I ' d like to start a discussion about the value of outlining software and human short time memory. As a lawyer, I have to remember at least hundreds of hierarchically structured checklists and outlines and used for a cuple of time an outliner of a german criminal law professor ( http://www.normfall.de ). In his theory, Prof. Haft explains that lawyers should build structures of basically not more than 7 items ( one parent and not more than six childs ), arguing that doing otherwise would not fit to the human short time memory's capacity as revealed the research done by George Miller in 1956 about our limitation of processing information ( " The magical number seven - plus or minus two " ).

Therefor an outline should not have more than 6 main chapters, each of them having not more than 6 sub-chapters. The amount of the sub - levels should not grow larger than 7. Building bigger chunks of information does not fit in our limitated short time memory and will not be properly processed.

Working with this limitation of our short time memory in mind helped me to remember my mind-maps in an adequate and complete way. Understanding the brain's chunking mechanism is necessary for everyone working with outliner and mind-mapping software.

Hi Guido,

I'm not sure I grasp what you're getting it. Isn't the whole point of outliners and other information management systems to be able to dump information out of the brain so that no information gets lost due to our poor in-built memories and processing systems?

We have short attention spans and small short-term memories, so I understand why marketing messages to the public or, maybe in your case, points for the jury, need to be short, make impact and be "sound-bite"-like; but I can't see how this would need to translate into limitations in, or optimal numbers of, nodes in outliners.

If you need to memorise lists, that, I would imagine, is another matter!

Mind you, I confess to rarely using pure outliners - I use Flying Logic sometimes to sort out rambing thoughts. I don't notice any more or fewers problems in finding my way around software by number of nodes.

Graham

Graham Rhind wrote:

I'm not sure I grasp what you're getting it. Isn't the whole point of
outliners and other information management systems to be able to dump information
out of the brain so that no information gets lost due to our poor in-built memories and
processing systems?

Not necessarily; one may use such systems in order to better structure information with the end goal of _memorising_ it. Or, as in my case, of better delivering it to an audience.

So indeed I do get Guido's point. Tony Buzan notes two important advantages of mind maps: (a) better understanding of information and therefore (b) better recall of that information. For example, he suggests making mind maps of whatever books one reads.

Buzan also suggests that mind maps (that in terms of structure are nothing more than outlines) are more suitable for the human brain to comprehend and recall, because its neural layout is similar. Furthermore, he suggests complementing item titles with colours and images to increase their impression on the mind.

I don't know whether neural research supports Buzan's claim on the layout, but I believe it does support the one on images. So, going back to Guido's initial point, I'd say that it sounds quite reasonable and complementary: structure information in groups of no more than six per level, show it two-dimensionally (in a mindmap), and increase the impression with images and colours.

By the way, the 'no more than six' rule is often used in powerpoint "six points with six words each" resulting in some really dumb presentations (see also http://www.wired.com/wired/archive/11.09/ppt2.html ) so I find outlines and mindmaps much preferable.

alx

Hi Alex,

I ' d like to add one important advantage of Mind-Maps that indeed are more than outlines : they reflect the way we think. We think in a network of pictures, feelings, sounds and information expressed in different manners. Though, a Mind-Map reflects better this network of different bits of information in our brains connected by billions of associations one to another.

That means in terms of outlining that we as human beings do have to communicate linearly, being restricted to a chronological order of presenting our ideas by speach or text. An outline is a good medium for presenting and communicating information. But it is not in terms of thinking, remembering or learning. In an outline, one has to process information in the following way : First 1., than 1.1, than 1.2, than 1.3, than 2, than 2.1 etc. etc. The brain thinks 1 and than at the same level : 1.2, than perhaps 1.1 and than 1.3 - the sequence simply doesn't matter, what matters is the connection between 1 and its childs. It's on the connection that a Mind-Map focusses. That's what Buzan calls "radial thinking" and I invite everyone to make a short break and to contemplate the own thinking process.

If we would think in outlines, we would process and think of a lot of unneccessary information ( nodes in our mental outline ) in a given situation before reaching the essential point. Fortunately, we do not think in this way but jump in a few seconds to the only interesting information node.

Guido wrote:

Hi Alex,

I ' d like to add one important advantage of Mind-Maps that indeed are more
than outlines : they reflect the way we think. We think in a network of pictures,
feelings, sounds and information expressed in different manners. Though, a
Mind-Map reflects better this network of different bits of information in our brains
connected by billions of associations one to another.

