or to clinical prediction:
People's impressions of how they reason, and how well
they reason, could not be taken at face value.'
D. Kahneman, P. Slovic & A. Tversky (1982)
Judgment Under Conditions of Uncertainty: Heuristics and
Biases
Earlier in 1977, reviewing the Attribution Theory literature evidence
on individuals' access to the reasons for their behaviours, Nisbett
and Wilson (1977) summarised the work as follows:
'................................... there may be little
or no direct introspective access to higher order
cognitive processes. Ss are sometimes (a) unaware of the
existence of a stimulus that importantly influenced a
response, (b) unaware of the existence of the response,
and (c) unaware that the stimulus has affected the
response. It is proposed that when people attempt to
report on their cognitive processes, that is, on the
processes mediating the effects of a stimulus on a
response, they do not do so on the basis of any true
introspection. Instead, their reports are based on a
priori, implicit causal theories, or judgments about the
extent to which a particular stimulus is a plausible
cause of a given response. This suggests that though
people may not be able to observe directly their
cognitive processes, they will sometimes be able to
report accurately about them. Accurate reports will
occur when influential stimuli are salient and are
plausible causes of the responses they produce, and will
not occur when stimuli are not salient or are not
plausible causes.'
R. Nisbett & T. Wilson (1977)
Telling More Than We Can Know: Public Reports on Private
Processes
Such rules of thumb or attributions, are of course the intensional
heuristics studied by Tversky and Kahneman (1973), or the 'function
approximations' computed by neural network systems discussed earlier
as connection weights (both in artificial and real neural nets, cf.
Kandel's work with Aplysia).
Mathematical logicians such as Putnam (1975,1988); Elgin 1990 and
Devitt (1990) have long been arguing that psychologists may, as
Skinner (1971,1974) argued consistently, be looking for their data in
the wrong place. Despite the empirical evidence from research in
psychology on the problems of self report, and a good deal more drawn
from decision making in medical diagnosis, the standard means of
obtaining information for 'reports' on inmates for purposes of review,
and the standard means of assessing inmates for counselling is on the
basis of clinical interview. In the Prison Service this makes little
sense, since it is possible to directly observe behaviour under
relatively natural conditions of everyday activities. The clinical
interview, is still the basis of much of the work of the Prison
Psychologist despite the literature on fallibility of self-reports,
and the fallibility and unwitting distortions of those making
judgments in such contexts has been consistently documented within
psychology:
'The previous review of this field (Slovic, Fischoff &
Lichtenstein 1977) described a long list of human
judgmental biases, deficiencies, and cognitive
illusions. In the intervening period this list has both
increased in size and influenced other areas of
psychology (Bettman 1979, Mischel 1979, Nisbett & Ross
1980).'
H. Einhorn and R. Hogarth (1981)
The following are also taken from the text:
'If one considers the rather typical findings that
clinical judgments tend to be (a) rather unreliable (in
at least two of the three senses of that term), (b) only
minimally related to the confidence and amount of
experience of the judge, (c) relatively unaffected by
the amount of information available to the judge, and
(d) rather low in validity on an absolute basis, it
should come as no great surprise that such judgments are
increasingly under attack by those who wish to
substitute actuarial prediction systems for the human
judge in many applied settings....I can summarize this
ever-growing body of literature by pointing out that
over a very large array of clinical judgment tasks
(including by now some which were specifically selected
to show the clinician at his best and the actuary at his
worst), rather simple actuarial formulae typically can
be constructed to perform at a level no lower than that
of the clinical expert.'
L. R. Goldberg (1968)
Simple models or simple processes?
Some research on clinical judgments
American Psychologist, 1968, 23(7) p.483-496
'The various studies can thus be viewed as repeated
sampling from a uniform universe of judgement tasks
involving the diagnosis and predication of human
behavior. Lacking complete knowledge of the elements
that constitute this universe, representativeness cannot
be determined precisely. However, with a sample of about
100 studies and the same outcome obtained in almost
every case, it is reasonable to conclude that the
actuarial advantage is not exceptional but general and
likely to encompass many of the unstudied judgement
tasks. Stated differently, if one poses the query:
Would an actuarial procedure developed for a particular
judgement task (say, predicting academic success at my
institution) equal or exceed the clinical method?", the
available research places the odds solidly in favour of
an affirmative reply. "There is no controversy in social
science that shows such a large body of qualitatively
diverse studies coming out so uniformly....as this one
(Meehl J. Person. Assess, 50,370 (1986)".'
The distinction between collecting observations and integrating it is
further brought out vividly by Meehl (1989):
'Surely we all know that the human brain is poor at
weighting and computing. When you check out at a
supermarket you don't eyeball the heap of purchases and
say to the clerk, "well it looks to me as if it's about
$17.00 worth; what do you think?" The clerk adds it up.
There are no strong arguments....from empirical
studies.....for believing that human beings can assign
optimal weight in equations subjectively or that they
apply their own weights consistently.'
P. Meehl (1986)
Causes and effects of my disturbing little book
J Person. Assess. 50,370-5,1986
'Distributional information, or base-rate data, consist
of knowledge about the distribution of outcomes in
similar situations. In predicting the sales of a new
novel, for example, what one knows about the author, the
style, and the plot is singular information, whereas
what one knows about the sales of novels is
distributional information. Similarly, in predicting the
longevity of a patient, the singular information
includes his age, state of health, and past medical
history, whereas the distributional information consists
of the relevant population statistics. The singular
information consists of the relevant features of the
problem that distinguish it from others, while the
distributional information characterises the outcomes
that have been observed in cases of the same general
class. The present concept of distributional data does
not coincide with the Bayesian concept of a prior
probability distribution. The former is defined by the
nature of the data, whereas the latter is defined in
terms of the sequence of information acquisition.
The tendency to neglect distributional information and
to rely mainly on singular information is enhanced by
any factor that increases the perceived uniqueness of
the problem. The relevance of distributional data can be
masked by detailed acquaintance with the specific case
or by intense involvement with it........
The prevalent tendency to underweigh or ignore
distributional information is perhaps the major error of
intuitive prediction. The consideration of
distributional information, of course, does not
guarantee the accuracy of forecasts. It does, however,
provide some protection against completely unrealistic
predictions. The analyst should therefore make every
effort to frame the forecasting problem so as to
facilitate utilising all the distributional information
that is available to the expert.'
A. Tversky & D. Kahneman (1983)
Extensional Versus Intuitive Reasoning: The Conjunction
Fallacy in Probability Judgment Psychological Review
v90(4) 1983
'The possession of unique observational capacities
clearly implies that human input or interaction is often
needed to achieve maximal predictive accuracy (or to
uncover potentially useful variables) but tempts us to
draw an additional, dubious inference. A unique capacity
to observe is not the same as a unique capacity to
predict on the basis of integration of observations. As
noted earlier, virtually any observation can be coded
quantitatively and thus subjected to actuarial analysis.
As Einhorn's study with pathologists and other research
shows, greater accuracy may be achieved if the skilled
observer performs this function and then steps aside,
leaving the interpretation of observational and other
data to the actuarial method.'
R. Dawes, D. Faust and P. Meehl (1989)
ibid.
David Longley
In article <357DF2D6...@sandpiper.net>
j...@sandpiper.net "Jim Balter" writes:
>
> It would seem from this that the *common sense* assessment of the
> importance of the detection of covariance to the ability to achieve
> "harmonious social interchange" is flawed. Yet Longley's prescription
> is to stick to covariance and other "actuarial" methods *to the
> exclusion* of "layperson" methods.
Perhaps you'd like to re-write the above. The covariation
principle is a folk psychological judgemental heuristic.
Analysing observations and their relations statistically is a
normative process undertaken form the extensional stance.
>
> Of course, as someone without Longley's precise professional background,
> it's unlikely that I could reach an understanding merely by reading
> a bunch of material, even if it's precisely the material Longley
> quotes.
If you were seriously interested in learning anything rather than
irritatingly and ignorantly abusive you could read one or two of
the key references (asterisked below) to see whether what I am
saying is representative of the wider literature or not. They
*are* after all, quite substantial review articles.
Ref.
