Touch interfaces and user control

Before you click a button or link you subconsciously weigh up the consequences of doing so. What will happen? What are the risks that it will trigger some negative or unexpected experience?

On a desktop application or website there’s usually plenty of information to help you determine the risk of clicking on something. Tooltips commonly appear when you hover over buttons and icons explaining what it’s for. Other links are textual, offering an unambiguous description of what you’re about to do.

However, this comfortable status quo does not translate well from the desktop to smaller, touch screen devices. On smartphones, where screen real estate is limited and tooltips do not exist, the consequences of each action can be hard to anticipate.

For example, in Amazon’s Kindle iPhone app, there is a ‘+’ button at the bottom of the screen (pic above left). What does this do? There’s no way to find out without pressing it.

If you decide to find out, a small black triangle appears at the top of the page (pic right). This looks like a paper page is folded over, but what this means is unclear. Is it a bookmark, a note, a highlight or link to another interaction?

This type of interaction, encountered for the first time, can be worrying. Users won’t feel in control if they don’t understand:

• What actions will do

• What actions have done

• How the consequences of an action can be undone

This sense of control can be at least partially achieved though. In the Guardian’s app (pic left), when you click on an icon for the first time a dialog appears explaining what has just happened. While this is only a retrospective description, it allows users to quickly learn how the interface works.

The conventions for taking the guesswork out of interacting with touch screens are yet to be established, but it’s clear that this is a problem that every app designer faces. Ensuring that users feel in control of their actions and the consequences of these is essential to retaining a strong user experience on mobile devices.

26/11/10 Dan Littler said:

Agreed! I guess it makes consistent controls more important than ever, so that users can rely more on muscle memory. But that's only one bit of the solution. Interesting stuff. There are similar issues with the 'thin air' gestural model of xBox Kinect, and voice control - lack of cues, feedback, perceived risk of wrongly interpreted actions, etc. I wrote a less-thorough post than yours about it (and weirdly enough, on the same day): http://www.danlittler.com/2010/10/27/would-you-rather-make-a-gesture-or-push-a-button/