Magical numbers: the seven-plus-or-minus-two myth

Jean-Luc Doumont denounces the 7plus or minus 2 myth in this article (available from IEE as a PDF, membership required) in IEEE Transactions on Professional Communication. Doumont calls the myth a conventional absurdity that most people cite out of context.

Most interesting to me is Doumont's identification of this detail quoted from George Miller's paper in Psychology Review, "For memory, a chunk of information is loosely defined as, precisely, one of those items that the immediate memory can hold up to seven of. As Miller puts it, “The span of immediate memory seems to be almost independent of the number of bits per chunk”

Jean-Luc Doumont denounces the 7plus or minus 2 myth in this article in IEEE Transactions on Professional Communication. Doumont calls the myth a conventional absurdity that most people cite out of context because they've heard it somewhere. The article reviews George Miller's original article in Psychology Review to put it in perspective.

Most interesting to me is Doumont's identification of this detail quoted from Miller, "For memory, a chunk of information is loosely defined as, precisely, one of those items that the immediate memory can hold up to seven of. As Miller puts it, “The span of immediate memory seems to be almost independent of the number of bits per chunk." The crux of Miller's message is that "Our capacity for processing information and, specifically, our span of unidimensional absolute judgement is severely limited. To communicate effectively, then, we should take this limitation into account and, rather than attempting to quantify it."

Miller proposes that there are various devices one can use to get arround this limitation of the mind. One such device is to arrange the task in such a way that we make a sequence of several absolute judgments in a row. The example Doumont gives is recoding bits into chunks to expand our span of short term memory. Given the two groups of items below, if the 3x3 list of items is meaningfully chunked, it will be easier to process than the 9 item list below:

    * item
    * item
    * item
    * item
    * item
    * item
    * item
    * item
    * item   * item
    * item
    * item

    * item
    * item
    * item

    * item
    * item
    * item

What I took away from the article was that some visually presented depth helps the mind to deal with a broad list that might be harder to hold in memory. Doumont goes on to describe attributes of smaller integers, which was interesting, but was not of any practical use to me.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

10% Brain Use Myth

This reminds me of the "We only use 10% of our brains" myth that a lot of people like to quote as a fact.

The 7+-2 principle applies to

The 7+-2 principle applies to the limits of short-term memory. Applying it to the number of links on a page is a stretch. Developers, looking for rules to follow, have glommed onto it without really understanding the original principle.
And this: "The span of immediate memory seems to be almost independent of the number of bits per chunk." That means that "731-8014" is 7 chunks (unless your own number starts with 731). 1-800-555-1000 is 3 chunks. A chunk is an individual remember-able blob of information. Your 3-groups-of-3-items-each example is a very good one.

Can I read it? and more on the matter

Yeah, I think you have to be a member

I think you have to be a member (will share with you though).