How Kinesthetic Memory Impacts the Need for Consistency
Kinesthetic memory—remembering with our body—is a powerful factor in learning to use a software user interface. We learn by doing—through repetitive actions. We remember the sensations of our body’s positions and muscular movements in relation to ourselves and our spatial environment.
According to “Kinesthetic Cues Aid Spatial Memory,” by Desney Tan, Jeanine Stefanucci, Dennis Proffitt, and Randy Pausch:
“Kinesthetic cues, or the awareness of parts of our body’s position with respect to itself or to the environment, are useful for recalling the positions of objects in space.”
While information about the prevalence of different learning styles differs, the University of Illinois reports that 50% are kinesthetic learners who learn best through movement and manipulation, 40% are visual learners who learn by watching, and only 10% are auditory learners who learn by listening. Research by Asselin and Mooney found that kinesthetic learners retain best; and people remember 90% of what they say and do, but just 30% of what they see and hear. Of course, the majority of people are multimodal learners.
Because of the power of kinesthetic learning, once users learn the position of an affordance on a screen, they reach for it without even looking. Interactions become habits, and habits are powerful and hard to break. Thus, inconsistencies in interaction models wreak havoc on our ability to learn new user interfaces that differ from those we already know. Which interaction models will stick in our mind, causing us to make the same mistakes over and over again when using software that deviates from the standards in our mind?
Good and Bad Disruption
As UX professionals, we all want to create great user experiences. But opportunities to design something new from scratch that will disrupt a market are rare, so most of our design work involves iteratively improving the designs of existing user interfaces.
Innovating Interaction Models
If you have the opportunity to design a new operating system, a brand-new application, software for a new business domain, or even a feature that no company has ever implemented before, that’s the time to think outside the box and innovate. Don’t rush the process. Create something great!
If you can innovate an interaction model that is leagues better than what already exists or is current best practice and will differentiate a product in the marketplace, go for it!
Breaking Interaction Models
But what you really shouldn’t do is make little, incremental changes to interaction models that disrupt users’ established behaviors. If you’re thinking about tweaking anything that would impact a user interface’s interaction models, don’t!
Making users change their long-established behaviors drives them crazy. Nobody likes making mistakes, and tweaking interaction models is guaranteed to result in user errors or, at the very least, irritate users.
Unfortunately, we see these sorts of design changes all the time. For example, designers move controls from the top to the bottom of a screen or vice versa for no good reason. They change the way affordances behave. They change interactions in small ways that don’t add value, but disrupt user behaviors. Companies release applications that have similar functionality, but implement them using inconsistent interaction models.
Recently, I’ve experienced what were to me some irritating design changes to interaction models that have disrupted my behaviors when interacting with some of my favorite products. That’s what prompted me to write this article. These are the changes that bothered me:
- Apple added widgets to iOS 10 and disrupted the behavior of the iPhone lock and home screens and the Camera.
- Apple overloaded the Control Center in iOS 10 and changed its behavior.
- Apple hid the message toolbar in Mail for Mac OS X.
- Amazon changed the behavior of toolbars in Kindle for iPad.
I’ll tell you stories about my experiences from the standpoint of a user to convey how users might feel when we make them change their behaviors.
Widgets and the iOS 10 Lock and Home Screens and Camera on iPhone
This is the change that annoyed me the most. I discovered this change when using my iPhone for the first time after a software update. I had grabbed my locked iPhone to take what would have been an incredible photo of a huge flock of birds flying swiftly over my top-floor balcony at very close range and wasn’t able to get to the Camera app in time to take the shot. I had only seconds!
Notifications were completely covering the lock screen, as shown in Figure 1, and there was no longer a Camera icon, shown in Figure 2, in the lower-right corner. My habitual behavior was to tap, not swipe to get to the Camera app. There was nothing to tap. Once Notifications were out of the way, I did not notice that the dot at the right in the page control was actually a tiny image of a camera, shown in Figure 3, because of the background image on my phone.
Apparently, Apple changed this long-standing interaction model for taking photos on a locked phone to accommodate its new widgets user interface, shown in Figure 4, which is a copycat feature that I have absolutely no use for.
iOS 10 Control Center on iPad
This is just a minor annoyance, but I used to have easy access to all of the settings I needed on the Control Center. Now, they’re inconveniently spread across two pages, as shown in Figure 5, so I have to swipe back and forth to get to all of them. The controls for playing music or videos are buried on the second page of the Control Center, and those are the ones I use the most. If the Control Center extended the full width of the screen and all of the touch targets were of a reasonable size—many are huge—the designers could easily have fit all of the controls on one page.
Message Toolbar in Mail for Mac OS X
In Mail windows that display messages in my Inbox, there’s no reason at all why there should be two toolbars with the same controls on them, one of which is visible only on hover. But, in addition to that, the designers have now removed the controls for viewing and saving attachments from the visible interface—Figure 6 shows the old user interface with a visible toolbar—and placed them only on a hidden toolbar, as shown in Figures 7 and 8, making them very inconvenient to use. I’m really not a fan of having to move the mouse pointer around to discover whether there are any hidden features and where they might be. Plus, Quick Look is buried at the bottom of a menu listing all of the attachments, so instead of it’s being a click away, I have to click, move the pointer to the bottom of the menu, then click again. It takes longer to acquire hidden click targets.
Toolbars in Kindle for iPad
Over the years, Kindle’s designers have moved the toolbar from beneath a book’s pages to the top of the screen—with the exception of the page slider, which is still at the bottom. And now they’ve hidden all of the controls, so I have to tap a page to display them. That means I have to tap twice instead of once to do anything with them, and I use some of the controls often.
The new behavior of the app when I tap a page to view the toolbar is very disconcerting. The page zooms out, as shown in Figure 9, then once I tap an icon on the toolbar, in most, but not all cases, I have to tap the page again to get it to zoom back in. How am I supposed to adjust the size of the type when I can’t see the content at its actual size? I don’t find the zoomed out view at all useful.
To keep users happy, design interaction models that are consistent with those that have gone before—except on those relatively rare occasions when you can create a truly disruptive innovation that will improve users’ lives. Don’t needlessly—or inadvertently—tweak interaction models in ways that would make users change their behaviors. If changes to interaction models disrupt habitual user behaviors, they must add significant value to the user experience. So, when deciding whether to change an interaction model, carefully balance the value of that design change against the effort it will take for users to learn a new behavior.