Inductive Reasoning as Heuristics
Agnoli F & Krantz D. H. Suppressing Natural Heuristics by Formal
Instruction: The Case of the Conjunction Fallacy Cognitive Psychology
21, 515-550, 1989
Eddy D M Probabilistic Reasoning in Clinical Medicine: Problems
and Opportunities In Kahneman, Tversky and Solvic (Eds) Judgment Under
Uncertainty: Heuristics and Biases Cambridge University Press 1982
Fong G T, Lurigio A J & Stalans L J Improving Probation Decisions
Through Statistical Training Criminal Justice and
Behavior,17,3,1990,370-388
Fong G T & Nisbett R E Immediate and delayed transfer of training
effects in Statistical reasoning. Journal of Experimental Psychology
General; 1991,120(1) 34-45
* Nisbett R E & Wilson T D Telling more than we can know: Verbal
reports on mental processes Psychological-Review; 1977 Mar Vol 84(3)
231-259
* Nisbett R E & Ross L Human Inference: Strategies and Shortcomings
of Social Judgment Century Psychology Series, Prentice-Hall (1980)
Nisbett R E, Fong G T, Lehman D R & Cheng P W Teaching Reasoning
Science v238, 1987 pp.625-631
Ross L & Nisbett R E The Person and The Situation: Perspectives of
Social Psychology McGraw Hill 1991
* Sutherland S IRRATIONALITY: The Enemy Within Constable: London 1992
Tversky A & Kahneman D Extensional Versus Intuitive Reasoning: The
Conjunction Fallacy in Probability Judgment Psychological Review
v90(4) 1983
* Arkes H R & Hammond K R Judgment and Decision Making: An
interdisciplinary reader Cambridge University Press 1986
Dawes R.M The Robust beauty of improper linear models in decision
making American Psychologist, 1979 34,571-582
* Dawes R M Rational Choice in an Uncertain World Orlando: Harcourt,
Brace, Jovanovich 1988
* Dawes R M, Faust D & Meehl P E Clinical Versus Actuarial Judgement
Science v243, pp 1668-1674 1989
Elstein A S Clinical judgment: Psychological research and medical
practice. Science; 1976 Nov Vol 194(4266) 696-700
* Einhorn H J & Hogarth R M Behavioral decision theory: Processes of
judgment and choice Annual Review of Psychology (1981), 32, 53-88
Faust D Data integration in legal evaluations: Can clinicians
deliver on their premises? Behavioral Sciences and the Law; 1989 Fal
Vol 7(4) 469-483
Goldberg L R Simple models or simple processes? Some research on
clinical judgments American Psychologist,1968,23(7) p.483-496
* Kahneman D, Slovic P & Tversky A Judgment Under Uncertainty:
Heuristics and Biases Cambridge University Press 1982
** See SCIENCE article 1974, 185 1124-1131 for the major artcile
Lundberg G A Case Studies vs. Statistical Methods - An Issue
Based on Misunderstanding. Sociometry v4 pp379-83 1941
* Meehl P E Clinical vs. Statistical Prediction: A Theoretical
Analysis and a Review of the Evidence University of Minnesota Press,
Minneapolis. 1954
Meehl P E When Shall We Use Our Heads Instead of the Formula?
PSYCHODIAGNOSIS: Collected Papers 1971
Meehl P E Causes and effects of my disturbing little book J
Person. Assess. 50,370-5,1986
Oskamp S Overconfidence in case-study judgments J. Consult.
Psychol. (1965), 29, 261-265
Sarbin T R Clinical Psychology - Art or Science? Psychometrica
v6 pp391-400 (1941)
Ramsey W, Stich S & Garon J Connectionism, eliminativism, an the
future of Folk Psychology in Greenwood J D (Ed) The Future of Folk
Psychology Cambridge University Press 1991
Stich S From Folk Psychology to Cognitive Science: The Case
Against Belief Bradford Books 1983
David Longley
In article <357E12A8...@sandpiper.net>
j...@sandpiper.net "Jim Balter" writes:
I've provided more than enough to enable those who are genuinely
interested to identify what is required in order not to have to
resort to intensional heuristics. One of the papers provides
(Agnoli and Krantz) even provides experimental data.
I, and colleagues see the sort of responses being posted here all
the time - don't think for a second you are being original.
There's over 30 years of research behind what I have said. Far
too few people in the AI field have grasped its significance
(that's been at least part of what I have had to say here) -
laymen generally have little idea of the work at all - and my
objective is not to educate them - the fact that heuristics and
their biases are characteristic of folk psychology is *why* a
professional applied behaviour science has something to offer.
And *that's* another part of what I have had to say here. The
two points together amount to a new perspective on "AI" as I have
said elsewhere.
David Longley
In article <357f62f1....@pop.americasttv.com>
sfri...@americasttv.com "Stanley Friesen" writes:
> On Tue, 09 Jun 98 18:40:36 GMT, Da...@longley.demon.co.uk#
> wrote:
>
> >In article <357e1a93....@pop.americasttv.com>
> > sfri...@americasttv.com "Stanley Friesen" writes:
> >> I doubt there is anything in this book I really disagree with!
> >> Yes, heuristics can produce erroneous conclusions, and, under
> >> some circumstances, can even lead to harmful biases.
> >>
> >> BFD!
> >>
> >> This is *not* my point. I in no way contend that these results
> >> are incorrect. What I disagree with is the *conclusions* which
> >> *you* draw from these results.
> >
> >No..the conclusins are *yours* - attributed to me through lack of
> >carefully reading of what I have said.
>
> We are *both* drawing conclusions from the evidence presented.
> You are concluding that heuristic procedures are *not* a
> necessary component of intelligence, and that they actually
> *detract* from intelligence. If you had not drawn this
> conclusion, you would not be so insistant on leaving heuristic
> procedures out of AI work.
There is a very large body of research in behavioural decision
theory which can, and should (I argue) serve as the basis for a
profession of applied behavioural science. That discipline
has a concrete field of application, and I have provided material
a considrable amount of material at the website below which
illustrates how that works out in practice.
>
> I, on the other hand, conclude that for timely action under
> chaotic conditions using limited resources, some heuristic
> approach is necessary. Thus I conclude that for an AI to be
> more than a laboratory curiosity it must use heuristics.
>
Under chaotic conditions you still have to apply the extensional
stance. The application of statistics depends on the extensional
laws of probability theory, not the implicit application of folk
psychological heuristics. Outside of the extensional stance,
decision making is not professional.
> > You should study carefully
> >what I have said about the nature of intensional contexts.
> >>
> >> You keep sending me mere evidence that heuristics are fallable.
> >> Since I do not disagree with that, this is unlikely to influence
> >> me particularly. It is the step from fallability to uselessness
> >> that I object to, especially in the light of good *laboratory*
> >> evidence that heuristics actually *are* useful in many
> >> circumstances. You seem to be ignoring this evidence (it is too
> >> bad that I cannot keep old issues of _Science News_, or I could
> >> give you some good citations on that).
> >>
> >>
> >The alternatives and conclusions you keep drawing are your own,
> >not mine. You should look carefully into what I have actually
> >written and not confabulate it with your own incorrect
> >inferences and paraphrases.
>
> If you are not saying that heuristic processes are not a
> component of intelligence, then what *are* you saying? And why
> should it be considered relevent to AI research?
I've been very clear on this. We assess "intellience"
relationally by differential performance on tests which are
fundamentally tests of extensional skills.
> >
> >Of course they're useful as a modus vivendi - but they are *not*
> >something which AI systems should seek to incporporate
>
> Which is a *conclusion*. One that I disagree with. I consider
> that the utility of heuristics as modus vivendi are the sina qua
> non of intelligent behavior.
>
Fine - you work with your model - but I thought you were trying
to understand mine. If you wnat to disagree, fine... but that's
surely irrelevant to the whole issue at hand.
Show you understand before disagreeing.
> To put it another way, my question is, why *shouldn't* an AI be
> concerned with a way of living?
It may well do - but the way we advance our "way of living" is
through the development or rules and technologies on the basis of
application of the extensional stance.
>
> > - the
> >point is that in all of the professions we invest considerable
> >efforts in trying to replace such heuristics with actuarially or
> >logically based systems - ie procedures developed from the
> >extensional stance.
>
> So? That has to do with producing and applying systematic
> knowledge - that is with epistimology. I consider that as being
> in a *different* catagory than "intelligence".
>
What you want to consider under different categories is not the
issue. At least not in this thread.
lsa...@aol.com
In article <897469...@longley.demon.co.uk>,
Da...@longley.demon.co.uk# wrote:
>
> In article <357DF2D6...@sandpiper.net>
> j...@sandpiper.net "Jim Balter" writes:
> >
> > It would seem from this that the *common sense* assessment of the
> > importance of the detection of covariance to the ability to achieve
> > "harmonious social interchange" is flawed. Yet Longley's prescription
> > is to stick to covariance and other "actuarial" methods *to the
> > exclusion* of "layperson" methods.
>
> Perhaps you'd like to re-write the above. The covariation
> principle is a folk psychological judgemental heuristic.
> Analysing observations and their relations statistically is a
> normative process undertaken form the extensional stance.
In artice <897469...@longley.demon.co.uk>,
Da...@longley.demon.co.uk (David Longley) wrote:
>In article <357DF2D6...@sandpiper.net>
> j...@sandpiper.net "Jim Balter" writes:
>>
>> It would seem from this that the *common sense* assessment of the
>> importance of the detection of covariance to the ability to achieve
>> "harmonious social interchange" is flawed. Yet Longley's prescription
>> is to stick to covariance and other "actuarial" methods *to the
>> exclusion* of "layperson" methods.
>
>Perhaps you'd like to re-write the above. The covariation
>principle is a folk psychological judgemental heuristic.
>Analysing observations and their relations statistically is a
>normative process undertaken form the extensional stance.
I would like to point to a couple of recent studies that
directly contradict Longley's stance:
[begin quote]
"Research on Causal Reasoning"
by Laura R. Novick
[...]
