Take the 2-minute tour ×
Sound Design Stack Exchange is a question and answer site for sound engineers, producers, editors, and enthusiasts. It's 100% free, no registration required.

My mind has been preoccupied lately with the concept of negative space and figure-ground perception. As an example of what I'm talking about, check out Rubin's face / vase or the Young Lady / Old Hag. You've also probably seen Escher's tessellations before. But what I'm trying to come to a conclusion on is whether or not this phenomenon is possible to accomplish with sound.

Certainly we can construct backgrounds that evolve through a scene or level that become the foreground object. But we make the transition for the listener. We mix and edit with the transformation as our goal. But is it possible to design sound that can be perceived two different ways by people upon one listening? How about by the same person upon repeated listening?

I'm taking the stance that it cannot, simply because we all work along a linear timeline and this is a concept from 2D art. For these images to work you have to essentially overload your visual input to the point that you can switch your mind to begin seeing the other image. But I really, really want to be proven wrong. Can you rationalize this for me? Have you or some sound artist you've heard of explored this terrain?

I may be out in left field here. But this notion has been rolling over and over in my mind.

share|improve this question
    
by the way, great question! we need more like this on the site. –  Shaun Farley Feb 17 '12 at 0:35
    
Ah, thanks @Shaun. –  Steve Urban Feb 18 '12 at 8:47

10 Answers 10

I think you've hit on something with the fact that audio is a temporal idea, as opposed to the 2D non-temporal nature of those pieces of art. In those pieces, we have time for our brain to process and pick them apart from different perspectives, because it remains static. People have a hard time doing that in the acoustic realm, since acoustic events are undergoing change while we attend them. Attack, decay, sustain release and all that.

That tells me that it might be possible if the acoustic event is cyclical in nature; that we could pull off this trick if we can predict what is going to happen in the future of the event. We may need something to help us shift perspective though. Which is convenient, since we work on audio to accompany visual material. Audio adds meaning and depth to the visuals. There's no reason the reverse can't happen. So, I'd say it's possible...but not without the effort and coordination of multiple departments.

Update:

So, why I didn't think of this earlier is beyond me...especially considering the article I just posted on both my site and Designing Sound this week. You can define audio/acoustic negative space as those things that we don't hear (or, myabe on occassions, those things to which we don't attend). "Figure" would equal the sonic content that accompanies the picture, while "ground" equals the sonic content that, while pertinent to the picture, does not. The trick then would be to provide some way to draw attention to what the viewer is not hearing. Maybe it's pairings or groups of sounds that you establish early on, then use only specific elements later in the piece. This might provide a means to infuse two distinct interpretations within a scene; one based on the overt use of sounds, and one based on the overt omissions. Thoughts?

share|improve this answer
    
Good point. Maybe something like this as an experiment, crossing with @Iain & @aross001's answer: Director shoots two conversations in a party scene (dialog A and B), you overlap them in the track and add background crowd walla and bury the recordings a bit into the background. If you loop the scene and swap picture between dialog A and B, does your auditory focus shift and you're able to pick out the respective dialog or does it sound like the same jumbled mess? What if both visuals are displayed as a split screen? –  Steve Urban Feb 18 '12 at 8:45
    
@Steve Urban - it would be an interesting experiment, that's for sure. –  Shaun Farley Feb 21 '12 at 1:12

You can partially achieve this effect with two overlapping voices. Try editing a conversation so that the participants talk over each other. If the timbre is similar, as well as the dynamic range then you will find that you have a slightly different listening experience each time you listen to the track.

share|improve this answer

There's an office scene in Girl With the Dragon Tattoo (which they explain in this video http://soundworkscollection.com/dragontattoo) where there's a floor shining machine of sorts that is at the exact same pitch as Reznor's score.

When I watched the film I thought it was just the sound of the cleaner intensifying and didn't realize it was part of the score until the scene changed. The person I was with heard it as score right away. That might be along the lines of what you're talking about.

share|improve this answer

It seems like that may happen more than you think. Something that sounds mysterious to one person can sound kablam different to another.

share|improve this answer

We do this in the real world when listening to a conversation in a room full of conversations. Our brain filters for us, just like when we interpret an ambiguous 2d image. We will also shift at will (recall those moments that a background conversation becomes super interesting and you have a hard time understanding what the person right in front of you is saying) There are folks that have difficulty with this. See http://en.wikipedia.org/wiki/Auditory_processing_disorder

I agree - it's a fascinating subject. I particularly like MarcoP's example. Perhaps because I really like Reznor ;)

share|improve this answer

Figure/Ground is explained as:

Normally the brain classifies images by what surrounds what - establishing depth and relationships. If something surrounds another thing, the surrounded object is seen as figure, and the presumably further away (and hence background) object is the ground, and vice versa. - Wikipedia

So, we'd have to completely remove the idea of a foreground/background relationship just to even begin to create an aural equivalent.

