Pattern Recognition

Written on 23 January 2012

Unless you’ve been on Mars for the last 20 years you probably don’t need to be told what, or who, these objects are. They’re immediately familiar even though they’re constructed from nothing more than a few coloured bricks.

This image gives our brain the chance to show off one of its most impressive skills – pattern recognition. Pattern recognition allows us to understand complicated things even when we’re only given limited information about them. So even though the object on the right is made up of three Lego bricks, representing only nine bits of information, pattern recognition makes our brain ‘see’ something far more intricate:

Pattern recognition also comes into play when we’re using websites or other interactive systems, and is therefore something that interaction designers work with every day.

Computer systems are complicated things, containing so much logic and data that to fully understand them might take several weeks or more. But when someone uses a computer system they don’t have that much time, and they don’t need to understand the system in its entirety – they only care about the bits that are relevant to them.

As interaction designers we’re often challenged with making sure users understand the systems we design without having to absorb a lot of information about them. Three major constraints apply:

  • Users have limited time – they often want to complete a task in a matter of seconds
  • Users have limited screen space at their disposal, so the element we’re designing can’t be too large
  • We can’t expect users to have trained themselves in using our system: RTFM is never an option.

Pattern recognition helps overcome these constraints by helping users fill in the gaps between what they perceive (the pixels on the screen) and what they’re actually dealing with (a horribly complicated computer system). For example, the user might see a grid of numbers between 1 and 31 and a couple of arrow icons…

…but if the designer’s done a good job, the user will almost immediately ‘see’ something much more complicated and intricate:

In this way, pattern recognition helps the user extrapolate from a simple graphic to a complicated range of functions without having to read, think, or experiment. Very little information has been exchanged between the system and the user, but she now has a detailed mental model of this part of the system and will be able to use it comfortably.

Pattern recognition isn’t always beneficial, of course. If a design element plants expectations in a user’s mind which it doesn’t go on to fulfil – if, say, clicking “December” in the calendar above did something other than show a dropdown – this can lead to a jarring experience which is known as cognitive dissonance. It’s important for designers to remember that the people who use their systems are always looking for patterns subconsciously, and if the design isn’t well thought out these users’ brains can easily be sent down a wrong and frustrating path.

The example of the Lego Maggie Simpson shows how easy it is to get a pattern wrong. Imagine there was one more blue brick, or if the yellow was replaced with brown – the illusion would be completely shattered.

As designers, we have to think about getting the minor details right if we want to make pattern recognition work in our favour. And if we fail, our users won’t ‘see’ the forms, structures and connections behind the systems we create – they’ll just see a small pile of randomly selected Lego bricks.


More from the blog