Human Causal Inference Is Unbiased
Early research began with the proposal that people
determine the causes of events much as scientists do, by
assessing (intuitively) the degree of covariation between
a potential causal factor and the effect. Subsequent
research, however, identified many ways in which people's
causal attributions deviated from the predictions of
covariational models. Pat Cheng and I hypothesized that,
contrary to the claims in the literature, people's causal
inferences are in fact rational. We argued that to address
this issue, it is necessary to distinguish between the data
on which the inference process operates and the process of
inference computation itself. Previous research either had
not made this distinction or had not accurately identified
the information people use to assess causality. Another
weakness of the earlier research is that the models against
which subjects' performance was compared were not always
normative.
[end quote]
And,
[begin quote]
"Complex Adaptive Systems as Intuitive Statisticians:
Causality, Contingency, and Prediction"
by Patricia W. Cheng Keith J. Holyoak
[...]
At least for the past quarter century, many psychologists
have seriously considered the possibility that untutored
humans as well as other animals are capable of acquiring
and using statistical knowledge about the structure of
the environment. Peterson and Beach (1967) called people
"intuitive statisticians", and Kelley (1967) proposed
that people are "intuitive scientists". In the context
of experimental paradigms investigating classical
conditioning, other theorists have suggested that lower
animals operate as intuitive statisticians (e.g.,
Gallistel, 1990; Miller & Schactman, 1985). Although
there has in fact been broad agreement that various forms
of causal induction depend on the implicit computation of
statistical information, the question of precisely what
is computed has yet to be resolved. In the field of animal
conditioning, as well as in human categorization and causal
induction, various theorists have proposed that animals
perform some implicit computation of statistical contingency:
the difference between the proportion of events for which an
effect occurs when a factor is present and that proportion
when it is absent.
[end quote]
Here are the URLs for the above:
http://peabody.vanderbilt.edu/psych_and_hd/faculty/novickl/causal.html
http://www.cs.qub.ac.uk/~P.Hanna/Philosophy/Papers/
complex_adaptive_systems_as_intuitive_statisticans.html
David Longley has apparently made a career out of pointing
out that human beings err. It should be noted that if the
errors are contrasted with the things that humans do well,
they become insignificant. My own estimate is that human
behavior is over 99.9% rational and reliable. Longley loves
to bring attention to the fact that humans must be trained
to use "normative" processes. May I also point out that
humans must be trained to do almost everything. The training
can be either autonomous or done with the help of an
instructor. This should not be seen as a negative. On the
contrary. When was the last time any "normative" system
learned much of anything new from observing its environment?
Can we expect Deep Blue to learn checkers?
Reliability in any trainable system (like humans) is
dependent on the degree of exposure. For example, if we
were exposed to mathematics as much as we are exposed to
walking, we would all be accomplished mathematicians. The
very fact that Longley can see the irrationality of
biological systems is a powerful testament to the utility
of the heuristics that he loves to demean. How else could
he see the flaws? The contradiction is there for everyone
to see.
For a scientist of Longley's professed caliber to insist
that a trainable robot using contingency-based statistical
heuristics, a robot which is capable of learning to drive a
car in a major city without getting into an accident is not
a valid scientific pursuit, does not speak well for certain
sectors of academia. To further insist that such a pursuit
must not be considered part of AI research borders on the
pathological.
Louis Savain
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
David Longley
In article <6lmsqk$huv$1...@nnrp1.dejanews.com> lsa...@aol.com writes:
>
> David Longley has apparently made a career out of pointing
> out that human beings err.
Is this an example of this "over 99.9% rational and reliable"
behaviour? How did you arrive at this?
> It should be noted that if the
> errors are contrasted with the things that humans do well,
> they become insignificant.
Quantify this.
Why on earth do we bother send people to school? Why do we go to
such lengths to select people for higher education?
> My own estimate is that human
> behavior is over 99.9% rational and reliable.
Am estimate I advise nobody to take seriously.
> Longley loves
> to bring attention to the fact that humans must be trained
> to use "normative" processes.
The fact that I have to make this point at all is telling.
> May I also point out that
> humans must be trained to do almost everything. The training
> can be either autonomous or done with the help of an
> instructor. This should not be seen as a negative.
Who said anything about it being "negative" (whatever this
means). It's just a fact of empiricism, and that the structure of
the world is generally not presented to us in a format which is
readily analysed by our raw perceptual systems. At least not to
the extent that the world now demands of us.
The
> very fact that Longley can see the irrationality of
> biological systems is a powerful testament to the utility
> of the heuristics that he loves to demean.
How many times do you have to be told - it is you who's
attributing value in this rather emotional manner. The facts
speak for themselves. We have good evidence for the extent of
human irrationality (outside rare enclaves of normative
convention which sustain such behaviour via powerful
contingencies - cf. academia and the professions.
>How else could
> he see the flaws? The contradiction is there for everyone
> to see.
> For a scientist of Longley's professed caliber to insist
> that a trainable robot using contingency-based statistical
> heuristics, a robot which is capable of learning to drive a
> car in a major city without getting into an accident is not
> a valid scientific pursuit, does not speak well for certain
> sectors of academia. To further insist that such a pursuit
> must not be considered part of AI research borders on the
> pathological.
>
When you learn to drive a car, you are taught a series of
concrete skills to be applied in an rule based manner. The fact
that this is so is known to anyone who knows anything about
driving. The fact that there are so many discrete elements to be
learned and that we have not deemed it financially worth
investing in robotic systems to do this - except as drive by wire
technology for testing cars in R&D should surprise nobody.
Stanley Friesen
>In article <357e1a93....@pop.americasttv.com>
> sfri...@americasttv.com "Stanley Friesen" writes:
>> I doubt there is anything in this book I really disagree with!
>> Yes, heuristics can produce erroneous conclusions, and, under
>> some circumstances, can even lead to harmful biases.
>>
>> BFD!
>>
>> This is *not* my point. I in no way contend that these results
>> are incorrect. What I disagree with is the *conclusions* which
>> *you* draw from these results.
>
>No..the conclusins are *yours* - attributed to me through lack of
>carefully reading of what I have said.
We are *both* drawing conclusions from the evidence presented.
You are concluding that heuristic procedures are *not* a
necessary component of intelligence, and that they actually
*detract* from intelligence. If you had not drawn this
conclusion, you would not be so insistant on leaving heuristic
procedures out of AI work.
I, on the other hand, conclude that for timely action under
chaotic conditions using limited resources, some heuristic
approach is necessary. Thus I conclude that for an AI to be
more than a laboratory curiosity it must use heuristics.
> You should study carefully
>what I have said about the nature of intensional contexts.
>>
>> You keep sending me mere evidence that heuristics are fallable.
>> Since I do not disagree with that, this is unlikely to influence
>> me particularly. It is the step from fallability to uselessness
>> that I object to, especially in the light of good *laboratory*
>> evidence that heuristics actually *are* useful in many
>> circumstances. You seem to be ignoring this evidence (it is too
>> bad that I cannot keep old issues of _Science News_, or I could
>> give you some good citations on that).
>>
>>
>The alternatives and conclusions you keep drawing are your own,
>not mine. You should look carefully into what I have actually
>written and not confabulate it with your own incorrect
>inferences and paraphrases.
If you are not saying that heuristic processes are not a
component of intelligence, then what *are* you saying? And why
should it be considered relevent to AI research?
>
>Of course they're useful as a modus vivendi - but they are *not*
>something which AI systems should seek to incporporate
Which is a *conclusion*. One that I disagree with. I consider
that the utility of heuristics as modus vivendi are the sina qua
non of intelligent behavior.
To put it another way, my question is, why *shouldn't* an AI be
concerned with a way of living?
> - the
lsa...@aol.com
In article <897518...@longley.demon.co.uk>,
Da...@longley.demon.co.uk# wrote:
>
> In article <6lmsqk$huv$1...@nnrp1.dejanews.com> lsa...@aol.com writes:
> >
> > David Longley has apparently made a career out of pointing
> > out that human beings err.
>
> behaviour? How did you arrive at this?
No. It's just humor that your perfectionist blinders is
preventing you from seeing.
> > It should be noted that if the
> > errors are contrasted with the things that humans do well,
> > they become insignificant.
>
> Quantify this.
No need to. Everyone but you can see it for themselves.
Just walking or pouring tea into a cup without spilling it
is enough to show the amazing reliability and rationality
of human behavior. Ask any roboticist how hard it is to
accomplish such things. It's called common sense.
> Why on earth do we bother send people to school? Why do we go
> to such lengths to select people for higher education?
Because we know from watching them learn to reliably
perform amazingly complex tasks such as walking and talking
that given enough exposure they can learn everything else.
The same statistical and causal heuristics that allow us to
learn how to walk also allow us to learn scholarly things.
Do you expect humans to be experts at things that they have
no exposure to?
> > My own estimate is that human
> > behavior is over 99.9% rational and reliable.
>
> Am estimate I advise nobody to take seriously.
