Change blindness is defined as the inability to detect changes to one’s environment.
Daniel Simons and Daniel Levin conducted a well-known research study on real-world interaction and change blindness.
In their study, “Failure to detect changed to people during a real-world interaction,” participants were asked to give directions to strangers. During the conversation, they are interrupted by two people carrying a door. What many participants failed to notice was one of the workers and the person asking directions actually switch places. About half of the participants failed to notice the change.
This idea of change blindness is very applicable to user interface design. Participants in the door study were focused more on giving directions, and less on changes in their visual world. Online users are similar; they can easily miss a “Back” button or fail to recognize a form has been submitted. This isn’t because they lack intelligence or focus, but simply we as humans are limited by our memory. When too many things enter our working memory, it is very difficult to give adequate attention to all of them.
Just as the the moving door interrupted the visual perception study participants, there are similar factors in user interface design that can interrupt a user’s perception of a digital product on screen. Some factors could be a user blinking, looking away from the screen, eye saccading, page loading, page animation, etc. These factors can effect usability perception of error messages, status notifications, menu navigation, searching, etc.
If we apply change blindness to search filtering as an example, we can understand how to visually present it. Users want an interaction the responds quickly in order to achieve a certain search set. However, depending on the visual implementation, some users may not detect a change in search results, causing them to either replicate the task or continue on without realizing the change. Many search filters are instantaneous: check box, drop down, input field, etc. values are applied as soon as they are activated. There should always be an obvious interim indication that filters are in the process of being applied. This could be a message overlay stating “We are applying your search…” or a loading animation.
We can reduce interaction cost in the context of change blindness by using visually clear UI elements to alert the user about changes.