Re: Richard says Rupert Morrish is wrong

Seanpit wrote:
On Dec 22, 10:30 am, richardalanforr...@xxxxxxxxxxxxxx wrote:
On Dec 22, 7:44 am, eerok <krke...@xxxxxxxxxxxx> wrote:
Seanpit wrote:


[...]

Useful scientific conclusions based on valid
statistical analysis need not be published or even
written down.


I for one wouldn't mind a bit if you explored this "not
writing down" concept more fully.


Mind you, I'm rather taken by the idea of rats carrying out
statistical analyses. I wonder if Sean has access to a
breed of superintelligent rodents. But then he insists that
we all carry out such analyses all the time *without even
knowing it*!


You don't have to be superintelligent to carry out a useful
level of statistical analysis. All you have to do is be
able to remember past events. Even rats can do this. If the
location of the cheese in the maze is generally in a
particular place, even if that place is not always the
location of the cheese, it has been shown that rats can
figure out that it saves time, most of the time, to check
this location first. That, my friend, is a valid use of
statistical analysis. Statistically, the odds favor the
appearance of the cheese in a specific location - - and the
rats were in fact able to figure this out without writing
anything down or consulting any other rats.


You are stretching the term "statistical analysis" so that it
encompasses just about any kind of mental processing. Today
it's raining -- I arrived at that conclusion by statistical
analysis? I use statistical analysis to step around puddles?

I think the term you're looking for most often is "guess,"
sometimes "recognize," and sometimes "judge."

And I hope it was an accident of context that led you to
suggest that rats arrive at useful scientific conclusions.


You are also able to do the same thing. That is why you
know that a highly symmetrical polished granite cube is in
fact "anomalous" from the perspective of non-deliberate
origin. In fact, it is so anomalous as to be clearly
artifactual for most people - to include most scientists
(even if they were on Mars or some other alien planet).


If your cube fell out of the sky and onto my lawn, I wouldn't
know what to make of it. I'd hope it wasn't deliberate,
because if it was I'd have to get mad at somebody. But what's
any of this supposed to mean?

If no one knows who made it or where it came from, then
that's the right answer. It's not as if it's encumbant on
anyone to come up with a specific answer based on no data. Of
course people like to speculate, but that's just guessing.

This notion of yours that we can calculate statistical models
on the fly that actually mean something, when in fact we don't
know anything much about what we're looking at, is quite a
leap into nonsense.

It's just guessing, and if you can't test it, then it's not
science. You're trying to make guesswork seem like a
methodology -- sort of like science but with some kind of
intuitive magic replacing the working parts. In fact, it
amounts to nothing more than an argument from incredulity.


The same thing is also true for your tool marks. You don't
need to write down your past experiences with tools and tool
marks to be able to recognize the markings of a particular
tool as being anomalous or "artifactual" right away. You
are able to do this because your memory is good enough at
remembering past events and adding up the numbers that you
don't need to make a written record of them before you can
use them to help you make useful predictions.


In the case of tool marks, you can test them by comparison.
When you have nothing to test, then you have to admit that you
don't know; or in your case it appears you make something up.

I fail to see how statistics has anything to do with this.


That, again, is a valid use of statistics. No one ever said
that statistics could only be done with pen and paper. Very
useful statistics can be done entirely in ones own head.
Arguing otherwise means that you don't understand the basic
nature of learning itself. Basic learning is based on
statistics.


No, it's not a use of statistics at all. As far as I can
tell, "statistics" for you is a neat-sounding word that you
hope will add credence and substance to an argument severely
lacking in both.

--
"It is not enough to conquer; one must learn to seduce."
-- Voltaire

.

Relevant Pages

  • Re: Richard says Rupert Morrish is wrong
    ... Even rats can do this. ... always the location of the cheese, it has been shown that rats can ... this out without writing anything down or consulting any other rats. ... statistics could only be done with pen and paper. ...
    (talk.origins)
  • Re: Richard says Rupert Morrish is wrong
    ... Even rats can do this. ... this out without writing anything down or consulting any other rats. ... The same thing is also true for your tool marks. ... That, again, is a valid use of statistics. ...
    (talk.origins)
  • Re: Richard says Rupert Morrish is wrong
    ... Even rats can do this. ... this out without writing anything down or consulting any other rats. ... The same thing is also true for your tool marks. ... That, again, is a valid use of statistics. ...
    (talk.origins)
  • Re: POTM nomination: Nothing but a fairytale
    ... The Earth has no moon. ... propablilities only show how unfit you are to do statistics. ... statistical analysis. ... starting point to come to these specific sequences. ...
    (talk.origins)
  • Re: Statistical Pattern Recognition
    ... the recent brain research is the *evidence* that shows brains perform ... statistical analysis during pattern recognition? ... that it is an *associative memory* process, not statistical analysis. ... I've already provided several references dealing with the statistics ...
    (talk.origins)