Noodling on a cloud

SUMMARY:

By teaching others, we also learn ourselves. How can we best use conversation as a tool to facilitate that?

Sensemaking

How do people make sense out of raw input?

Marvin Cohen suggests that it is usually a two-way process. Not only do we use the data to suggest a mental models to try for good fit, but also we simultaneously try to use mental models to select and connect the data. (LINK)

The same thing applies when the data is a cloud of vaguely associated concepts in our head. One of the ways that we can make sense of them, turn them into crystallized thoughts that we can then associate with a handle, is by attempting to verbalize them. The discipline of turning something asyndetic into a linear progression of connected thoughts forces us to select between possible mental models and actually pick just one, allowing us to then consider whether it fits the data well or not.

But the first possibility we pick won't necessarily be the one that fits best. Going around a loop, iterating, trying different starting points or angles of approach, trying different ways of stating things, and seeing what associations those raise to add to the cloud, takes longer but can often produce more useful results. However, its a delicate process, because of the way memory works.

Working memory

The size of cloud you can crystallize is limited. The type of short term memory that the brain uses to store them where you're aware of them lasts about 18 seconds. (LINK) For a concept or datum to persist longer than that, part of your attention needs to be used to 'revisit' it. The faster your ability to do that, the more mental juggling balls you can keep in the air without dropping one. Most adults can keep between 5 and 7 balls in the air, in their 'working memory'. (LINK)

There are a number of ways around this limitation. You can group multiple concepts together and treat them as a single 'ball', if you can attach to them a mental handle (a reference, such as a word or image, that recalls them). (LINK)

You can put things down on paper, rather than doing it all in your head, using the paper to store links to different parts of the cloud. So, for instance, rather than try to consider 12 things at once, split them into 4 groups of 3 (A, B, C & D), and systematically consider the concepts 6 at a time: A+B, A+C, A+D, B+C, B+D, C+D (and hope that the vital combination you needed wasn't larger than 6, or spread over more than 2 of your groups).

And you can use other parts of your short term memory as a temporary cache, to expand your stack. For example, the phonological loop, which gets used when we talk out aloud. (LINK)

Talk

In section 4 of their 2007 paper (LINK), Simon Jones and Charles Fernyhough say some very interesting things about the origins of thought, and also about Vygotsky's theory of how self-talk relates to how children learn to think through self-narration. (LINK)

It explains why talking aloud is actually one of the most effective ways of coming up with new thoughts and deciding what you actually think about something. And that's not limited to when you explicitly talk to yourself. The same process takes place when you are talking to other people; when you're having a conversation.

When this works harmoniously, your conversation partners acts as a sounding board, as additional sources of concepts to add to the cloud you're jointly noodling on, and the sound of the words (via the phonological loop part of your memory) works in effect as an expansion to the size of your working memory.

The downside is potential interruptions.

Interrupting the flow

A lot has been written about the evils of interrupting computer programmers (LINK, LINK):

and, to some extent, the same applies when you interrupt while someone else is talking, or totally derail the conversation onto a different topic when they pause.

People interrupt because they don't know better (children who have not yet learned how to take turns), because they are egotistic (they think that what they want to say is more important or interesting - they want the attention), as a domination power play (yes, that get's taught as a deliberate technique: LINK), because they are desperately impatient (they've have a thought and are sure they'll forget it unless they speak it immediately) or even because they believe they are being helpful (completing your sentence, making efficient use of time).

But what the people worried about efficiency of communication are not taking into account is that there's more than one conversation going on. When I talk aloud to you, I'm also talking aloud to myself. When you interrupt my words to you, you also interrupt those same words going to me, which help me think.

As one person put it, in the context of a notice on a door in a work environment:

When I’m busy working, please don’t interrupt me unless

what you have to share is so urgent and important that

it’s worth erasing all the work I’ve done in the past hour.

Points of order

So is interruption ever ok?

Yes. Sometimes people are not in the process of constructing thoughts that are new to them, on the very edge of what they can conceive. Sometimes people ramble, because they are used to a conversational style that encourages interruptions, and welcome someone else 'rescuing' them from having to fill a silence. And sometimes something new comes up which is not only important enough, but also urgent enough, to merit an interruption.

But I'd like to consider a different scenario. Not a contentious one, where the interruption happens against your will, but where two or more well intentioned people are having a conversation designed to evoke new ideas and where certain type of interruption are part of a pre-agreed protocol, designed to aid the process.

For example, suppose people in a particular conversational group agreed certain hand signals, that could be used to cue each other to:

• I'm currently trying to solidify a thought. Please give me a moment to finish, then I'll restate it from the beginning in better order or answer questions.

or:

• Stack Overflow. I want to follow your explanation, but I now have so many pending questions that I can't take in anything new that you're saying. Please could you find a pause point to let me off load some of those pending points, before you continue?

Does anyone here know of groups that have systematically investigated how best to use conversation as a tool to improve not the joint decision making or creativity, but the ability of individuals to conceptualise more complex ideas?