Everyone would do well *not* to heed your advice because my
99.9% estimate is actually a conservative one. You just need
to take time to appreciate the astronomical complexity of human
behavior before you can talk about its unreliability. That you,
a psychologist, have not come to such an appreciation does not
speak well for your profession. Luckily most other psychologists
will not agree with your pathological stance. I've already
referred you to two studies that are at odds with what you have
been writing. You purposely ignored them. It makes me wonder.
> > Longley loves
> > to bring attention to the fact that humans must be trained
> > to use "normative" processes.
>
> The fact that I have to make this point at all is telling.
The fact that you think you need to make it is also telling.
> > May I also point out that
> > humans must be trained to do almost everything. The training
> > can be either autonomous or done with the help of an
> > instructor. This should not be seen as a negative.
>
> Who said anything about it being "negative" (whatever this
> means).
It simply means that, contrary to what you have been
advocating, it should not be seen as a reason for AI scientists
to stop working on inductive learning systems. A contingency-based
adaptive system which uses statistical covariation is an inductive
learning system. Covariational induction is a statistical process,
i.e, one which induces causal correlations based on statistical
patterns, contrary to what you have written in response to Jim Balter.
> It's just a fact of empiricism, and that the structure of
> the world is generally not presented to us in a format which is
> readily analysed by our raw perceptual systems. At least not to
> the extent that the world now demands of us.
If that is the case, I wonder why we have managed to survive
this long. On the contrary, I would say that our senses are
perfectly adapted to the type of physical phenomena that we need
to detect in order to act reliably and rationally.
> > The
> > very fact that Longley can see the irrationality of
> > biological systems is a powerful testament to the utility
> > of the heuristics that he loves to demean.
>
> How many times do you have to be told - it is you who's
> attributing value in this rather emotional manner. The facts
> speak for themselves. We have good evidence for the extent of
> human irrationality (outside rare enclaves of normative
> convention which sustain such behaviour via powerful
> contingencies - cf. academia and the professions.
This is nonsense. I've given the examples of driving and walking
and talking to illustrate that we are indeed extremely rational in
the things that we do on an everyday basis. There is nothing rare
about walking and talking. In fact, we are so rational and reliable
that it is our *rare* mistakes that stand out. Not being an AI
scientist, you continually fail to appreciate the enormous complexity
of mundane human behavior.
> > How else could
> > he see the flaws? The contradiction is there for everyone
> > to see.
> > For a scientist of Longley's professed caliber to insist
> > that a trainable robot using contingency-based statistical
> > heuristics, a robot which is capable of learning to drive a
> > car in a major city without getting into an accident is not
> > a valid scientific pursuit, does not speak well for certain
> > sectors of academia. To further insist that such a pursuit
> > must not be considered part of AI research borders on the
> > pathological.
> >
>
> When you learn to drive a car, you are taught a series of
> concrete skills to be applied in an rule based manner. The fact
> that this is so is known to anyone who knows anything about
> driving. The fact that there are so many discrete elements to be
> learned and that we have not deemed it financially worth
> investing in robotic systems to do this - except as drive by wire
> technology for testing cars in R&D should surprise nobody.
Using your "extensional" deterministic programming methods to
create a general driving robot is out of the question. The only
way such a complex task can be accomplished is by creating a goal
oriented system that uses statistical methods to learn how to drive.
This is how biological system learn to perform complex tasks. It
is an inductive process and there is no other way to do it because
it is proven that the reliability of extensionally engineered
systems is inversely proportional to their complexity. The
complexity is intractable to normative means. It has nothing to do
with economical feasibility. It just cannot be done, period. If
you don't believe me, ask the programmers who wrote the code for
the Denver airport baggage handling system.
Bill Taylor
Da...@longley.demon.co.uk (David Longley) writes:
|> The covariation
|> principle is a folk psychological judgemental heuristic.
The concept that judgemental heuristics exist is classic folk psychology.
|> If you were seriously interested in learning anything rather than
|> irritatingly and ignorantly abusive
This form of ad hominem is a classic fallacy due to pop psychological
superficiality. Poor effort!
It's amazing that anyone would make such a blatant error, when Quine
has led the way for us all in defining such error.
-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
The solipsist society may well have more than one member, but who's counting?
-------------------------------------------------------------------------------
Bill Taylor
THINK !? THAT'S not very extensional! Poor effort!
|> finally grasping *some* of the material,
To speak of "grasping material" (unless you're speaking of a wanker) is
very much indicative of folk psychology. A terrible slip-up.
|> There has, unfortunately, over the years, been a strong degree of
|> resistance to the actuarial approach.
Resistance to an approach? Very fuzzy. Very intensional.
Go to the bottom of the class.
|> and is essentially the approach being advocated here
Surely we're not allowed to give personal evalustions of this type?
It seems as if Longley hasn't the slightest idea of extensional objectivity.
Humph!
-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
Any sufficiently primitive magic is indistinguishable from science.
-------------------------------------------------------------------------------
Bill Taylor
Da...@longley.demon.co.uk (David Longley) writes:
|> There is a very large body of research in behavioural decision
|> theory which can, and should (I argue) serve as the basis for
NOT GOOD ENOUGH! This personal evaluation denotes an intensional attitude
which nullifies any attempt to conform to the great Quine's standards.
|> I've been very clear on this.
WHAT A GIVEAWAY! HE thinks HE HIMSELF has been clear! Has there been
any clearer indication of rampant intensionality than this! Quine will
be turning in his grave, when he gets to it.
|> Fine - you work with your model - but I thought you were trying
|> to understand mine.
What a terrible denial of the extensional stance, and all it implies.
Heretical.
-------------------------------------------------------------------------------
Bill Taylor W.Ta...@math.canterbury.ac.nz
-------------------------------------------------------------------------------
No-one with raging toothache would be so silly as to say...
I THINK, therefore I am.
-------------------------------------------------------------------------------
Oliver Sparrow
mat...@math.canterbury.ac.nz (Bill Taylor) wrote:
"" It seems as if Longley hasn't the slightest idea of extensional
objectivity.
Go for it, Taylor. :=)
______________________________________
Oliver Sparrow
David Longley
In article <6lns4b$4u$1...@cantuc.canterbury.ac.nz>
mat...@math.canterbury.ac.nz "Bill Taylor" writes:
> Da...@longley.demon.co.uk (David Longley) writes:
>
> |> The covariation
> |> principle is a folk psychological judgemental heuristic.
>
> The concept that judgemental heuristics exist is classic folk psychology.
>
>
Just read the material I sent you, and if that doesn't persuade
you, let's just say we disagree.
David Longley
In article <35808477...@news.demon.co.uk>
oh...@chatham.demon.co.uk "Oliver Sparrow" writes:
Look at the evidence - weigh it up - if you aren't persuaded that
of the merits of the case I have provided over the past three
years, fine - we'll just have to disagree.
David Longley
> In article <897518...@longley.demon.co.uk>,
> Da...@longley.demon.co.uk# wrote:
> >
> > In article <6lmsqk$huv$1...@nnrp1.dejanews.com> lsa...@aol.com writes:
> > >
> > > David Longley has apparently made a career out of pointing
> > > out that human beings err.
> >
> > Is this an example of this "over 99.9% rational and reliable"
> > behaviour? How did you arrive at this?
>
> No. It's just humor that your perfectionist blinders is
> preventing you from seeing.
This isn't a humour group.
>
> > > It should be noted that if the
> > > errors are contrasted with the things that humans do well,
> > > they become insignificant.
> >
> > Quantify this.
>
> No need to. Everyone but you can see it for themselves.
> Just walking or pouring tea into a cup without spilling it
> is enough to show the amazing reliability and rationality
> of human behavior. Ask any roboticist how hard it is to
> accomplish such things. It's called common sense.
This is an odd way of looking at things. It's *difficult* to
program in general. It's easier to let an ANN abstract the
principles from being taken through the motions. The difficulty
with the latter is that it's unaccountable for its actions
amongst other things. We also know that such a way of learning is
prone to the biases listed elsewhere. This has been exmpirically
demonstrated in the learning theory literature as I have remarked
before.
>
> > Why on earth do we bother send people to school? Why do we go
> > to such lengths to select people for higher education?
>
> Because we know from watching them learn to reliably
> perform amazingly complex tasks such as walking and talking
> that given enough exposure they can learn everything else.
> The same statistical and causal heuristics that allow us to
> learn how to walk also allow us to learn scholarly things.
> Do you expect humans to be experts at things that they have
> no exposure to?
But peole don't just learn by being exposed to material at school
or university. It takes a lot of training, practice and testing
to ensure that they are proficient in the acquisition of rules
and their application.
>
> > > My own estimate is that human
> > > behavior is over 99.9% rational and reliable.
> >
> > Am estimate I advise nobody to take seriously.
>
> Everyone would do well *not* to heed your advice because my
> 99.9% estimate is actually a conservative one. You just need
> to take time to appreciate the astronomical complexity of human
> behavior before you can talk about its unreliability. That you,
> a psychologist, have not come to such an appreciation does not
> speak well for your profession. Luckily most other psychologists
> will not agree with your pathological stance. I've already
> referred you to two studies that are at odds with what you have
> been writing. You purposely ignored them. It makes me wonder.
