Science-- there's something for everyone

Friday, October 5, 2012

Change deafness

When something suddenly appears in our visual field, we usually notice right away. That isn’t the case when something changes or disappears. Even though the difference is equivalent, it takes us a lot longer to register the loss or alteration of an object within our field of view. This phenomenon has been termed ‘change blindness’. You can see an example here.

Apparently, the same is true for our sense of hearing. University College London researchers led by Francisco Cervantes Constantino tested people for ‘change deafness’. They played specific sets of 4 to 14 pure tones for groups of people with normal hearing. Each auditory scene lasted from two to four seconds. In some trials, a tone was either added or subtracted partway through the sound-scape. You can see a visual representation of these trials below.


No change: the scene was played unaltered.
Change-disappear: A single tone was removed.
Change-appear: A single tone was added.

Participants did significantly better at detecting when a tone was added than when one was subtracted. As more tones were added to the original mix, people became worse and worse at noticing the disappearance of one sound. In contrast, the subjects were close to 100% accurate in detecting the addition of a new tone even when listening to fourteen other notes. People were also far more adept at identifying exactly which tone had been added than which had been removed.

It’s not surprising that we’re attuned to pick up the commencement of new sounds in our environment. After all, a sudden sound may augur the arrival of a predator or other danger. However, it is interesting that we’re much less likely to notice the cessation of sounds. Anyone with a young child can tell you why this can be problematic. 

Francisco Cervantes Constantino, Leyla Pinggera, Supathum Paranamana, Makio Kashino, & Maria Chait (2012). Detection of Appearing and Disappearing Objects in Complex Acoustic Scenes PLoS ONE : doi:10.1371/journal.pone.0046167