That means in terms of
outlining that we as human beings do have to communicate linearly, being restricted
to a chronological order of presenting our ideas by speach or text. An outline is a good
medium for presenting and communicating information. But it is not in terms of
thinking, remembering or learning. In an outline, one has to process information in
the following way : First 1., than 1.1, than 1.2, than 1.3, than 2, than 2.1 etc. etc. The
brain thinks 1 and than at the same level : 1.2, than perhaps 1.1 and than 1.3 - the
sequence simply doesn't matter, what matters is the connection between 1 and its
childs. It's on the connection that a Mind-Map focusses. That's what Buzan calls
"radial thinking" and I invite everyone to make a short break and to contemplate the
own thinking process.

If we would think in outlines, we would process and think of a
lot of unneccessary information ( nodes in our mental outline ) in a given situation
before reaching the essential point. Fortunately, we do not think in this way but jump
in a few seconds to the only interesting information node.

Guido is correct about thinking, but the mind maps usually seen are, in fact, just pictorial representations of outlines. A web-type diagram, such as The Brain is a better representation, ofr thought because it allows subitems with different parents to be linked together. Outlines and other PIMs that allow unlimited linking among different items anywhere in the outline or PIM also represent the type of thinking Guido describes.

The problem with web-type diagrams or unlimited linking is that one can soon become completely lost, even if there is a trail of breadcrumbs to help one find one's way back. Many years ago, in DOS days, I had such a text program [possibly named Black Magic]. I couldn't keep track of where I was.

Although I dislike most mind maps, The Brain has the great advantage of one being able to chose any node as one's "center of the universe."

-c
-c

That is because we don't seperate here between thinking how it can be and how it should be. Every thougt can be in theory connected to every other thought, but in terms of good scholar thinking that should not be so. Even if hyperlinks can be useful, one should be aware that hyperlinks are not necessarily translated into meaningful connections between thoughts by your brain ( my main argument against the wiki approach of connected text ).

Guido wrote:

That is because we don't seperate here between thinking how it can be and how it should
be. Every thougt can be in theory connected to every other thought, but in terms of good
scholar thinking that should not be so. Even if hyperlinks can be useful, one should be
aware that hyperlinks are not necessarily translated into meaningful connections
between thoughts by your brain ( my main argument against the wiki approach of
connected text ).

I agree completely! That is why one should be careful and "stingy/minimalist" in creating hyperlinks. I, too, fear to use the wiki approach for the same reason. Of course, some might say that wikis or an abundance of hyperlinks might support "thinking outside the box," but I doubt it.

Speaking of "thinking outside the box," here is a recent, marvelous quote by Warren Buffett:

"I've reluctantly discarded the notion of my continuing to manage the portfolio after my death - abandoning my hope to give new meaning to the term 'thinking outside the box.'"
--Warren Buffett on his eventual passing and new management of Berkshire Hathaway, 2/29/08

-c

From the little I've read, Buzan was a believer in the idea that arbitrary connections between items didn't really correspond with how people actually think. His thesis was that people think mostly in terms of hierarchies (outlines) but can benefit from some spatial information added to those hierarchies.

I think he's probably right. The human mental instinct to think in terms of hierarchies is very strong. It seems to be related to our instinct for pattern recognition. Arbitrary collections of links seem foreign.

You see this everywhere. Programmers instinctively lean towards tree controls in their user interfaces. Doctors prefer the concept of a hierarchical "differential diagnosis" over Bayesian reasoning, etc.

-- Chris

I am far from an expert on memory or learning, but I'd still like to throw in my thoughts on this subject.

First of all, I am uncertain of the value in worrying about short-term memory. That's not learning or understanding. It enters our minds for a short time then goes away. As it should.

Isn't it long-term memory and understanding that we are talking about here? Or am I missing something? -- entirely possible.

Learning and even long-term memory is about context. When I understand the context of something, I am more likely to remember it and to "get it." Of course, context can come in many shapes and sizes. For instance, if I wanted to learn about the planets of our solar system, I could think of them by their distance from the sun, or by their relative sizes, or by the type of atmosphere each may have. In fact, I'd like to have all this information about each of them available... A program like MyInfo would allow me to build a list of the planets with columnar meta data on each of the three characteristics I listed above. Then I could sort them by any of those characteristics. I don't believe I would be able to view a mind map of the planets in the same way -- that is with different contexts.

Steve Z.

I agree with Steve, the most important thing is to understand first, before you try to learn sth.
The problem is when there isn't much to "understand", and one has to purely memorize things, sets, ...

