Trendy science

[From Bill Powers (920713.1730)]

Rick Marken (920712) --

Welcome back. I'm too flabbergasted by the reaction to your "Blind men"
paper to speak of anything else.

I found out why Psych Review didn't even send my "Blind men" paper
out for review. According to the editor it was because:

"It would need to speak more directly to current psychological issues
and theorizing. One would need to see more clearly a connection
between what you are talking about and the issues that dominate
psychological theorizing today."

What could be a better illustration of the trendiness of psychology? Only
13 years ago, I published an article on control theory in that very
journal. Already it has passed over the horizon and is no longer an "issue
that dominates psychological theorizing today."

Can you imagine what physics would be like with an event horizon of only 13
years? Physics would no longer be concerned with inverse square laws,
optical refraction, the Hubble Constant, gas laws, electrical phenomena,
lasers, transistors, or the Mossbauer Effect. Instead of building up a
coherent and growing picture of nature, physicists would be worrying about
whether they're working on things that are popular and current. There would
be as many schools of physics as there are of psychology, sociology, or
economics.

This is confirmation of my thesis that a science based on low-probability
facts can't create a coherent picture of nature. When only specific effects
under specific circumstances are studied, facts lie scattered around the
landscape in disconnected confusion. No argument involving more than a
small handful of facts can lead to deductions with a probability of truth
greater than chance. Reasoning is limited to the fourth-grade level. This
is why the great majority of observations that pour out into the literature
are forgotten the day after they are published, not to mention 13 years
later. There is no underlying body of understanding to which each new
observation adds, or even potentially adds. Today's hot subject is
tomorrow's phrenology, disappearing when the careers of those making a
living off of it end. There is no science of psychology. There's only a
Psychology Club, and its newsletter is called Psychology Today. I suppose
that it might have a Nostalgia Column titled "Three years ago this month."

In disgust,

Bill P.



[Martin Taylor 920713 15:15]
(Bill Powers welcome back to RIck Marken 920713.1730)

I'm still listening, but not contributing unless tweaked, until probably
this weekend or next week.

"It would need to speak more directly to current psychological issues
and theorizing. One would need to see more clearly a connection
between what you are talking about and the issues that dominate
psychological theorizing today."

What could be a better illustration of the trendiness of psychology? Only
13 years ago, I published an article on control theory in that very
journal. Already it has passed over the horizon and is no longer an "issue
that dominates psychological theorizing today."

I have a certain sympathy with both sides of this. If the editor really meant
what he said, then Bill's comment is quite justified, along with the rest of
it. But I prefer to read the comment in a more social sense (and I don't
mean Psychology Club and Nostalgia column). As I see it, the requirement is
to get people to read. To do that, one has to give them something they can
perceive as relevant to their interests. I have no doubt that they SHOULD
find PCT relevant to their interests, no matter which facet of psychology
they work on. That doesn't mean that they know that they should. So the
editor's comment makes a great deal of sense. As far as he/she is concerned,
to use up the valuable print space on something that the readers will simply
pass over is a waste, no matter how valuable the article might be found to be
ten years from now.

Seen in this light, I don't think the "trendy science" comment is justified.
All science is trendy. Science is a sociological function, based on the
belief structures held by scientists, and that includes beliefs about where
important new things will happen. You can't fault them for not agreeing
that your own position is such a place, just on your own say-so. You have
to show them in their terms, not yours.

Also, editors are people, and control for the stability of their own views.
I have had papers refused by editors who insisted on the use of significance
statistics, which I abhor (doesn't that sound funny, given my insistence that
statistics are/is very important). I refuse to publish a significance level,
and if an editor won't accept that, I go elsewhere or keep the paper in a
drawer. If I see a significance level in a published paper, my first thought
is that the author doesn't know what the data say.

This is confirmation of my thesis that a science based on low-probability
facts can't create a coherent picture of nature. ...
No argument involving more than a
small handful of facts can lead to deductions with a probability of truth
greater than chance.

An amusing self-contradiction! Also a take-off point for an argument. Low
probability facts do not contribute to a logical argument, but they can pool
to generate high-probability facts. Most perception works that way, I strongly
believe. Perception is not deduction.

Martin

[From Rick Marken (920713.1300)]

Bill P. -- Thanks for the sympathetic comments on the "Blind men" paper.

I'm tempted to forward your "trendy science" remarks to Kintsch (the
editor of Psych Review). But I'm pretty tired of arguing with these
people. Unfortunately, I'm also running out of places to send my papers;
places that might be willing to publish "non-trendy" science. Maybe I'll
just start saving these papers up for "Mind readings II".

By the way, does anyone out there have any idea what Kintsch might mean
by "current psychological issues and theorizing"? Maybe if I knew
for sure what these were I could add a sentence or two to the "Blind men"
paper relating control theory to these current issues and theorizings.

Hasta luego

Rick

···

**************************************************************

Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)

[From Rick Marken (920713.1500)]

Martin Taylor (920713 15:15) says

As I see it, the requirement is
to get people to read. To do that, one has to give them something they can
perceive as relevant to their interests. I have no doubt that they SHOULD
find PCT relevant to their interests, no matter which facet of psychology
they work on.

Unfortunately, the relevance of PCT to their interests is quite negative. PCT
shows that most psychologists are interested in an illusion of one kind or
another -- the illusion of control by reinforcement, the illusion of external
causation of behavior, the illusion of internal programming of behavior.

So the
editor's comment makes a great deal of sense.

From a PCT perspective, I suppose so.

As far as he/she is concerned,
to use up the valuable print space on something that the readers will simply
pass over is a waste, no matter how valuable the article might be found to be
ten years from now.

