Re: MR Evidence

From: Bill Spight (
Date: Tue Aug 07 2001 - 22:08:17 BST

  • Next message: "Re: Teleology"

    Received: by id WAA06717 (8.6.9/5.3[ref] for from; Tue, 7 Aug 2001 22:10:24 +0100
    Date: Tue, 07 Aug 2001 14:08:17 -0700
    From: Bill Spight <>
    Subject: Re: MR Evidence
    Message-id: <>
    Organization: Saybrook Graduate School
    X-Mailer: Mozilla 4.72 [en]C-CCK-MCD {Yahoo;YIP052400}  (Win95; U)
    Content-type: text/plain; charset=us-ascii
    Content-transfer-encoding: 7BIT
    X-Accept-Language: en
    References: <3B6ECBC0.14995.5A4B18@localhost> <002101c11f77$62cc25c0$f188b2d1@teddace>
    Precedence: bulk

    Dear Ted,

    > Let's begin with the rats.
    > In 1920 William McDougall of Harvard began training rats to learn to escape
    > from a water maze by choosing the correct exit. While the brightly lit exit
    > would give them an electric shock, when they picked the dimly-lit exit, they
    > got out undisturbed. McDougall found that the first generation of rats had
    > to endure 165 shocks before getting the message. But by the 30th
    > generation, only 20 transgressions were necessary to persuade the rats of
    > the error in their way. (McDougall, 1938. British Journal of Psychology
    > 28:321-345.)
    > McDougall assumed the rats were passing on acquired characteristics.
    > Wishing to disprove this "Lamarckian" (and Darwinian) interpretation of the
    > data, F. A. E. Crew replicated the experiment in Edinburgh. Right from the
    > get-go, Crew's rats needed only 25 errors to learn their lesson, as if
    > picking up where the Harvard rats had left off. (Crew, 1936. Journal of
    > Genetics 33:61-101.)
    > In Melbourne, W. E. Agar found the same effect. His trials went on for over
    > twenty years, and even when he tested control subjects that weren't
    > descended from trained rats, they still showed improvement over the
    > performance of previous generations. So it couldn't have been coming from
    > their parents. (Agar, 1954. Journal of Experimental Biology 31:307-321.)

    I suspect that the difference has to do with human culture. By the time
    I was in college, labs were brightly lit places, but I suspect that labs
    of the early 20th century were not so brightly lit, given the difficulty
    of installing wiring and lighting in old buildings that were not built
    for them. What difference does that make? A common error in experiments
    with rats running mazes has been not isolating the mazes from ambient
    stimuli. The rats use cues outside the maze. Possibly better external
    lighting, installed in McDougall's lab between 1920 and 1938, helped the
    later rats, and the other labs were also better lit.

    > Acquired traits have often been observed to pass
    > throughout a species with no known means of direct transfer from individual
    > to individual. For instance, in England in the 20s a small bird known as
    > the blue tit learned to open milk bottles at doorsteps. When one bird
    > learned the trick, others in the area learned it by simple imitation. But
    > the blue tit doesn't fly more than a few miles, and this habit spread to
    > several widely disparate areas in England by 1935 and continued popping up
    > in faraway places throughout the forties, including Scandinavia and Holland.
    > The habit appeared independently at least 89 times in the British Isles, and
    > the spread of the habit accelerated as time went on.

    Why should we think that these appearances were independent, especially
    as relevant memes might have been passed from neighbor to neighbor?

    > (Fisher and Hinde,
    > 1949. British Birds 42:347-357.) Milk bottles practically disappeared in
    > Holland during the war, and by the time they returned all the birds that had
    > been opening them before the war could not have survived to see their
    > return. Yet the habit rapidly returned when the bottles were re-introduced
    > in 1947.

    So what?

    > According to Sheldrake's model, the more a new trait is practiced
    > by the members of a given species, the more likely other individuals will
    > pick it up through resonance.
    > Arden Mahlberg, a psychologist, carried out a test of the ability to learn
    > Morse Code. He had one group of subjects learn actual Morse Code, while
    > another had to learn a newly-invented code that closely resembled it. He
    > found that subjects were able to learn the actual code far more rapidly than
    > the alternative, and he interpreted this as evidence that the subjects were
    > resonating with the millions of people who had already learned Morse code.

    But the Morse code should be easier to learn, even than a similar code.

    > Each time he replicated the experiment, he found that the difference in
    > learning time between Morse code and the new one progressively decreased.
    > This might mean that the initial results were false. But the fact that the
    > decrease was progressive suggests that the morphic resonance of the new code
    > was becoming progressively stronger as more and more students learned it.
    > (Mahlberg, 1987. Journal of Analytical Psychology 32:23-34.)

    It might also mean that there were experimenter effects that were not
    controlled for.

    > Countless people have learned to type on the QWERTY keyboard. If morphic
    > resonance is real, we should expect people to learn this layout more readily
    > than random layouts. This is indeed the case. Even the alphabetical
    > layout, which should be easier to learn, is often harder to learn, though in
    > a few experiments it was equally easy to learn as the QWERTY layout.
    > (Norman and Fisher, 1982. Human Factors 24:509-519.) (Hirsch, 1970. Journal
    > of Applied Psychology 54:484-490.)

    Well, as I recall, the Dvorak keyboard is easier to learn than the
    QWERTY keyboard. But the QWERTY keyboard should be easier to learn than
    a random keyboard or an alphabetic keyboard. The QWERTY keyboard was not
    chosen at random, although it is planned to be a bit difficult.

    The problem with doing such experiments with Morse code and the QWERTY
    keyboard is that they were designed to be easy to use, and thus it is no
    surprise that they are relatively easy to learn. Better to take
    something that many people have learned, such as the Japanese syllabary,
    and compare learning it with learning other symbols that are as easy to
    write. Have the teaching done by a computer program, so that no teacher
    is learning how to teach them.



    This was distributed via the memetics list associated with the
    Journal of Memetics - Evolutionary Models of Information Transmission
    For information about the journal and the list (e.g. unsubscribing)

    This archive was generated by hypermail 2b29 : Tue Aug 07 2001 - 22:14:39 BST