This post is part of a series about Designing the Obvious, a book about common sense Web application design. Learn more about this series.
If the term “mental model” is off-putting to you, I understand. It sounds starchy and academic. It also is a great name for what it is. Just like a model train can help you understand what the real trains look like, mental models provide a way to understand how users think. Mental models help to avoid thinking like a technician.
The book describes it this way:
“A mental model is what we believe to be true… It has to help us understand how to use a computer and understand what it is, but not necessarily what it really does.”
There is an example from the book that is so good, you may be buying the mental model yourself.
The trash can or recycle bin is where we put files we don’t want anymore. When we’re absolutely sure we won’t need the files, we empty the trash. In neither case is the computer actually destroying the file.
There’s an even better mental model hidden in the above example. Can you find it? Files! This model is so strong that most programmers don’t even have to acknowledge that files are just a way to disguise the details of storing data.
When a programmer doesn’t share the user’s mental model, it’s called an implementation model:
“An implementation model is something that has been designed, usually by not being ‘designed’ at all, to reflect the underlying details of the system.”
Designing the Obvious says to look for implementation models in our applications and find ways to replace them with mental models–the stuff that makes sense to users.
As a technician, I’m often thinking in the wrong direction, as Peter Merholz described in The Experience is the Product:
If you can pretend to be a user, you’ll have a leg up on creating something really easy to use. If you are a user, even better. Otherwise, you need to grab a user or two and talk to them.
Find the places where you’re “removing sectors from the hard drive index” and change it to “dragging icons into trash cans.”
Got any horror stories of implementation models beating out mental models? Dare you share any of your own mistakes? I’d love to hear about it below. As a special treat for sharing, you could get an autographed copy of Designing the Obvious. I’ll randomly select three commenters when I finish the series.