Designing for Working Memory
Your user’s working memory has a storage capacity of ~4. How much is your design taking up?
As designers, we know our designs should reduce cognitive loads. We’ve heard that good design is invisible design. But why? What is the physiological or psychological reasoning for this? Hard to answer, right? It’ll be much easier to answer once we understand what our working memory is.
What is working memory?
The working memory is where we learn, where we do math, where we reason. It’s what we use as we’re reading this sentence and when we’re using any app. Essentially, working memory is what processes information, the information that’s stored temporarily in our short term memory and more permanently in our long-term memory.
Here’s an example. Close your eyes and do the following equation. Try it, then open your eyes.
Short term memory allows you to remember that you need to add 24 and 55. You know how to add because that’s been stored in your long term memory since elementary school. Working memory is when you’re actually doing the operation.
But working memory is severely limited. Unlike long term memory, which is unlimited, short term and working memory can only hold and process 4 +/- 1 discrete pieces of info (if you’ve previously heard 7, read the footnotes to learn more). If I changed the operation to 247 + 551, notice how much harder it’s become. If I asked you to close your eyes again, just keeping the numbers in your head is quite difficult. Processing it becomes even harder.
When working memory is so limited, it’s imperative that design is intuitive. Otherwise, you’re relegating the user’s working memory from focusing on their goal to focusing on how to use your tool.
And remember, working memory only has ~4 spots!
Is your design an abacus or a calculator?
Using the same math example again, let’s say you must remember 247 + 551 because you can’t write it down or refer back to this post. Now pretend I gave you an abacus to do the math problem. Chances are you’ve never used it before, so now remembering your goal of adding 247 + 551 has become much harder because your memory has now been relegated to learn how to use a new tool.
That mental anguish, the mental taxation, that’s because all your working memory is at its ~4 capacity. That is cognitive load. A psych term, it literally means the amount of working memory used for a task.
Compare using an abacus to using a calculator. A calculator, a tool that you’re familiar with, disappears into the background, so that all your working memory is focused on punching in 247 + 551 instead of figuring out how the tool works. Much less cognitive load.
So why should design fall into the background? Literally, it’s because working memory, when focused on its own goal, has less capacity to figure out how to use a tool.
With only ~4 capacity storage, it’s a zero sum game. If working memory is focused on the tool, it’s not focused on its goal. If it’s focused on the goal, it has little room for a complex tool. If it wants to focus on both, then we’ll feel a lot more mental fatigue and be more error prone.
Using our eyes to “understand”
Recently I’ve been fascinated by our eyes. In another article, I wrote about how 30% of our brain is composed of neurons dedicated to vision and how 90% of transmitted info to the brain is visual. As a matter of fact, our eyes are optical neurons that outgrew from the brain!
Our eyes quickly and easily make tremendous sense of the world around us. When things are laid out nicely, our eyes can “understand” the page or screen. When a page has good information architecture and visual hierarchy, when buttons and links have good affordances, and when objects follow the Gestalt principles, our eyes can quickly and correctly make sense of a screen.
This clears the brain from being involved in the “understanding” so more of the working memory can be used to reach its original goals.
But when a page has a lot of text, when objects don’t follow the Gestalt principles, when a screen isn’t laid out to be easily understood, that computation has now been passed on from our eyes to our brains. Our brains are more actively required to make sense of the screen, and that increases the cognitive load by using up more of our working memory.
So the next time we’re designing, remember that working memory is very limited and precious. Our users should be using all of it to reach their goals, not to understand how to use our tools. Leave the “understanding” to our eyes.
Or just remember our little experiment: are you designing an abacus or a calculator?
Next week I’ll be writing a follow up about how emotions affect our working memory and what that means for our designs. Follow me to be updated!
Also, my last article explored our 3 brains and how to design for them. This article was actually a continuation of the last, exploring how to design for our new brain.
Footnotes / Further Explanations:
If you thought our working memory can hold ~7 pieces of info, that’s from George Miller, who actually said that in a joke. Since then Nelson Cowan has proven that the capacity of our working memory is actually much smaller, holding only ~4 pieces of info.
4 +/- 1 also depends on the person, their knowledge, schemas and expertise. For example, how many of “FBINATOCIANSA” letters can you remember in order? What if I broke it down to FBI NATO CIA NSA? Chances are you can remember all letters in order no problem now. But for a child or a foreigner, breaking down the letters won’t help, because they don’t have this knowledge stored in their long-term memory.
Design is the same. Depending on who your users are, they probably already have quite a lot stored in their long-term memory that they can pull from. Designing for millennials who know technology probably better than the back of their hands is very different from designing for users in their 60s, who have difficulties finding the period on the keyboard.
I should mention that the ~4 storage capacity aren’t always even being delegated to the user’s goal while they’re using your product. Anything else the user might be doing/thinking or anything from the environment is also vying for those ~4 precious working memory spots. Let’s say you’re using Google Maps. Chances are you’re out in traffic, trying to navigate between cars and pedestrians while paying attention to cyclists and traffic lights. Those are all using up those ~4 working memory spots! And if you’re new to driving. Well, then you’re just f*ed. Working memory overload!
