Cognitive Bandwidth Is Like Dial-Up

🧑‍🎨 Creator(s)
🗓️ Publish Date
January 29, 2005
📚 Publisher(s)
🍿 Media Type(s)

🗃️ Archival copy:


A couple days ago I got an email from Steve Krug, author of the wonderful web usability book Don't Make Me Think, which is in my Top Ten Computer Books list on


I thought about how our books could have been named just the opposite of his--DO Make Me Think, since much of our approach is about how to get learners to process new information more deeply. In other words, we work hard to make people think.

But then I realized that both his book and our approach could have been named:

Don't make me think about the wrong things.

I can't speak for Steve, but my interpretation of his message is something like:

When someone comes to your vintage vinyl store, they want to think ONLY about the records.

They do not want to think about whether that picture over there is the thing they're supposed to click. They do not want to think about where they are on your site, how they got there, and how the hell they get back to where they wanted to be. Worst of all (for the store, anyway), they do not want to think about whether your website actually is an online vinyl store.

If I'm digging for just the right record for my perfect remix, that's what my brain wants to focus on. I want your site to stay in character, and not take me out of the digging-for-vinyl experience by forcing me to think about your user interface. I want to be in flow, just as I would in, say, a real bricks and mortar record store, where the experience is intuitive.

Cognitive bandwidth is precious.

We try to reflect this in our learning books in two main ways:

1) Use a strict 80/20 approach with the material.Rather than taking a topic, making a chapter out of it, and doing it to death, we try to focus on just the part that gives you the power you need to be creative, and leave off everything else. Because we assume you're not reading our book as an intellectual exercise or to skim every possible factoid about the topic. We assume you actually want to do something.

2) Don't use an example that comes with cognitive overhead.We had a Java course at Sun where one of the early exercises was on the looping constructs of the language. But the exercise itself was a task that, among other things, involved converting newtons to kilograms. The scenario was some kind of package shipping system, or something like that.Of course what happened is that when the students got to that exercise, they focused their brain on the whole newton-to-kilogram thing, and struggled with understanding the shipping domain. In other words, they were thinking about the wrong things. All we wanted them to do at that point in the course was understand the basics of looping. But the exercise added so much cognitive overhead that looping was the last thing they were thinking about. [Disclaimer: we don't always succeed at this... I've authored more than one chapter where I forgot the point. But we're trying. Hard.]

When someone has trouble applying knowledge, it's usually because they really never had knowledge. They had information, and that's not the same thing. You can get information just through listening or reading, but knowledge requires thinking... thinking about the RIGHT things.

Our advice to our authors, teachers, and web/software developers is this:

Figure out what you really want users to think about. This is almost always the cool thing they want to do (pick the right record, learn how loops work, etc.). Do whatever it takes to keep them from having to think about anything else!

Imagine your users all have thought bubbles over their heads that say, "Don't make me think about the wrong thing!" If a user has a confused look, it should be because she's struggling with whether the sea foam green bustier really works with the neon pink skirt (it doesn't), or whether the iPod Shuffle is better than therapy (it is).