Psychologist Harold Gouzoules on What We Know About Screaming (Not Much) | Q&A

We scream when we’re excited or happy; we scream when we’re fearful or in pain; we scream when we are exasperated; we scream when we’re charging into battle; we scream during sex. But we rarely stop to wonder what those screams, even the ones that erupt from us, signify or if they can be differentiated. Emory University psychologist Harold Gouzoules thinks in those terms, but despite being probably the world’s foremost expert on screaming, he doesn’t speak in absolutes. For decades, Gouzoules studied screams in macaques and other nonhuman primates. He’s only worked with Homo sapiens for three years and answers to even the most basic research questions remain elusive.

How do you collect your data? Do you have people come into your lab and scream?

You know, our IRB — the committee that reviews research proposals — probably wouldn’t like us actually scaring our participants. The approach that we’ve taken is to have participants listen to screams delivered via headphones from a computer, and then answer a series of questions about those screams. It’s very focused on the perception and interpretation of screams, rather than having people produce screams or generate screams in people by scaring them.

A lot of the screams that we use in our research come from movies or TV shows, and another set comes from real screams — because, of course, people upload their entire lives to YouTube now. You can harvest actual events, whether it’s a response to an earthquake or the meteor exploding over Russia a couple years ago.

An interesting empirical scientific question is whether the screams that are produced by an actor are identical, or very similar, to those produced by people experiencing real events. There’s a concern among critics who would say an acted scream doesn’t represent the real emotion, but that assumes that people communicate to honestly signal their internal states. When, in fact, what’s more likely is that, yes, sometimes we’re communicating in an honest way, but many times we’re not. We manipulate others by conveying thoughts and emotions that we don’t really feel.

Some people are much better at this than others. And who are the best? Well, I would suggest that those people who are paid $20 million to play roles in movies.

Have you found that people can tell the difference between a genuine scream on YouTube and a scream from an actor?

We have pilot data for this, and we are pursuing this with one of my new graduate students. But the preliminary data suggests that people are not good at judging whether a scream is real or not, confirming my initial suspicion. Again, we don’t have a completed study on that. We have a start of a study and preliminary data. But so far, the suggestion is that people are not good at saying whether it is an acted version or a “real” scream.

I have no doubt that if I took a Jamie Lee Curtis scream from one of the Halloween movies and played it out on the street corner here at Emory, that I would get a reaction.

Is it because there’s this wealth of YouTube videos and other social media records that lets us study screaming in new ways?

I think you’re absolutely right. I think that the technologies now available over the last 15 years or so will give people an opportunity to study events that, in the kinds of worlds most of us live in, are rare. When I study monkeys I can go out and predictably record fights and screams, but I would be hard-pressed to do that with the folks living in Atlanta. Mind you I also collect newspaper clippings and so forth where there’s reference to victims of a robbery or a sexual attack and they screamed, and people hear and respond. Sometimes they’re the cops and so forth. It’s not to say that people don’t scream in our everyday world, but it’s rare.

How do you delineate between a scream, let’s say, and a moan or a yell? What makes a scream a scream?

That’s one of the things I’m trying to figure out. And the way I do that is through a very simple experiment where we have a range of vocalizations, some of which I would label as scream, some of which would be crying, laughter, moaning. We ask participants, simply, is this a scream or is this not a scream? Essentially it’s getting at the classification system that we use, either inherently or as a result of the experiences that we have. In the modern world, that’s through exposure on TV or movies, and presumably in ancestral times, through observation of real-world traumatic events.

And people are remarkably consistent. With 80, 90 percent consistency people will label vocalizations as screams even if they are acoustically very different. So going back to the different kinds of screams, what we might call a startle scream — those tend to be what my wife would give when she sees a cockroach or a mouse. Or when you’d go around a corner and somebody comes at you and startles you. That vocalization, people label that a scream, even though it’s different from the scream that would be produced when somebody comes at you with a knife or whether you’re giving birth.

I’m sure I’m conditioned by horror movies, but it sort of intuitively makes sense that there would be a difference between a terrified scream and a thrilled scream.

Here’s the interesting thing: It’s an empirical question. Meaning, we might suspect that we should be able to discriminate and possibly interpret those different kinds of screams, but whether we’re actually able to do that is not established.

You hear kids screaming down the street, and you’re able to say, oh they’re just having fun at the swimming pool, and it’s summer and they’re out of school. But is that conclusion based on the fact that you know there’s a swimming pool down there, and that the kids collect and have fun every afternoon? Or are there acoustical differences that allow you to say, that’s just a scream of people having fun? I don’t know if it’s all that easy to distinguish.

Whereas an aggressive scream — somebody’s in a fight and they scream — people are pretty good at judging that as different from fear or excitement.

Is there an acoustic aspect to the aggressive scream that stands out?

There is, and I say that based on people’s responses. We have not yet been able to track down exactly what that is. That’s a harder, more challenging question.

I’ve got my suspicions that there’s a deeper frequency in the scream. Fear is more associated with a higher pitch, and aggression is associated with a lower pitch.

Is there a burning question about screaming that you would love to see answered?

I’m interested in the male-female differences. It’s not to say that males don’t scream, because you go to a sports event and they will scream as well. But there are clear sex differences.

Let me raise one other interesting context: screaming at rock concerts. In my generation it was The Beatles, today it might be One Direction, and the generation prior to mine it would have been Elvis Presley, and you go back even further, Frank Sinatra. Screaming at those concerts — and who screams? Guys or girls? It’s girls. Why is that?

Some people might dismiss this as sort of a popular culture phenomenon, ‘Oh my grandmother screamed at Elvis Presley and I scream at One Direction.’ I think it’s more interesting than that. Because if you go back into the literature and you look at widely disparate kinds of settings, you find screaming directed at powerful or otherwise attractive individuals. You go back to the pre-World War II Nazi rallies in Germany and when Hitler spoke, women would scream.

What’s exactly going on there? It’s hard to say. There’s just so much room for speculation in this whole area. It’s important to keep in mind that screams attract attention. Is it competitive — competitive screaming? I’m not suggesting that they’re consciously competing with one another for attracting attention — I don’t think it’s a conscious thing at all. But I think it might be how these screams, in this kind of context, are functioning.

There’s a hysterical component to it, you know? And that’s exactly how the historians described these screams to Hitler, as if there was a religiosity or hysterical component to it. That was really perplexing to them.

In the spirit of Halloween and popular screams, are you familiar with the Wilhelm scream?

Of course!

Do you have an idea why this is such an iconic scream?

It’s been used in literally hundreds of movies and TV commercials. It’s amazing how few people actually recognize it — they’ve heard that scream hundreds of times. It’s a joke among the directors and the producers and the sound technicians. They insert it, especially in Spielberg and Indiana Jones movies and Star Wars.

That relates to what I was talking about using acted screams. Clearly, anybody who would use the Wilhelm scream as a representation in an experiment would be an idiot. Do you remember the original Star Trek show with William Shatner? One of the running jokes in there was that the red shirt characters, the journeyman crew on the Enterprise, they get knocked off every episode. And of course nobody invested much in the way of high-priced actor or sound effects on them.

So our rule of thumb is never use a red shirt character’s scream. If you want to do Jamie Lee Curtis or Kim Basinger — they’re doing their own stuff. Not a red shirt character.