Do you think timbre would be the correct equivalent of the geometry in Rubin et al?

Shaun, your mentioning of cyclical characteristics makes me think of LaMonte Young's 'Drift Studies', one of which had sine tones playing from small speakers in a space, and as the listener moved the relationship would change, causing phasing and beating. This is close as I can imagine the effect being carried out purely by sound.

share|improve this answer
    
The issue of foreground/background is a good point. Our eyes have the ability to focus on specific points...somewhat to the exclusion of other visual sensory information. Our ears don't have that ability. They hear everything that is audible at all times, and our brain has to filter through all of that information to focus on something. There are greater processes of deconstruction required for acoustic stimuli than there are visual. You may be right about timbre being the equivalent. RE: Lamont Young, a bunch of us tried to go to the "Dream House" in New York while at AES. It wasn't open. :\ –  Shaun Farley Feb 17 '12 at 12:43
    
So, essentially we need to change the listeners brain chemistry. Only one way to do that....!!! –  Brendan Rehill Feb 17 '12 at 19:05

I'm surprised no one has mentioned Steve Reich's early work in this discussion! His tape loops drift in and out of sync, creating evolving harmonies, letting some words shift in emphasis over time, and tricks our ears into making up hybrid "words" as phonemes are juxtaposed and overlapped. Re-listening does pull out different relationships on each listen, although I'm not certain how to measure if 2 people are having similar or different perceptual experiences moment to moment, but in work that complex and weird, I'd bet they are.

This isn't meant as a discrete answer, but rather the closest example I can possibly think of to Steve's original (excellent) question.

I wonder if anyone has ever done a test wherein which of two channels are put into a side chain compressor, and then automate/interpolate/fade which signal is triggering the compressor and which is being compressed. Could be an interesting way to mess with figure-ground relationships over time.

share|improve this answer
    
cool example, Nathan. not familiar with any of Steve Reich's work myself. i'll have to look into that. –  Shaun Farley Feb 21 '12 at 1:13
    
If you want a place to start, here's a classic example, indeed the one that I had in mind when I wrote this post: esnips.com/displayimage.php?pid=10967872 You need to give it a while - the time offset is SLIGHTLY out of sync, take a while. Dated, probably, but NO ONE had done explorations like this before. –  NoiseJockey Feb 21 '12 at 2:08
    
Hooray, @NoiseJockey returns! –  Steve Urban Feb 21 '12 at 6:15

Pretty deep.

I think you'd have to set the listeners up with a way to alter the way they hear a sound - like on hi-fi 20Hz-20K speakers and let them hear something rumbly and the sound sound on computer speakers and all of a sudden you hear something else in the track because the rumble is gone - because the viewers of those optical illusions are shifting how they are viewing it..

Perhaps you can have 2 sounds so similarly balanced and similar tonality that one listen you might hear one sound out of the two, and then next listen you'll hear the other one.

I personally like how in The Grey, there is a character who falls asleep at night and the rest of the characters are looking out for wolves and the sleeper awakes violently from a dream and makes a wolf-like sound and it scared the sh&* out of me in the theater (makes me wonder if the sound designer meant it to sound like a wolf) - something like this might be up your alley.

share|improve this answer

I'm reminded of Sonnenschein's idea of sound spheres. I still need to read all of his published writing on the subject more in depth, but the general idea appears to be focused around how our perception of a sound influences and is influenced by our level of awareness of its context.

Leaving academia behind, I often think of personal experiences I've had with mistaken sound. One morning outside a gas station, amidst the beginnings of traffic rumble, bird ambience and light walla, I heard what seemed to be a woman's scream. I turned, alarmed at first, only to see that the source of the sound was really an industrial vacuum.

In terms of horror and thriller films, I think this device has been used quite a bit through the years, though it's frequently heavy handed and almost becomes a cliche (yet somehow all film examples slip my mind - funny how cliches work like that). But I think there's still room for experimentation with the technique.

Another personal experience, I was sitting in a museum and became aware of a layer of sound hiding just below the general "fill". It seemed like water droplets on a skylight from within a house. I slowly began to walk towards where the sound appeared to be coming from, and gradually realized that it was coming from an air vent. As I moved closer, the proximity effect began to take place, but my overall perception of the sound was also altered as I discovered its source. Initially, I had associated the sound with liquid and transparency, and suddenly I was hearing the sound as metal on metal and air pushing dense materials.

I'm not entirely sure of how recreating this moment could be used in terms of narrative structure, but it was still interesting in revealing to me that the sound source reveal could be used in more subtle ways apart from horror misdirection.

share|improve this answer

Might not be the right answer but, in Blade Runner when Roy says 'fucker' but it also sounds like 'Father' for that kinda double meaning effect.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.