How flat design uses psychology to trick us into believing it’s better

It looks simple, and therefore it must be easy to use…right?

But of course this isn’t always true. The disconnect between perceived complexity and operational complexity is one of the issues highlighted in a book I’m reading, Living with Complexity, by Donald A. Norman. Think of a remote control for a TV – we need 20-30 buttons to make it usable. We could make it incredibly simple and just put 1 button, but a remote that only does 1 thing is useless. Apple has tried to re-invent the remote by giving it 7 buttons (as on Apple TV) but as an Apple TV owner, I hate this remote. Entering a username or password, which I’m asked to do nearly daily, is a painful, painful experience. The only problem solved by the remote is that it’s lower in perceived complexity. And it’s pretty. The operational complexity is higher than just about any remote I’ve ever used. It’s a design failure, in my opinion.

So it is with ‘flat design’, pioneered in the modern operating system by Microsoft, who introduced it in Windows 8 in 2012. Not that flat design is new by any stretch, but it’s adoption by Microsoft in 2012, and by Apple in 2013 signaled a major shift away from representing objects as they might exist in the world, with shadows, gradients etc. Flat design tricks us into believing that a simple aesthetic is equivalent to simple operational use. Flat designs therefore must be easier to use than non-flat designs. It’s a trick, and people are waking up to the limitations of flat design. In the last few months, Google has released ‘Material’, a compromised version of flat design with some lighting and shadows.

Google probably recognized, as many have, that ‘flat’ design introduces complexity of it’s own. Flat buttons now look like graphic blocks, instead of what we’re used to clicking on a as ‘button’. How is a user supposed to differentiate a ‘flat’ button from a graphic block of text? In many cases the difference is simply in context, or by a slight rounding of the button edges. In other words, not nearly enough.

‘Flat’ buttons remove all chrome on the button – but the ‘chrome’ is part of what makes us understand that an element is clickable, and will lead us somewhere.

A gradient on a button mimics round buttons in real life – and lets us know that on the computer, this element is not just a graphic block with a word in it, but a clickable element that controls the interface at hand.

Flat design in it’s current form is dogmatic – it removes nearly all visual complexity – even the complexity that helps us make sense of the interface. Think about looking at a a piece of wood, a beach, pavement and a couch. All have complex surfaces, but those complex surfaces give us information about how to decode the object in question. Is it rough or smooth? Will it give way or retain it’s shape? We’ve learned to use complexity to our advantage, to understand the world in which we operate.

I appreciate the movement towards simplicity that flat design represents. But I’m worried that instead of taking the time to design mental models that are familiar and easy to learn, we’re spending our time in the aesthetic realm. We’re conflating the look and feel with the more difficult to understand concept of how it actually works. And how it functions as a system. How each of the parts work together.

So even though the flat design debate seems to be resolving somewhat, we can’t stop with a self-congratulatory pat on the back. Flat designs can still suffer from confusing mental models, muddled interactions and poor layout. Any design can, that’s the point. We need to look beyond the surface, understand the psychology hoodwink, and keep designing, down to the very bones. Our applications will be better for it.

Featured image credit Ryan McGuire at Gratisography