System 1 and visual meaning

We love Daniel Kahnemann.  And we’re not aloneThinking, Fast and Slow has given us a whole range of useful tools for managing our work and making it more meaningful.

For those of you not familiar with it, Daniel’s work centres around the characterisation of the mind as composed of two “systems” – system one which understands things instinctively (“thinking fast”) and system two which processes information in deliberate conscious steps (“thinking slow”).  So for example, system one knows how this person is feeling without having to think about it:


But it’s clueless about how to solve …


… which requires System 2.

It’s just a metaphor, but it’s a neat one.  The practical implication of the book is that our lives work better when we understand the two systems, and know which one is appropriate for each situation.

It’s an idea we’ve started using more and more in our work to describe how useful (or not useful!) our pictures are.  If everyone we show it to gets what our picture is trying to say, then we describe it as “System 1 friendly”.  If they spend a lot of time trying to figure it out, then something’s gone wrong.  The same of course is true of all visual tools – maps, human machine interfaces, dashboards, charts, you name it.

So how do you make visual tools more System 1 friendly?  The answer, as ever, lies in connecting with the experience of your audience.  Think, for example, about the visual language of maps.  What makes road maps intuitive?

Well, the experience that maps connect to is that of the physical landscape and its geographical features.  So the more closely the visual language of the map replicates the visual experience of the geography, the faster System 1 can understand what things are.

Sometimes this is obvious.  I haven’t encountered a full-colour map that has not shaded woodland in green, because that’s the colour we prototypically associate with trees:


But if you imagine for a second someone was creating a new series of maps and decided that woodland would be purple, you can see that suddenly System 2 will be called into action, because the symbols no longer connect intuitively with experience:


Here’s a subtler example.  In the UK, roadmaps invariably have motorways shaded in blue and A roads in green.  Here the visual language isn’t taking its cue from the physical characteristics of roads (which are all black with white stripes), but the small number of things that differentiate one road from another.  In this case, motorways are larger than other roads, and have blue signs compared to the green signs on A roads, so this is reflected in the visual language:


Also notice how dual carriageways have a line down the middle, giving System 1 a convenient connection with the experience of seeing a central reservation.

Now imagine this visual language is taken away – all roads are the same size and the same colour.  Reading the map is now more like asking someone a long multiplication problem i.e. System 2 has to step up.


And this is the point:  System 2 hates stepping up!  It takes way more mental energy than System 1.  This is why we’re drawn to interfaces that are clean and intuitive, and repelled by complex, hard-to-read charts.

So, try this yourself.  Whether you’re making a picture of a complex system, a visual for a Powerpoint slide, a User Interface, or whatever.  Feed System 1 with cues that it can use to connect to its experience of the world, and you’ll end up with much happier users!