I've read them, and I've read a lot else besides. Unlike you, my
reading is not limited by what I can find on the WWW.
>
> > > Longley loves
> > > to bring attention to the fact that humans must be trained
> > > to use "normative" processes.
> >
> > The fact that I have to make this point at all is telling.
>
> The fact that you think you need to make it is also telling.
Fine - just ignore what I have to say, and ignor ethe evidence
too. It's your loss not mine.
David Longley
The fact that I use natural language and it's idioms here is
really not the point at all. Those familiar with 'anomalous
monism' will appreciate that convention forces us to use such
idioms, although in sciene we do not use such idioms in an
explanatory capacity.
All of this has been said before.
Seth Russell
lsa...@aol.com wrote:
> [snip] because
> it is proven that the reliability of extensionally engineered
> systems is inversely proportional to their complexity. The
> complexity is intractable to normative means.
You know that has the ring of truth to it. After a life time of
trying to make "normative means" work for my clients, I can
personally attest to it. However, I wonder, can we prove it ?
--
Seth
See "Bozo's Conjecture" at
http://www.clickshop.com/ai/conjecture.htm
And then on to the AI Jump List ...
David Longley
In article <3580c032....@pop.americasttv.com>
sfri...@americasttv.com "Stanley Friesen" writes:
> On Wed, 10 Jun 98 22:49:21 GMT, Da...@longley.demon.co.uk# wrote:
> >... The facts
> >speak for themselves.
>
> The problem is that in this case, *no* *they* *don't*. They may
> *seem* to do so *to* *you*, but not everybody draws the same
> conclusions from these facts.
>
> > We have good evidence for the extent of
> >human irrationality
>
> Yep.
>
> And we *also* have good evidence for the *efficacy* of human
> reason in many circumstances. Some of which you *deleted* from
> Savain's post!
>
There's a judgement call here. I'm not interested in quibbling
over this. I've made a philosophical/metyhodological point and if
you look carefully at what's been said you will have a hard time
proving that the efficacious behaviour which you think "natural"
is not empificlaly acquired procedures learned either formally or
serendipidiously.
Note - the emphasis has been on the extensional stance with all
of the strengths and merits of knowing what one is talking about
as opposed to the demonstrable breakdown of logic within
intensional contexts of propositional attitude.
The focus of the contra-arguments I have seen to date miss the
overall point and merely assert a religious faith in what is
purportedly 'natural'.
> > (outside rare enclaves of normative
> >convention which sustain such behaviour via powerful
> >contingencies - cf. academia and the professions.
> >
> Except in cases where "normative" approaches are actually
> deletrious - which you deny are relevent.
Normative processes are the only processes we can coherntly
discuss. I have made that point many times and illustrated how
and why this is so not only on the basis of the intensional
heuristics, but also via the logical pathology of the idioms of
propositional attitude (cf. failure of substitutivity of
identicals salva veritate/failure of logical quantification).
>
> >When you learn to drive a car, you are taught a series of
> >concrete skills to be applied in an rule based manner.
>
> Only partially. For such things as recognizing what other
> drivers are doing, or estimating margins and distances, or
> judging probably outcomes of many actions on the controls, a
> human uses learned heuristics, not normative behaviors.
This is all missing the point. Try and grasp where this all
starts - in the anomalous nature of the idioms of propositional
attitude.
>
> I know what my car will do when I press the brakes not from any
> rule-based reasoning, bu simply from extrapolation from prior
> behavior of the car. This is an example of exemplar based
> reasoning, not statistical reasoning.
>
>
What you *know*, *believe*, *think* etc may all seem very
famikair and sound to you. But then your whole concept of self-
hood is a folk psychological notion.
To understand what I have been saying, you neeed to put all of
this aside - look atthe evidence and the nature of professional
behaviour.
You begin your post by missing the point about concrete
application - look to what I have said, not what you think I've
said.
Jerry Hull
On Thu, 11 Jun 98 18:21:06 GMT, Da...@longley.demon.co.uk (David
Longley) wrote:
>
>Note - the emphasis has been on the extensional stance with all
>of the strengths and merits of knowing what one is talking about
>as opposed to the demonstrable breakdown of logic within
>intensional contexts of propositional attitude.
>
>The focus of the contra-arguments I have seen to date miss the
>overall point and merely assert a religious faith in what is
>purportedly 'natural'.
Wad some Power the giftie gee us to see ourselves as ithers see us, or
something like that. We can always "know what we are talking about"
if we ignore the complexities of reality that go beyond our pathetic
preconceptions. I also recommend that you stick to Aristotelian logic
(that Frege/Russell stuff is much too contrived), & Newtonian physics
is surely much more comfortable than all that quantum & relativity
biznez.
Or, just maybe, the "breakdown" of standard logic in propositional
attitudes points up an opportunity to develop a more encompassing
logic that can handle the complications of the actual world. Cuz, you
know, not all substitutions of identities (&c.) in opaque contexts
result in falsehoods: only those that reveal errors/limitations of
the individual whose attitudes we are propositioning.
--
Jer
"Stop, stop. Do not speak. The ultimate truth
is not even to think." -- Sakyamuni
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>Note - the emphasis has been on the extensional stance with all
>of the strengths and merits of knowing what one is talking about
>as opposed to the demonstrable breakdown of logic within
>intensional contexts of propositional attitude.
Right. But it is all part of the grand Longley illusion.
Let's remove all intensions from language. Is there anything left?
Longley's illusion is that extensions remain. I'll bet that Longley
cannot define as much as a single extension -- not even the empty
set -- without relying on intensions.
And if I am right, extensionalism (what is left after removing the
intensions) is completely empty.
Stanley Friesen
On Wed, 10 Jun 98 17:51:00 GMT, Da...@longley.demon.co.uk# wrote:
>In article <357f62f1....@pop.americasttv.com>
> sfri...@americasttv.com "Stanley Friesen" writes:
>> You are concluding that heuristic procedures are *not* a
>> necessary component of intelligence, and that they actually
>> *detract* from intelligence. If you had not drawn this
>> conclusion, you would not be so insistant on leaving heuristic
>> procedures out of AI work.
>
>profession of applied behavioural science. That discipline
>has a concrete field of application, and I have provided material
>a considrable amount of material at the website below which
>illustrates how that works out in practice.
>
the scope of that "concrete field of application".
>>
>> I, on the other hand, conclude that for timely action under
>> chaotic conditions using limited resources, some heuristic
>> approach is necessary. Thus I conclude that for an AI to be
>> more than a laboratory curiosity it must use heuristics.
>
>stance. The application of statistics depends on the extensional
>laws of probability theory, not the implicit application of folk
>psychological heuristics. Outside of the extensional stance,
>decision making is not professional.
>
Applying "extensional" decision making processes will often take
too long given a reasonable, attainable computing capacity (which
is what I meant by "limited resources" above). If, in order to
take advantage of a situation, or to avoid harm from a situation,
one *must* take action before one is able to collect sufficient
evidence for applying statistics to it, then one must, of
necessity, either ignore the situation or use some less precise
method to try and guess the proper action. Trying to collect a
statistically valid sample will take so long that the situation
will be gone before one can respond.
Applying statistics is simply not alway a viable option, unless
one has infinite resources for collecting an evaluating evidence.
>
>> Which is a *conclusion*. One that I disagree with. I consider
>> that the utility of heuristics as modus vivendi are the sina qua
>> non of intelligent behavior.
>>
>
>Fine - you work with your model - but I thought you were trying
>to understand mine. If you wnat to disagree, fine... but that's
>surely irrelevant to the whole issue at hand.
>
Actually, I have decided it is the *sole* issue at hand.
>Show you understand before disagreeing.
>
Well, I understand the science you are presenting. None of the
citations or articles you sent me contained anything I have not
read and understood previously.
It is the road from the scientific facts to the applications that
I cannot seem to grasp, as you seem to think it is "obvious" from
the evidence, and cannot seem to explain that jump.
Well, it is *not* obvious from the evidence. The evidence you
cite can be construed to imply quite a different course of
practical application than you suggest.
>> To put it another way, my question is, why *shouldn't* an AI be
>> concerned with a way of living?
>
>It may well do - but the way we advance our "way of living" is
>through the development or rules and technologies on the basis of
>application of the extensional stance.
>
Up to a point, yes. But that point is quickly reached in the
humble-jumble of living.
I would *never* trust a purely "extensional" agent to drive an
automobile! At least not unless it was simultaneously driving
*all* automobiles on the road, and thus have full knowledge of
all relevent variables all of the time. Even one automobile not
under its control would make it unable to adequately predict
outcomes in real time using purely statistically valid
techniques.
[This is why all proposals for an auto-drive lane include
provisions to utterly exclude cars that do not transfer control
to the system].
Stanley Friesen
On Wed, 10 Jun 98 22:49:21 GMT, Da...@longley.demon.co.uk# wrote:
>... The facts
>speak for themselves.
