Casual Fridays: Can sound cause us to experience a visual illusion?

Last week's Casual Friday study was inspired by a recently-discovered illusion which showed that sound could influence what people percieve visually. I was planning to report on the study confirming that illusion yesterday, but my computer wasn't cooperating with me, and I couldn't generate a demo of the illusion. I think I've figured it out now, but now it's Casual Friday, so you'll have to wait until Monday to see it.

In the meantime, we've created our own study of visual illusion and sound. What we wanted to know is if sound could push a viewer over the threshold from not perceiving an illusion to perceiving it. We selected two relatively ambiguous illusions from Akiyoshi Kitaoka's amazing illusion pages. Many of Akiyoshi's illusions are quite dramatic, and sound would be likely to have no effect on the illusion. But we found a couple that were a somewhat difficult to perceive. The first is the "Rocket," in which the two rockets appear to be moving up:


The second one is ambiguous in a different way -- the inset appears to move, but in no particular direction:


For the study, we made these illusions even more ambiguous by reducing them in size. We wanted to know if playing a musical scale moving in a particular "direction" -- up or down -- would make more people perceive the illusion. I created QuickTime movies to present each illusion with a musical soundtrack. Here's the Rocket with a scale going up:

Respondents were divided into two groups. One group saw the rocket movie with a scale going up and the diamond movie with a scale going down, followed by a silent version of each illusion. The other group saw the rocket movie with a scale going down and the diamond movie with the scale going up, again followed by silent versions. After each movie, they were asked which direction the relevant part of the picture was moving.

The results? Most people still didn't see any motion at all. For all the movies, "not moving" was the most popular choice, always garning more than 70 percent of the votes. However, the sound did appear to have an effect on those who indicated a direction of motion:


When a direction was chosen, "up" was always most the most frequent choice for those viewing the rockets, while "right" was the most common for the diamonds. But significantly more people indicated the rockets were moving up when the scale went up compared to when the movie played silently. A similar effect was found for the diamonds: when the scale was ascending, people were more likely to say the center of the figure was moving to the right compared to when the figure was viewed in silence.

It's more difficult to say whether the "direction" of the music influenced the direction of the illusion. There was no significant difference in the perceived direction of either illusion when the scale went up compared to when the scale went down. Perhaps the sound simply made viewers more aroused, and therefore more likely to see the illusion. Or perhaps after seeing the same illusion a second, viewers were less convinced that they saw motion.

Nonetheless, this is a fascinating result, and I think it's possible that there might be an actual non-casual study lurking in this data.

Be sure to come back next week for a new Casual Fridays study (and if you're interested in illusions, come back on Monday for my report on a peer-reviewed study of how sound can cause a visual illusion).

More like this

Can sound cause us to experience a visual illusion?
If you have to ask, you've never been to a Dead show....

One difference between the two images that should be taken into account: in the rocket image, there is a visual asymmetry, namely the fact that the rockets come to a point at one end, which suggests direction. On the other hand, in the square image there is complete symmetry both vertically and horizontally - no visual suggestion of direction to the right or left. This might account for why there was an asymmetry in the results with more people claiming to see the rockets go up than down while there was no such asymmetry with regard to the direction of the square.