The sound of fear

31 October 2016

Interview with 

Professor Dan Blumstein, University of California, Los Angeles

Share

One of the things people do when a film is getting too spook, is turn down the An adult yellow-bellied marmot volume. Sound is a really big character in horror films, but what makes a sound scary? Dan Blumstein is a professor at the University of California, Los Angeles, and he explained to Georgia Mills why small ground squirrels might hold the answer...

 Dan - I study marmots which are big ground squirrels, and I run a long term study of marmots that have been studies in Colorado since 1962. And for this study we catch marmots, and we mark them, and we release them, and we study them throughout their lives. So I know marmots pretty well and I'm enthused by marmots.

I'm casually handling a baby marmot one day and we put marks on them. I'm holding it gently and it starts screaming and I almost dropped it and I wondered why I had that feeling - why did I have an emotional response to a marmot screaming in my hand?  I'd never had one scream in my hand before. I hear them alarm call all the time - I don't have emotional responses there. And that got me asking what's going on with this scream, what's going on with this sound? 

Georgia - Ladies and gentlemen, the marmot scream:

Eeek eeek

Georgia - How did you make them scream?

Dan - well we weren't making them scream, we were just gently holding them and pulling hair from their back to get DNA and putting marks on their back so we could tell who they were. And every once in awhile they would scream and we'd say oh no, not again.

Georgia - Thankfully no squeezing of marmots then, but why did this sound have such a pronounced effect?

Dan - So I'm reading around, and looking and  listening to screams and it turns out that Darwin didn't know what marmots were - nowhere in Darwin is the word marmot. But Darwin knew what screams where. Darwin said "that screams where cries from young to assist help from their parents." And I said "okay that's interesting" - all these species scream. And I listened to this hunting tapes which are horrible things that hunters use a lures to lure in predators usually to kill then or trap them and, if you look at the structure of these screams, they're all the same. If you listen to humans scream, they're very similar as well.

Screams are produced when animals overblow their vocal production systems - what's that mean? Screams are produced when a system becomes over-stimulated and works outside its tolerance. So if you turn up your stereo it sounds good, it sounds good as it gets louder, you're rocking along and then suddenly it's way too loud and the sound becomes sort of predictably unpredictable. It becomes noisy, as rapid frequency shifts it just sounds bad. Well that system is no longer acting in a linear way. And there's a series of these acoustic attributes that are associated with overblowing a system, and none of them is noise. Specifically, it's deterministic chaos but it sounds like noise. It sounds raspy, it sounds rough. And these things are found in all animal screams and human screams as well.

So it seems like there's something because of how screams are produced. There's something common among many species that makes screams particularly evocative that we listen to other species vocalisation and maybe have an emotional response.

I'm giving a popular talk about this in LA. There are no marmots in LA but there are a lot of people who make music and film soundtracks and I said "I bet this works for a movie soundtracks. I bet musicians do this to influence emotions." And one guy comes up later and says "yeah you know I'm a film score composer and I study how emotions in music are communicated. And I said "oh that's interesting" and he goes "I bet you're right." and I go do you want to collaborate," and he's like "yeah."

So we ended up getting an honour student to work with us and the three of us sat down together and we're listening to a bunch of these sounds. We're looking at movie soundtracks and we're getting the best movies, the best horror movies, the best adventure movies, sad, dramatic movies. And we had a series of predictions. We said that these nonlinearities are really emotionally evocative at a fundamental level, then we should expect more of them in really scary scenes and really scary movies, and fewer of them in sad, dramatic films, and things like that.

We scored them and we found something really interesting, we found that these nonlinear acoustic attributes - this noisy type stuff - was over-represented in climatic scenes in horror films, and was under-represented in sad, dramatic films. Suggesting, in a correlative way, well yeah the noise and these nonlinearities are emotionally evocative and they're used specifically to get people's attention and make them aroused.

Georgia -   Next, Dan wanted to see how this noise impacted our emotions. He teamed up with composer Greg Bryant, they made marmot inspired music to go along film clips- little ditties either with or without this 'noise.'

Dan - And what we found was that there were emotional responses to these sounds, specifically the noise and abrupt frequency shifts whereby people felt more negative and more aroused when they heard these things.

So basically, the story is by listening to my inner marmot, I discovered that there's a way to understand why we're scared, what's arousing, what's fearful. And what's fearful are these nonlinear type sounds that are produced honestly when animals are really scared and we listen to these and respond to these because our ancestors listened to these and responded to these. And we listen to these and respond to these because they're normally associated with bad situations. When animals are normally screaming they're being killed or in serious dire straights and they're sort of unbluffable, honest, messages that you should pay attention to what's going on and be scared.

Georgia - So what about it is unbluffable?

Dan - When animals are really scared, if they sort of overblow their system they're reliably producing these sorts of sounds. And if they're reliably producing these sorts of sounds only in situation when they're really scared and overblowing their vocal production system, that becomes a very honest signal. There's no reason to necessarily have that if you're not scared, but it might be an inadvertent honest signal that's associated with being scared or a cue that others can cue in to and can associate the producer of that sound as being in a bad situation.

Comments

Add a comment