In fact, it's quite interesting that when one has to learn a "set of information", it is easier to remember the "ordered set", which in fact contains more information than the unordered set. Like the example with planets in our solar system, usually people learn them ordered by the distance from the Sun.

When building my knowledge base, I often consult the SuperMemo website, the articles on memory and knowledge management are excellent, http://www.supermemo.com/

Guido and Cassius,
can't help myself, but I need to respond:
(i) Even if hyperlinks can be useful, one should be aware that hyperlinks are not necessarily translated into meaningful connections between thoughts by your brain ( my main argument against the wiki approach of connected text ).

(ad i) I couldn't agree more. But the idea is not to translate hyperlinks "into meaningful connections between thoughts," but to translate (possibly) meaningful connections between thoughts into hyperlinks (to test and explore them) or to create a hierarchy of some sort

(ii) I agree completely! That is why one should be careful and "stingy/minimalist" in creating hyperlinks. I, too, fear to use the wiki approach for the same reason. Of course, some might say that wikis or an abundance of hyperlinks might support "thinking outside the box," but I doubt it.

(ad ii) If (i) is the only reason for the fear, it is no reason, because that's getting it "back a...wards," as one of my friends likes to say. You should use just as many (or just as few) hyperlinks, as it is necessary to express meaningful links. And, you know, you are not eternally committed to links. they are just as easily removed as they are added.

I AM sorry.
Manfred

Guido wrote:

In his theory, Prof. Haft
explains that lawyers should build structures of basically not more than 7 items ( one
parent and not more than six childs ), arguing that doing otherwise would not fit to the
human short time memory's capacity ...

...

Therefor an outline should not have more than 6 main chapters,
each of them having not more than 6 sub-chapters...

In my outlines, I try to limit myself to six children for each item. I don't do that for my memory, but rather because I have found that it forces me to think harder about the structure of my outline and speeds up information retrieval in the future.

It's easy for me to just dump a huge list of information into one level of an outline. Unfortunately, a huge list is pretty much useless: to find information in it, you have to read the entire list every time you need one of the items!

I guess it's basically a search algorithm problem. I want faster searches, so I want smaller numbers of children on each branch. Setting an arbitrary limit of 6 children per item forces me to really consider where each item should go and how it relates to other items.

Stephen Zeoli wrote:

First of all, I am uncertain of the value in worrying about
short-term memory. That's not learning or understanding. It enters our minds for a
short time then goes away. As it should.

I may be wrong, but isn't something supposed to go through short-term memory before it goes into long-term, i.e. short-term memory effectively operating as a filter for what may be further remembered? The way I see it, if it's not worth remembering for a short while, then it's probably not worth remembering for a long one.

alx

Alexander, you are right. Research shows that it is the pattern we use to process in our short - term memory that is registered in our long - time memory. Stephen confounds short - time memory with ultra - short - time memory which indeed is used for very very short but necessary reactions so as crossing the road when the lights are green etc. and for that reason has to be quickly evacuated.

Not so for short - time memory. When reading a book, you use your long - time memory for the context and your short - time memory for the new information. You simply cannot process information i.g. learn, understand, think without short time memory.

Being limited to seven items would be quite frustrating. We have to build chunks of information. An experienced "speed-" reader is used to deal with a whole phrase as one chunk and remembers up to three, four, five phrases in his short - time memory. An unexperienced "slow" reader deals with one word or two = one chunk and has to re-read frequently.

When using outliner software or, better, mind - mapping software we should be aware of the brain's necessity of chunking information. Creating huge lists simply will not help to learn, to memorise and finally to understand better. I propose to create a network of Mind-Maps that have not more than 4 or 5 main branches and 5 or 6 levels ( childs to each branch ). By repeating the Mind-Map, it will become one chunk with the time and you will be able to memorise 4, 5, 6 Mind-Maps and so on.

Guido, many thanks for the summary.

The issue seems to be quite more complex than I initially imagined. Indeed, Miller's original paper ( http://www.musanim.com/miller1956/ ) seems to have been surpassed, as well as been overused in areas unrelated to its original scope, such as user interface design ( http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0000U6 ).

From what I've read these days, I think your description is one of the most clear and accurate ones regarding the outcome: whether we are to memorise or process information, we need to recognise patterns and deal with chunks rather than mutiple isolated concepts. I think that, one way or another, this method is what many of us have been trying to apply with the tools we discuss at this here forum.

alx