Re: Re. 7 plus or minus 2

["Nagai, Paul" <pnagai -at- VISA -dot- COM>]

> (I'd dearly love to obtain the
>  literature citation for any such study so that I can follow
>  up on this subject.)

My undergraduate degree was in psychology and I don't remember any studies
which applied this memory and learning research to practical applications
beyond the original scope of the study.

>  Has anyone followed up on Miller's
>  research to demonstrate that this principle extends beyond
>  memorizing phone numbers?

I don't recall that the research was specifically aimed at memorizing phone
numbers although the research may have covered this topic or been used to
support a 7 digit phone number. The study definitely examined more than just
numbers. Random strings of characters, words, objects, geometric shapes,
etc. were all used to arrive at the 7+/-2 figure.

>  This finding has been used and abused ever since to justify
>  all sorts of strange ways of limiting information (e.g.,
>  bulleted lists, steps in instructions, etc.) to groups of
>  no more than 7 points.

Before extrapolating the findings of this research it is critical to
understand that the original research examined *short term memory* which has
a very specific definition in the context of the study and any conclusions
drawn on the findings of the study. Short term memory is that memory used to
store and recall things for brief periods of time. I don't recall any
specific measures of time used to define short term memory, but storing a
phone number while calling directory assistance and recalling it after
hanging up and immediately attempting to dial it is a good example.

Long term memory is storing and recalling anything over a greater period of
time.  For example, if you are staying at a hotel for a week, you store your
room number in long term memory *not* short term memory (the directions to
that room given by the desk, however, are stored in short term memory) even
though you may forget it only a few days after checking out.

Of course, we used to joke that anything stored at 2am Friday morning,
recalled at 9am Friday morning, and forgotten by 11am Friday morning had
been stored in mid term memory ;-{)

The second critical thing to understand about this study is that the 7+/-2
items we are capable of storing refer to "chunks" (to use the terminology of
the study). A "chunk" is any item or collection of items capable of being
stored as a unit. For example, you recall that directory assistance is 411
... you don't recall that as 4-1-1 or three chunks. In fact, a phone number
is often broken into three or four chunks (counting the area code). As
another good example, think of URLs. The "significant" chunks I must
remember to find CareerPath are: web, careerpath, com. From that, I can
easily construct http://www.careerpath.com ... if I were a newbie ... I'd be
lost without a pen and paper.

How people chunk and how well they chunk can vary wildly. If you are
learning about a topic which you know well, you will chunk information in
very large (relatively speaking) chunks and retain much, much more than
someone who knows nothing about the subject and who can only chunk little
chunks, if any at all.

All this supports the idea that you must know your audience, you must know
your subject, and you must know *how* you intend to communicate. Your
readers will never need to digest an entire reference guide. Your readers
might, however, need to read and retain "Five Things to Never Say or Touch
While Visiting Japan" as they walk by your sign.

 ------
Paul Nagai
Visa International
pnagai -at- visa -dot- com
415-432-1101
415-432-3678 fax