The problem is that in this case, *no* *they* *don't*. They may
*seem* to do so *to* *you*, but not everybody draws the same
conclusions from these facts.
> We have good evidence for the extent of
>human irrationality
Yep.
And we *also* have good evidence for the *efficacy* of human
reason in many circumstances. Some of which you *deleted* from
Savain's post!
> (outside rare enclaves of normative
>convention which sustain such behaviour via powerful
>contingencies - cf. academia and the professions.
>
Except in cases where "normative" approaches are actually
deletrious - which you deny are relevent.
>When you learn to drive a car, you are taught a series of
>concrete skills to be applied in an rule based manner.
Only partially. For such things as recognizing what other
drivers are doing, or estimating margins and distances, or
judging probably outcomes of many actions on the controls, a
human uses learned heuristics, not normative behaviors.
I know what my car will do when I press the brakes not from any
Oliver Sparrow
ric...@cs.niu.edu (Neil Rickert) wrote:
"" I'll bet that Longley
"" cannot define as much as a single extension -- not even the empty
"" set -- without relying on intensions.
I imagine that you are right, although there may be abstract or trivial
things to be said. At the base, though, after tracking th e fringe of this
for four years now, I cannot see the point. What, behind all the fluff, does
this discriminate between, and how does this discriminant help me or anyone
else to achieve their goals? There seems to be no punch line.
There is a major effort afoot at the moment to base public health provision
around what is called 'evidence based medicine'. I have some peripheral
involvement with this, but organisations such as the Welcome Trust have just
published synopses which make a great deal of sense. They leap frog what
Longley is trying to say to its implications.
Allocation of resources in the hundred billion plus health provision arena is
plainly a major task. At present, it depends upon two features: the judgement
of the individual health professional and the institutional means which the
society has put in place for what is potentially limitless ambitions for
health product consumption. Neither of these are based on entirely rational
grounds. The grounds exist, however: we can study what works and estimate the
benefits and the resource choices associated with various portfolios of
health provision. Thus, we can be rational. The punch line is, however, that
if we choose to be rational, then the power base of the individual
practitioner is removed, because they are now required to prioritise patients
in certain ways, and to use specific treatments for them. The political
choices which are implicit in current systems of rationing (and the hard
cases which rigorous systems always throw up) are both severe tests of
political will.
Boring, you say. True, but this is what the Longley Line takes one to when it
is applied. What has this to do with AI? Peripheral stuff, I suggest, very
peripheral. So why is Longley publishing in this newsgroup rather than in a
social science of public policy forum? I cannot possible project myself into
his mind state, and thus cannot say. Extensional or what?
______________________________________
Oliver Sparrow
David Longley
> Da...@longley.demon.co.uk (David Longley) writes:
>
> >Note - the emphasis has been on the extensional stance with all
> >of the strengths and merits of knowing what one is talking about
> >as opposed to the demonstrable breakdown of logic within
> >intensional contexts of propositional attitude.
>
> Right. But it is all part of the grand Longley illusion.
>
> Let's remove all intensions from language. Is there anything left?
Well, if you read Quine carefully you'll see that comes pretty
close to what he advises for scientific language. For natural
language, the case is slightly different, but only because we
have no natural alternative at present. This is anomalous monism.
When it comes to the other intensional idioms, ie the modal
operators, the problem is not so acute - in logic, Quine advises
elimination (despite the protestations of modal logicians). It is
the task of behaviour science to come up with alternatives to the
idioms of propositional attitude - and this is something I think
we, as a profession have already made some sort of progress
towards.
>
> Longley's illusion is that extensions remain. I'll bet that Longley
> cannot define as much as a single extension -- not even the empty
> set -- without relying on intensions.
>
The success of the extensional system of science indicates that
there is something radically wrong with your reasoning.
> And if I am right, extensionalism (what is left after removing the
> intensions) is completely empty.
>
And I suggest you are just wrong. Your passion for the
intensional is just that - a passion.
David Longley
In article <3581d898...@news.demon.co.uk>
oh...@chatham.demon.co.uk "Oliver Sparrow" writes:
> ric...@cs.niu.edu (Neil Rickert) wrote:
>
> "" I'll bet that Longley
> "" cannot define as much as a single extension -- not even the empty
> "" set -- without relying on intensions.
>
> I imagine that you are right, although there may be abstract or trivial
> things to be said. At the base, though, after tracking th e fringe of this
> for four years now, I cannot see the point. What, behind all the fluff, does
> this discriminate between, and how does this discriminant help me or anyone
> else to achieve their goals? There seems to be no punch line.
I suspect the reason why you don't see this is because you are
not aware of the alternatives. I've trie dto illustrate this in
the supporting papers to http://www.longley.demon.co.uk/Frag.htm.
If you have read those, and done so in the context of the
material presented in "Fragments of Behaviour: The Extensional
Stance" - *and* understood the Quinean context which I have gone
to some lengths to explicate here - and you still don't
understand - perhaps you need to be working in the field I am
talking about to appreciate the problem and the solution I am
proposing.
If the reactions you have observed here are not enough to give
you a clue - just consider the following. You can put your
rational had on and pretend you don't see the obstacle if you
wish...
COURSE TEXT:
JACK: Meehl's study set off, or at least much
inflamed, the 'statistical versus clinical judgment'
controversy, which has rumbled on ever since, though
it's somewhat less fashionable than it was.
PENELOPE: Why?
JACK: Cynically, because the human judges didn't like
the results and made sure that they or their
authors didn't get the funding, circulation or
promotion they deserved. Closed shops (as most
professions are to some extent) are not likely to vote
for what they see as de-skilling, and alternative
approaches that showed more respect for the human
judge became fashionable and fund worthy (especially
the expert systems we shall meet in the session
after next). Uncynically, the methodological problems
in policy-capturing research are real: it IS difficult
to establish the external validity of the results.'
Page 63 Volume 1 Introductory Text 2
David Longley
A little light relief I thought I'd pass on (despite this not
being a humour group - I think it can be considered apposite <g>).
>
>The story behind the letter below is that there is this nutball in
>Newport, RI named Scott Williams who digs things out of his backyard and
>sends the stuff he finds to the Smithsonian Institute, labeling them
>with scientific names, insisting that they are actual archaeological
>finds.
>
>This guy really exists and does this in his spare time! Anyway...here's
>the actual response from the Smithsonian Institution. Bear this in mind
>next time you think you are challenged in your duty to respond to a
>difficult situation in writing.
>
> ------------------------
>Smithsonian Institute
>207 Pennsylvania Avenue
>Washington, DC 20078
>
>Dear Mr. Williams:
>
>Thank you for your latest submission to the Institute, labeled "93211-D,
>layer seven, next to the clothesline post...Hominid skull." We have
>given this specimen a careful and detailed examination, and regret to
>inform you that we disagree with your theory that it represents
>conclusive proof of the presence of Early Man in Charleston County two
>million years ago.
>
>Rather, it appears that what you have found is the head of a Barbie
>doll, of the variety that one of our staff, who has small children,
>believes to be "Malibu Barbie." It is evident that you have given a
>great
>deal of thought to the analysis of this specimen, and you may be quite
>certain that those of us who are familiar with your prior work in the
>field were loathe to come to contradiction with your findings.
>
>However, we do feel that there are a number of physical attributes of
>the specimen which might have tipped you off to its modern origin:
>
>1. The material is molded plastic. Ancient hominid remains are typically
>fossilized bone.
>
>2. The cranial capacity of the specimen is approximately 9 cubic
>centimeters, well below the threshold of even the earliest identified
>proto-homonids.
>
>3. The dentition pattern evident on the skull is more consistent with
>the common domesticated dog than it is with the ravenous man-eating
>Pliocene clams you speculate roamed the wetlands during that time.
>
>This latter finding is certainly one of the most intriguing hypotheses
>you have submitted in your history with this institution, but the
>evidence seems to weigh rather heavily against it. Without going into
>too
>much detail, let us say that:
>
>A. The specimen looks like the head of a Barbie doll that a dog has
>chewed on.
>
>B. Clams don't have teeth.
>
>It is with feelings tinged with melancholy that we must deny your
>request to have the specimen carbon-dated. This is partially due to the
>heavy load our lab must bear in its normal operation, and partly due to
>carbon-dating's notorious inaccuracy in fossils of recent geologic
>record. To the best of our knowledge, no Barbie dolls were produced
>prior to 1956 AD, and carbon-dating is likely to produce wildly
>inaccurate results.
>
>Sadly, we must also deny your request that we approach the National
>Science Foundation Phylogeny Department with the concept of assigning
>your specimen the scientific name Australopithecus spiff-arino.
>
>Speaking personally, I, for one, fought tenaciously for the acceptance
>of your proposed taxonomy, but was ultimately voted down because the
>species name you selected was hyphenated, and didn't really sound like
>it might be Latin.
>
>However, we gladly accept your generous donation of this fascinating
>specimen to the museum. While it is undoubtedly not a Hominid fossil, it
>is, nonetheless, yet another riveting example of the great body of work
>you seem to accumulate here so effortlessly.
