Dear Don,
Dear Don,
Silly question(s): How will the design field move forward and gain respect
if we borrow knowledge from other disciplines that, as your example on STM
suggests, is questionable or no longer valid? If we know (I know now after
reading your email) that the magical number seven theory has been debunked,
wouldn't we want to refer to updated knowledge?
Best,
María
---
María del Mar Navarro
PhD Fellow 2017
Interdisciplinary Studies, Graduate College
University of Arizona
On Wed, Mar 17, 2021 at 3:13 PM Don Norman <[log in to unmask]> wrote:
> External Email
>
> I write simply to reassure the readers of this ist that I will not engage
> in one of those contracted back-and-forth arguments.
>
> But one amusing comment (amusing to me, at least), triggers a completely
> different set of thoughts, having nothing to do with the implicit, tacit,
> non-declarative, experiential memory question.
>
> So consider this a new post about an irrelevant comment about the nature of
> science. I had fun writing the material below. Some of you will have fun
> reading it. If you are not interested, no problem: hit delete.
>
> -----------------------------------------
> Richard (Herriott) said that "A similar condition applies to the
> Gibson/Gregory debate." To which I responded, "huh"? After all, I am a
> reasonable friend of both Gibson and Gregory. Well, was a friend: they are
> now both dead. Yet I had no idea what their debate might have been. So I
> looked it up.
>
> Ah, Gibson was interpreted as believing in bottom-up processing. Gregory
> believed in top-down processing. Actually, whenever I would say the word
> "processing" to Gibson he would scowl at me and sometimes yell. As far as
> he was concerned it was "information pickup," from the light (sound touch,
> smell,...) to the person's response. No processing need be done. (There we
> go again: naming something and therefore thinking that it has been
> explained.. Gibson and I had many wonderful arguments, usually after dinner
> and a lot of wine, beer, or both whiskey and whisky. Gregory wasn't as much
> fun because we agreed about everything.)
>
> As for the controversy, like most controversial arguments between two very
> intelligent, sensible people, the answer is: both are correct.
>
> - Context, prior history, and expectations have a huge impact on our
> perceptions (Gregory). TRUE.
> - The signal itself has a huge effect on our perception. (Gibson). TRUE
>
> The answer (that is, what people believe today): It is both: A combination
> of top-down and bottom-up processing, plus considerable iteration (feedback
> loops).
> --
> I suspect that the various classifications of memory systems are completely
> wrong. This means that all my early work on short-term, long-term, and
> working memory is wrong. That's OK -- that means science has progressed.
> Mind you, the early work was not completely wrong -- think of it as simply
> very rough, first-order approximations. But the notion of a discrete and
> distinct set of different memory systems is much too simple-minded. Modern
> neuropsychology has yield dozens of distinctions. Which makes me believe
> that we've got it all wrong. However, we still do not know the final
> answer.
>
> --
> HOWEVER, as a designer, we simply need to know the rough-approximation.
> That's good enough for design. I have long pointed out that there is a
> model of short-term memory (STM) that is so wrong that nobody believes in
> it, but that nonetheless is of great value to designers. Here it is: STM
> consists of 5 slots, each can hold one item (five, not 7 plus or minus 2,
> that famous paper by George Miller was an earlier approximate model)*..
> Whenever a new item comes in, an old item (usually the oldest) is dropped.
> But what I tell people is that no psychologist who studies memory thinks
> that this model is correct. (Most would call it sheer nonsense.)
> Nonetheless, when designing something where we need to know what a person
> can remember in STM or in Working memory, this model is simple enough to
> use on our calculations and "good enough" in its predictability.
> ---------
> * If you don't know the famous paper, no problem. It was wonderful and
> insightful when it was written, but that was 65 years ago.
>
> https://en.wikipedia.org/wiki/The_Magical_Number_Seven,_Plus_or_Minus_Two
> The paper itself can be found at
>
> https://www.semanticscholar.org/paper/The-magical-number-seven-plus-or-minus-two:-some-on-Miller/4023ae0ba18eed43a97e8b8c9c8fcc9a671b7aa3
>
> ----------
> (What do I mean by "item"? Hah. An item in memory is "a meaningful,
> coherent unity": 5 numbers. or 5 words. or 5 well-crafted sentences. Yup.
> Pretty vague. I once claimed that it is five-pointers, and a pointer was
> pointing at some unitary concept in memory. But what is a pointer? We know
> what it is in a computer programming language, but how on earth would that
> be implemented in the brain. Answer: nobody knows.)
>
> Don
>
>
> -----------------------------------------------------------------
> PhD-Design mailing list <[log in to unmask]>
> Discussion of PhD studies and related research in Design
> Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
> -----------------------------------------------------------------
>
-----------------------------------------------------------------
PhD-Design mailing list <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------
|