I agree that this appears to be the basis for the rejection -- but I find
it appalling; I can't believe that the role of an editor of a scientific
journal is to attract readers; National Enquirer, yes; Psychological
Review, no! I would hope that the editor of a scientific journal would have
the integrity to "waste" precious print space on non-trendy science if s/he
thought the article actually made a valuable contribution. I get the impression
that this is indeed the way some editors go about their business -- Estes
for one, bless his fair (but unquestionably conventional s-r) heart.

Seen in this light, I don't think the "trendy science" comment is justified.

In this light it seems just as trendy -- the editor is evaluating on the
basis of what will "sell" in the current market, not on scientific merits.

All science is trendy. Science is a sociological function, based on the
belief structures held by scientists, and that includes beliefs about where
important new things will happen. You can't fault them for not agreeing
that your own position is such a place, just on your own say-so. You have
to show them in their terms, not yours.

There is no place in the "blind men" paper where I ask the reader to accept,
on the basis of my say-so, that mine is a new, important position. Instead,
I present an analysis of a closed loop negative feedback system -- an
analysis that the reader if free to question and test -- and show that aspects
of the behavior of this system look like s-r, reinforcement or cognitive
behavior (the latter being "their" terms for these types of behavior). So
I think I have tried to show "them", in their own terms, that what they consid-
er behavioral phenomena are (possibly) different perspectives on closed loop
perceptual control. I suggest that, if this is the case, then they can only
understand behavior if they start trying to figure out what perceptual var-
iables the systems is controlling. That's what the paper is about. It was
short, sweet and to the point. It seemed to be "relevant to what psychologists
care about" -- understanding behavior. So I don't understand what Kintsch
could have meant by his justification for not having the paper reviewed.
My guess is that he felt that the issue raised by my paper was not interesting
to him (and possibly to most other psychologists who already KNOW what
behavior is). That is a very poor basis for deciding what get's disseminated
to the scientific community.

I don't really know what I could have done to make the "Blind men" paper
more publishable, given Kintsch's criteria (and your interpretation of
them). Do you think there is a way to re-write the paper so that it could
meet these criteria? Any suggestions would be greatly appreciated.

Regards

Rick

···

**************************************************************

Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)

[from Joel Judd 920714.1255]

Rick laments:

Unfortunately, I'm also running out of places to send my papers;
places that might be willing to publish "non-trendy" science.

I for one would enjoy publishing ANYPLACE. I am awaiting what will no doubt
be interesting replies to an attempt to break into the SLA literature.
Which brings me to an alternative: you could try to cross fields and
co-publish with someone else on CSG-L, for example. Do you want to try out
language acquisition? There are still some editors that will entertain
pretty far-out ideas (for which label PCT seems to qualify). I'd be happy
to entertain any possibilities that interest you.

[I would have sent this direct but I'm not using my own e-mail disk. You
can reply just to me if you wish]

[From Rick Marken (920714.1300)]

Joel Judd (920714.1255) says:

I for one would enjoy publishing ANYPLACE. I am awaiting what will no doubt
be interesting replies to an attempt to break into the SLA literature.
Which brings me to an alternative: you could try to cross fields and
co-publish with someone else on CSG-L, for example.

You are right Joel -- I should be glad that my PCT stuff has been published
at all. I would be happy to try to cross fields and/or co-publish (I've done
that with Bill P.; I'd like to work with others on CSG-L too). I'm just
kvetching because I'm tired of trying to get published by wriggling through
hoops in order to make PCT palatable to editors and reviewers. When editors
or reviewers catch factual errors then I'm happy to change stuff. But when
they want changes that change the meaning of the paper (like the reviewer
who suggested that I make it clear how the target guides behavior) it becomes
tiresome. I think PCT should have reached the point by now where we can just
put our work before the public and assume that that public either has been
or can get educated about PCT. I don't want to have to write a "what PCT is
about and why you should care" section every time I submit a PCT paper for
publication. That's why I won't resubmit the "Behavior of perception" paper
to journals that say "we'd accept it if you could just explain how this
fits into conventional psychology". I don't think people who try to publish
papers based on other, "trendy" theories of behavior are made to jump through
such hoops. I'm just tired of having PCT treated differently than the
"trendy" stuff; and I'm tired of jumping through the hoops.

Although I'm willing to cross fields in order to publish -- I also think
that it is important to hit the right audience. I submitted to Psych
Review, for example, because I thought the "blind men" paper was most relevant
to an audience of theoretical psychologists. Obviously, my thoughts were
not consistent with those of the editor.

Hasta luego

Rick

···

**************************************************************

Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)

You people seem to have lost sight of the fact that you can still be
thought of as working within the broader systems science/cybernetics
fields. I can point you to a number of journals and conferences that
would probably be happy to have you, and can personally refer good
papers to some editors.

Many of you are aware of Bill's and others' experiences with the ASC.
But there are others as well.

It is true that the Systems literature is pretty much a ghetto, with a
low signal/noise ratio and a relatively high crackpot ratio. But at
least they're open to otherwise far-out views (not that I think CT is
REALLY far-out). But if you're REALLY desparate, it IS available.

O----------------------------------------------------------------------------->

Cliff Joslyn, Cybernetician at Large, 327 Spring St #2 Portland ME 04102 USA
Systems Science, SUNY Binghamton NASA Goddard Space Flight Center
cjoslyn@bingsuns.cc.binghamton.edu joslyn@kong.gsfc.nasa.gov
^^^^^^ !NOTE NEW EMAIL! ^^^^^^

V All the world is biscuit shaped. . .