>
>You should know that our Director has reserved a special shelf in his
>own office for the display of the specimens you have previously
>submitted to the Institution, and the entire staff speculates daily on
>what you will happen upon next in your digs at the site you have
>discovered in your Newport back yard.
>
>We eagerly anticipate your trip to our nation's capital that you
>proposed in your last letter, and several of us are pressing the
>Director to pay for it. We are particularly interested in hearing you
>expand on your
>theories surrounding the trans-positating fillifitation of ferrous metal
>in a structural matrix that makes the excellent juvenile Tyrannosaurus
>rex femur you recently discovered take on the deceptive appearance of a
>rusty 9-mm Sears Craftsman automotive crescent wrench.
>
>Yours in Science,
>
>Harvey Rowe
>Chief Curator-Antiquities
David Longley
> ric...@cs.niu.edu (Neil Rickert) wrote:
>
> "" I'll bet that Longley
> "" cannot define as much as a single extension -- not even the empty
> "" set -- without relying on intensions.
>
> I imagine that you are right, although there may be abstract or trivial
> things to be said. At the base, though, after tracking th e fringe of this
> for four years now, I cannot see the point. What, behind all the fluff, does
> this discriminate between, and how does this discriminant help me or anyone
> else to achieve their goals? There seems to be no punch line.
>
> There is a major effort afoot at the moment to base public health provision
> around what is called 'evidence based medicine'. I have some peripheral
> involvement with this, but organisations such as the Welcome Trust have just
> published synopses which make a great deal of sense. They leap frog what
> Longley is trying to say to its implications.
>
> Allocation of resources in the hundred billion plus health provision arena is
> plainly a major task. At present, it depends upon two features: the judgement
> of the individual health professional and the institutional means which the
> society has put in place for what is potentially limitless ambitions for
> health product consumption. Neither of these are based on entirely rational
> grounds. The grounds exist, however: we can study what works and estimate the
> benefits and the resource choices associated with various portfolios of
> health provision. Thus, we can be rational. The punch line is, however, that
> if we choose to be rational, then the power base of the individual
> practitioner is removed, because they are now required to prioritise patients
> in certain ways, and to use specific treatments for them. The political
> choices which are implicit in current systems of rationing (and the hard
> cases which rigorous systems always throw up) are both severe tests of
> political will.
>
> Boring, you say. True, but this is what the Longley Line takes one to when it
> is applied.
It may be where it leads when *you* apply it. I certainly haven't
applied it in that context. If you can demonstrate that I have,
please point this out explictly rather than attribute such
falsehoods.
I have been very explicit about this sort of behaviour - and I
have been equally explicit about the domain within which I advise
application of what I have had to say.
> What has this to do with AI? Peripheral stuff, I suggest, very
> peripheral. So why is Longley publishing in this newsgroup rather than in a
> social science of public policy forum? I cannot possible project myself into
> his mind state, and thus cannot say. Extensional or what?
>
You don't need to project yourself - just read the beginning of
"http://www.longley.demon.co.uk/Frag.htm". There is enough
material there to make it all quite clear.
As to why here than elswehere - well, it has been posted and
discussed in other areas and lists - what I have had to say about
AI here pertains to the philosophy of AI - and the aspirations of
GOFAI in particular.
Jerry Hull
On Fri, 12 Jun 98 10:48:32 GMT, Da...@longley.demon.co.uk (David
Longley) wrote:
>> And if I am right, extensionalism (what is left after removing the
>> intensions) is completely empty.
>>
>And I suggest you are just wrong. Your passion for the
>intensional is just that - a passion.
Longley's view is not just non-intuitive & "unnatural": it is
logically incoherent. Here we have him saying that a person who
disagrees with him is "just wrong." How is that person wrong? Does
he drive on the wrong side of the road or take more than 8 items into
the express lane? No, the guy's beliefs on the subject of
intensionality are wrong. But if people can have wrong beliefs, it
logically follows that one cannot validly draw inferences from
person's beliefs on the basis of what is true. Viola, intensional
contexts!
So Longley finds himself in the awkward position that he is unable to
disagree with his opponents without demonstrating the existence of the
very thing he disputes.
David Longley
In article <897652...@longley.demon.co.uk>
Da...@longley.demon.co.uk# "David Longley" writes:
> In article <3581d898...@news.demon.co.uk>
> oh...@chatham.demon.co.uk "Oliver Sparrow" writes:
>
> > At the base, though, after tracking th e fringe of this
> > for four years now, I cannot see the point. What, behind all the fluff, does
> > this discriminate between, and how does this discriminant help me or anyone
> > else to achieve their goals? There seems to be no punch line.
> >
Who do you think you're kidding here? If you've genuinely only
"tracked the fringe" is it surprising if you "cannot see the
point"? Or are so so arrogant that you think that it should only
require you to ""track the fringe" or a complex issue in order to
understand what's being said.
If all of this could be said simply and applied with ease I
certainly wouldn't have put so much time and effort into it.
It's the fact that ordinary folk psychology just isn't up to it,
and is fundamentally unsound that makes it so difficult. But the
evidence is there for anyone who is prepared to look into what I
have said. Not only that, but once you look at it in context it
is difficult (from a rational perspective) to understand why
there is such resistance to putting it into practice - until one
appreciates the extent to which folk psychology guides so much of
natural judgement and how contextually constrained our skills
within the extensional stance actually are.
'Regardless of how much we stand to gain from supposing
that human behavior is the proper subject matter of a
science, no one who is a product of Western civilization
can do so without a struggle. We simply do not want such
a science.'
B F Skinner (1953)
Can Science Help? - The Threat to Freedom
(in Science and Human Behavior p.7)
David Longley
In article <3581334e.499949@news-server>
ZZZg...@stny.lrun.com "Jerry Hull" writes:
And you have the temerity to claim that what *I* say is logically
incoherent....
You really should try to grasp what's being said before posting
such nonsense. The point is that intensional contexts are
logically anomalous, not that they don't "exist". "Existence",
when used in the context of ontology, requires that the referent
takes on the value of a variable - one of the characteristics of
intensional contexts is that they are resistant to logical
quantification from the outside. Instensional contexts are a folk
psychological modus vivendi - for the extensional alternative one
looks to the ontology within which a science's professionals
operate.
I've provided that elsewhere - for those prepared to work within
it.
Neil Rickert
ZZZg...@stny.lrun.com (Jerry Hull) writes:
>So Longley finds himself in the awkward position that he is unable to
>disagree with his opponents without demonstrating the existence of the
>very thing he disputes.
And, of course, this is why Longley can never present a coherent
argument of his own, and mainly resorts to dumping the output of his
optical scanner (often in violation of copyright).
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>> In article <3581d898...@news.demon.co.uk>
>> oh...@chatham.demon.co.uk "Oliver Sparrow" writes:
>> > At the base, though, after tracking th e fringe of this
>> > for four years now, I cannot see the point. What, behind all the fluff, does
>> > this discriminate between, and how does this discriminant help me or anyone
>> > else to achieve their goals? There seems to be no punch line.
>Who do you think you're kidding here? If you've genuinely only
>"tracked the fringe" is it surprising if you "cannot see the
>point"?
Oliver cannot see the point, because there is no point (other than
allowing Longley to exercise his ego).
David Longley
If anyone seriously wants to understand more about what I have
written, please read what is available at my web site carefully
first, and then post specific queries/comments relating directly
to what is written there, copying the post to me as an e-mail.
Neil Rickert
oh...@chatham.demon.co.uk (Oliver Sparrow) writes:
> ric...@cs.niu.edu (Neil Rickert) wrote:
>"" I'll bet that Longley
>"" cannot define as much as a single extension -- not even the empty
>"" set -- without relying on intensions.
>I imagine that you are right, although there may be abstract or trivial
>things to be said. At the base, though, after tracking th e fringe of this
>for four years now, I cannot see the point. What, behind all the fluff, does
>this discriminate between, and how does this discriminant help me or anyone
>else to achieve their goals? There seems to be no punch line.
>There is a major effort afoot at the moment to base public health provision
>around what is called 'evidence based medicine'. I have some peripheral
>involvement with this, but organisations such as the Welcome Trust have just
>published synopses which make a great deal of sense. They leap frog what
>Longley is trying to say to its implications.
That sounds like what is happening in this country. What it amounts
to, is a political system for a crude system of rationing health
care, but done in such an obfuscatory manner that the politicians can
deceive their constituents and perhaps deceive themselves into the
belief that it is something other than rationing.
I have a lot of respect for the way this was done in the state of
Oregon, which was far more open about the fact that they were
developing a rationing system.
> So why is Longley publishing in this newsgroup rather than in a
>social science of public policy forum?
Deep confusion in Longley's thinking.
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>In article <6lpku4$s...@ux.cs.niu.edu> ric...@cs.niu.edu "Neil Rickert" writes:
>> Da...@longley.demon.co.uk (David Longley) writes:
>> >Note - the emphasis has been on the extensional stance with all
>> >of the strengths and merits of knowing what one is talking about
>> >as opposed to the demonstrable breakdown of logic within
>> >intensional contexts of propositional attitude.
