Pareidolia: People Hear What They Want to Hear

We’re going to apply this same concept using DJ software and any song your friends will readily recognize.  We’ll mix this song with white noise—noise generated by a totally random signal. When the audio from the song is reduced to zero, will your friends imagine that they can still hear the song playing somewhere behind all the white noise?

Problem

Can you condition people to perceive music in the absence of music?

Materials

• Computer with a headphone jack
• Headphones
• Demo version of Traktor Pro 2 (download here)
• White noise sound file from AudioCheck.net (right click and save-as)
• Stopwatch
• MP3 file of a popular song your friends will be instantly familiar with (make sure the file has no DRM restrictions, otherwise it won’t play in the DJ software you downloaded)
• 15 or more volunteers

Procedure

1. Install the demo version of Traktor Pro 2.
2. While the program is installing, download the white noise sound file and the MP3 of the song you chose.
3. Open Traktor Pro 2.
4. During the Setup Wizard, specify that you will not be using any external devices. When asked what kind of interface you’d like to use, select “2 Track Decks.”
5. Click the gear icon in the top right corner and select “Audio Setup.”

6. Select the drop-down menu next to “Audio Device” and select your computer’s soundcard or whatever built-in audio device your computer has installed (this will usually be the only available selection).
7. Next, click on “Output Routing.” Under “Output Master,” assign the L channel to “Headphones L.” Assign the R channel to “Headphones R.”

8. Click on the drop-down menu next to the left of the magnifying glass icon and select “Mixer.”

9. You should now have a series of knobs and faders in the center of the program.

10. Click the gear icon again and select “Mixer”.
11. Under the section titled “Crossfader,” click and drag the slider to change the Auto Crossfade Time to 60 seconds.

12. Drag your MP3 file into Deck 1. You should see a brightly colored waveform of the song show up.

13. Drag your white noise sample into Deck 2.
14. Hit the “Play” button on each deck to make sure you can hear both your MP3 file and your white noise sample in your headphones. Hit the “Play” button a second time to stop each track. Click the “Jump to Track Start” button to reset each track so that each will start from the beginning when you play both for your volunteers.

15. Have a volunteer put the headphones on. Don’t let them see the computer screen! Tell your volunteer that they will hear both a song and white noise, and that you’d like him or her to indicate when they can no longer hear the music.
16. Hit “Play” on each deck so that both audio sources are playing at the same time. Hit the “Autofade Right” button so that the crossfader slowly begins to cut out the audio from the deck playing the music.

17. Start your stopwatch after the crossfader moves all the way over to the right. Stop your stopwatch when your volunteer states that they can no longer hear the music.
18. Record how long it took your volunteer to recognize that music was no longer playing.
19. Stop and reset your decks. Repeat steps 15-18 with a new volunteer each time.

Results

People hear what they want to hear! Most of your subjects will have thought that music was still playing even when the audio channel playing the music was reduced to 0 decibels.

Why?

By playing music for your listeners, you’re exposing them to auditory information that is in fact structured in a meaningful way. By doing so, you’re conditioning your listeners to expect to extract meaningful information from the sound they hear. Even when the volume of the deck playing the music is reduced to zero, your listener will continue to think they’re hearing the music somewhere in the white noise, even when all they’re hearing is randomly generated sonic information!