mar 2
2010

Pink Noise at the Movies

NYT's Natalie Angier has a very poetic piece on some new research showing that over the last 50 years, the pacing of movies has tended toward the natural rhythm of the brain (and the universe). It's hard to summarize in a sentence, so Angier explains at length:

The basic shot structure of the movies, the way film segments of different lengths are bundled together from scene to scene, act to act, has evolved over the years to resemble a rough but recognizably wave-like pattern called 1/f’, or one over frequency -- or the more Hollywood-friendly metaphor, pink noise. Pink noise is a characteristic signal profile seated somewhere between random and rigid, and for utterly mysterious reasons, our world is ablush with it. Start with a picture of Penelope Cruz, say, or a flamingo on a lawn, and decompose the picture into a collection of sine waves of various humps, dives and frequencies. However distinctive the original images, if you look at the distribution of their underlying frequencies, said Jeremy M. Wolfe, a vision researcher at Brigham and Women's Hospital, "they turn out to have a 'one over f' characteristic to them."

Researchers analyzed the length of shots in films and noticed the trend, which Angier suggests may explain why movies are so captivating even when they aren't that good. The researchers also seemed surprised that a montage from Rocky IV showing Rocky and Drago training separately featured matching shots of equal length for each boxer. As with the golden ratio, it seems like pink noise is the sort of thing that artists and audiences figure out before scientists do.

An accompanying graph shows how various films align (or not) with the 1/f ratio, objectively and as compared to the average for its year of release. Of all the films analyzed, Back to the Future matched 1/f’ most closely. Even so, researchers noted that there is no consistent correlation between a film's adherence to pink noise principle and its popularity with viewers. --ADM

9 comments

PS: Fans of aptonyms will note that one of the researchers is named "Cutter."

posted by adm at 1:06 PM on March 2, 2010

I LOVE this post.

posted by marrina at 2:17 PM on March 2, 2010

thanks. it is interesting to think that there may be some natural rhythm in the universe that we can create (and therefore behave) according to. if we could figure out the right syllable counts, maybe we could be more captivating in meetings and on dates.

posted by adm at 2:44 PM on March 2, 2010

I'm gonna call bullshit on this one. That last sentence is what kills it. Think of it like the subliminal messaging scare--it doesn't actually work, but it brings up our fear that popular culture is controlling our minds. Lots of recent film theory focuses on cognitive processes that go on in the spectator, but this seems like a very simplistic version of that.

posted by Jared at 2:46 PM on March 2, 2010

Thanks for the thoughts, Jared. I agree the last sentence renders the rest of it a little disappointing. Can you clarify which part you are calling bullshit on, though? I think the article isn't drawing many conclusions, other than that the shot duration/frequency aligns with the pink noise signal, which can (as I understand it) be measured objectively. The cognitive *effect* of that -- which, let's face it, is where all the interesting stuff would happen -- is yet to be researched, as far as I could tell.

posted by adm at 3:00 PM on March 2, 2010

Well, first of all it's suspicious that the Times article itself doesn't do a very good job of explaining what pink noise actually is. After doing five minutes of web browsing, I think the article is not saying anything more than that there is more of a pattern to shot lengths today than in the past. One look at the graph is enough to make me suspicious that average is "approaching" 1/f--sure, it's getting closer, but it's still a long way off. (Plus, that line-of-best-fit looks really weird to me.) This is interesting, but you're right to point out that it has nothing to do with cognitive effect--despite the tone of the article.

I'm interested to read the original study, and I will when I have more time (right). But it seems to me that before we even think about what this means for film attention, we should wait to hear what the Cinemetrics people think about this. They're already got a huge amounts of data that can be analyzed for this. http://www.cinemetrics.lv/database.php

For more on average shot length, and how it's getting shorter, you can also look at David Bordwell's The Way Hollywood Tells It. I'm still skeptical that all these numbers are just going to distract us from more critical analysis of how films work on our brains. Apparatus theory is out of fashion, but still for my money the most critical approach.

posted by Jaredq at 3:42 PM on March 2, 2010

I love that somewhat had to watch "Spies Like Us' thousands of times.

posted by FS at 4:14 PM on March 2, 2010

Great comment, Jared. Thanks.

FS -- that is pretty much my favorite part of the article, too.

posted by adm at 4:21 PM on March 2, 2010

hi ilove this things

posted by bappu at 10:29 AM on March 8, 2010




NOTE: The commenting window has expired for this post.