>> Right. But it is all part of the grand Longley illusion.
>> Let's remove all intensions from language. Is there anything left?
>Well, if you read Quine carefully you'll see that comes pretty
>close to what he advises for scientific language.
Quine is quite clear that he takes science to be an extension of
common sense.
> It is
>the task of behaviour science to come up with alternatives to the
>idioms of propositional attitude - and this is something I think
>we, as a profession have already made some sort of progress
>towards.
No, it should be the task of behavior science to find a scientific
basis for human behavior. Any prior committment to religions beliefs
(such as extensionalism) can only interfere with this work.
>> Longley's illusion is that extensions remain. I'll bet that Longley
>> cannot define as much as a single extension -- not even the empty
>> set -- without relying on intensions.
>The success of the extensional system of science indicates that
>there is something radically wrong with your reasoning.
Longley is completely incapable of understanding the extent to which
the success of the physical sciences depends on common sense and
human judgement, and on the use of intensional language.
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>In article <3581d898...@news.demon.co.uk>
> oh...@chatham.demon.co.uk "Oliver Sparrow" writes:
>> ric...@cs.niu.edu (Neil Rickert) wrote:
>> "" I'll bet that Longley
>> "" cannot define as much as a single extension -- not even the empty
>> "" set -- without relying on intensions.
>> I imagine that you are right, although there may be abstract or trivial
>> things to be said. At the base, though, after tracking th e fringe of this
>> for four years now, I cannot see the point. What, behind all the fluff, does
>> this discriminate between, and how does this discriminant help me or anyone
>> else to achieve their goals? There seems to be no punch line.
>I suspect the reason why you don't see this is because you are
>not aware of the alternatives.
Longley offers no alternatives.
> COURSE TEXT:
> JACK: Meehl's study set off, or at least much
> inflamed, the 'statistical versus clinical judgment'
> controversy, which has rumbled on ever since, though
> it's somewhat less fashionable than it was.
> PENELOPE: Why?
This should be obvious. The overall level of scientific competence
among psychologists is so low, that abandoning human judgement in
favor of actuarial methods would often be beneficial. By contrast,
most physical scientists have long since learned how to combine human
judgement with formal methodologies so as to achieve effective
results.
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>In article <3581d898...@news.demon.co.uk>
> oh...@chatham.demon.co.uk "Oliver Sparrow" writes:
>> There is a major effort afoot at the moment to base public health provision
>> around what is called 'evidence based medicine'. I have some peripheral
>> involvement with this, but organisations such as the Welcome Trust have just
>> published synopses which make a great deal of sense. They leap frog what
>> Longley is trying to say to its implications.
>> ...
>> Boring, you say. True, but this is what the Longley Line takes one to when it
>> is applied.
>It may be where it leads when *you* apply it. I certainly haven't
>applied it in that context.
Your choice not to apply it in that context must have been
*intensional*.
Seth Russell
Oliver Sparrow wrote:
> if we choose to be rational, then the power base of the individual
> practitioner is removed, because they are now required to prioritise patients
> in certain ways, and to use specific treatments for them. The political
> choices which are implicit in current systems of rationing (and the hard
> cases which rigorous systems always throw up) are both severe tests of
> political will.
Thanks for shedding light on the real issues!
Rationality (reason) is a very broad word, too broad me thinks for our current
context. "Reason" used so broadly contains the biases (choices) of the author,
and as such are political, not objective scientific, statements. I was first
acquainted to this "reason" with a strap wielded by my father. I can now
evaluate my father's reasons and see which ones benefited me and which ones
benefited my father's selfish interests. Now you say: that is not the reason
of which we speak, we are talking about 2nd order predicate calculus where every
variable and context has been extensionaly verified and cannot be doubted. And
I will tell you, as has Godel, that no such objective reason yet exists. Which
is not to say that we should cast reason to the wind, but rather that we should
not worship it as another false God. Objective reasoning is a powerful tool that
we should use with heightened awareness as a bullet, not as a shotgun.
Why is this important now? Those of you who have used the computer at the
highest level of an organization, know why you used the computer. I'll bet you
used the computer to control a situation and the criteria for judging the
efficacy of the technology was whether the situation was changed to your
liking. The computer with its associated automated procedures is a powerful
tool against uncertainty, but it is also a shotgun that can be wielded against
creativity and the human values which are not in our best interests to
relinquish. The very real possibility of the emergence of robust AI makes it
imperative that we understand this tool from both perspectives.
PS: As to the health care issue you raised, if I were forced to decide today, I
would vote to be rational in that case.
David Longley
In article <6lrk4j$a...@ux.cs.niu.edu> ric...@cs.niu.edu "Neil Rickert" writes:
>
> > JACK: Meehl's study set off, or at least much
> > inflamed, the 'statistical versus clinical judgment'
> > controversy, which has rumbled on ever since, though
> > it's somewhat less fashionable than it was.
>
> > PENELOPE: Why?
>
> among psychologists is so low, that abandoning human judgement in
> favor of actuarial methods would often be beneficial. By contrast,
> most physical scientists have long since learned how to combine human
> judgement with formal methodologies so as to achieve effective
> results.
It's got nothing to do with the levels of scientific competence
of psychologists specifically. It's a demonstrable feature of
natural human reasoning, as anyone who is prepared to read the
material will discover.
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>In article <6lrk4j$a...@ux.cs.niu.edu> ric...@cs.niu.edu "Neil Rickert" writes:
>> > JACK: Meehl's study set off, or at least much
>> > inflamed, the 'statistical versus clinical judgment'
>> > controversy, which has rumbled on ever since, though
>> > it's somewhat less fashionable than it was.
>> > PENELOPE: Why?
>> This should be obvious. The overall level of scientific competence
>> among psychologists is so low, that abandoning human judgement in
>> favor of actuarial methods would often be beneficial. By contrast,
>> most physical scientists have long since learned how to combine human
>> judgement with formal methodologies so as to achieve effective
>> results.
>It's got nothing to do with the levels of scientific competence
>of psychologists specifically. It's a demonstrable feature of
>natural human reasoning, as anyone who is prepared to read the
>material will discover.
Yes, right. You forgot to mention that they must only read Longley's
approved reading list, and they must leave their own intelligence
behind while reading it. Otherwise the Longley indoctrination
program might fail.
Recentlty Longley suggested the book: Sutherland: "Irrationality: The
enemy within". I borrowed it and took a look. What a chuckle. It
should be great for quotes to use in the cocktail party circuit. If
the author is a professional psychologist, he should be required to
hand back his certificate. It just shows how, by taking a bunch of
perfectly rational behaviors and deliberately misconstruing them as
irrational, you can build a bogus theory of irrationality.
Why are psychologists, who are supposed to know all about self
delusion, so prone to delude themselves?
David Longley
The material he covers was accepted by editors of SCIENCE as
worthy of publication as major review artciles.
There is so much material, replicated under so many conditions
and in so many cultures that one must seriously question who is
the odd one out here....
Should we believe the results of thousands of research studies -
or the insights/rhetoric of one Neil Rickert....?
David Longley
Isn't it *more* likely that you have spent too long in a
community which is 1) unrepresentative of typical human ability
and 2) which actively shapes up the extensional stance as a
condition for membership?
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
>Should we believe the results of thousands of research studies -
>or the insights/rhetoric of one Neil Rickert....?
There is a third alternative, one which I would suggest:
Give considerable credence to those research reports which, loosely
speaking, could be described as extensional. That is, they report
observed behavior under controlled circumstances.
Be highly skeptical of those research reports which, loosely
speaking, are intensional. For example, be skeptical of those that
make assessments as to whether the test subjects acted rationally
or irrationally.
The trouble with Longley is that he is such a hypocrite. He swallows
intensional reports whole, no questions asked, and then uses them as
evidence to support his argument for extensionalism.
Neil Rickert
Da...@longley.demon.co.uk (David Longley) writes:
(Longley was responding to my earlier post)
>Isn't it *more* likely that you have spent too long in a
>community which is 1) unrepresentative of typical human ability
There may be some slight validity to this, although I think I am not
as out of contact with the sea of humanity as Longley might think.
>and 2) which actively shapes up the extensional stance as a
>condition for membership?
This, however, is quite wrong. Mathematical entities exist only in
the mind. As Longley uses the term, that makes them intensional.
Since the hard sciences make heavy use of mathematics, we must count
them as having a significant intensional component.
David Longley
It isn't a matter of asking *no* questions - it's one of not
asking *irrelevant* questions - something which probably only
comes through the years of study which it takes to become
proficient in any profession. The fact that you take great
offence at being told that you lack the requisite professional
skills is, I suggest, just a symptom of 1) your arrogance and 2)
your immersion in folk psychology. I have pointed you towards the
relevant literature, but to date, you keep reporting back not so
much that you disagree with it, but, through what you have to
say, that you just don't understand it.
You can take some comfort in the fact that you are not alone...