Friday, May 24, 2013


Failing to Notice Haircuts, Missing Buildings, and Changed Conversation Partners

So what do you think? I hate that question. People ask me that question when they've changed something - a new haircut, new clothes, new paint, rearranged furniture, whatever. Unfortunately, I fail to notice the changes so I don't know what to compliment. Why am I so hopeless at detecting changes?
As it turns out, I'm not alone in failing to notice changes. People fail to detect a great variety of changes - we regularly experience change blindness. Hollywood, for example, depends on all of us reliably suffering from change blindness. Directors make multiple takes of most scenes and then splice together pieces from different takes. The different scene takes include changes - changes in background objects, the location of objects, the clothing on actors, etc. Movies frequently contain these changes called continuity errors. But when I watch movies, I don't notice these continuity errors. If you're interested, Google "Continuity Changes in Movies" and you'll get several websites that provide examples from both hits and flops - a fun search if you are a movie fan.

Change blindness isn't limited to movies, however. Dan Simons and Dan Levin created a live version of a change blindness experiment. People were walking in a familiar environment when someone (a researcher actually) asked them for directions. As they were giving directions, people carrying a large door passed between the conversation partners. After the door passed, the people finished giving directions. Unbeknownst to them, their conversation partner changed when the door passed - the person they were talking to helped carry the door away while someone different stopped to finish taking directions. Their conversation partner changed in the middle of the conversation!
Would you notice if your conversation partner changed? Everyone thinks they would notice. Simons and Levin found, however, that about half the people failed to notice the change. (You can see a video from the original door study or a funny British TV recreation of the study in YouTube videos.)
Cognitive researchers have studied change detection and change blindness using a variety of methods. The good news is that if something changes right in front of you, you'll generally notice. The perceptual transition grabs your attention and you detect the change. If you are staring at the stoplight while waiting, you'll usually notice when the light turns green.
You may experience change blindness, however, if there is any visual disruption coinciding with the change. With visual disruptions, no single perceptual transition grabs your attention because all the visual input changes. This occurs whenever you move your eyes, during cuts in a movie, when a door passes by, or if you alternate between two slightly different pictures with a blank screen in between (called the flicker paradigm, created by Rensink, and fun to try at his website). Other researchers have found change blindness in comparing memories to pictures. If you show people pictures of familiar scenes, for example, they fail to notice missing buildings. In a variety of research paradigms, we fail to notice meaningful changes.
I've loved the change blindness studies since I first heard about them. I don't feel so bad about failing to notice haircuts anymore.
We experience change blindness because we maintain a limited representation of the world in our heads. I have, for example, a rather vague picture of what my friends look like. Surprisingly, my interaction with my friends and the world is better because of this. I need to recognize my friends in different clothes, lighting, locations, and after they get a haircut. A more general representation allows me to recognize people across these changes. This vague and limited representation also means that we may fail to notice interesting stimuli in our environment (such as unicycling clowns, see my earlier post).
If I had a more detailed representation, then changes might cause me to fail to recognize someone. For example, one of my students recently got a major haircut and dyed her hair. I still recognized her, thought she had changed something, but couldn't identify what had changed. I experienced change blindness, but successfully recognized my student. In contrast, when she walked into her mom's office with her new hairstyle, her mother failed to recognize her. Her mom's representation was detailed enough that the young woman no longer matched.
I'm happy suffering from change blindness and experiencing a consistent world. I continue to recognize people and places across changes. Of course, if you get a haircut, let me go ahead and apologize now. I probably won't notice. Luckily, it isn't just because I'm an absent-minded professor. Instead change blindness is a basic of human perception and attention.

No comments:

Post a Comment