Why Horror Seduces

by Mathias Clasen

  The horror genre, then, achieves its peculiar affective goal by targeting an evolved defense system, the fear system. That is why psychologists who study emotion are fond of using clips from horror films in their studies. When they want to examine fear in human subjects, they expose these subjects to clips from films such as The Shining (Kubrick 1980) and The Silence of the Lambs (Demme 1990) (Gross and Levenson 1995, Rottenberg, Ray, and Gross 2007). Doing so is effective because fiction produces emotions—not “quasi-emotions” (Walton 1990), but genuine emotional responses (Hartz 1999, Mellmann 2002). A good story can make us sad, happy, surprised, disgusted, afraid, and so on; what we feel when absorbed in such stories is real sadness, real happiness, real surprise. Horror fiction aims to instill negative emotion by having us hold in our minds fear- and anxiety-inducing thoughts—from simple images (a gigantic spider!) to complicated scenarios (what if a secret military experiment breaches the barrier between dimensions and unleashes monsters from another reality upon ours, with chaos and social disintegration as a result?). Films and computer games, being visual media, can directly feed us fearsome visual stimuli—such as a depiction of a scary monster—which are processed by some brain structures as though they were real (Grodal 2009, 101–102), whereas horror literature provides us with verbal descriptions that direct and constrain what we imagine, thus putting images into our heads, as it were. But horror in all media works by targeting the same evolved mental systems. As the neuroscientist Jeff Zacks observes, “when you read a book, you construct a model of the situation described in the book that is in important ways similar to the model you would build if you watched a movie of the same situation” (2015, 42). Representations engendered by films are, on a neurobiological level, like representations engendered by literature.

  It is no surprise that looking at an image of something nasty, say a photograph of a mangled corpse, produces a strong emotional and physiological response. But even holding a nasty image in one’s mind can evoke a forceful response. Remember those urban legends about psychopathic people handing out razor blades buried inside apples for Halloween? Imagine biting into such a nice, red apple. It’s unpleasant to even think about. That’s because our imaginations evolved to enlist appropriate emotional responses when we cognitively entertain hypothetical scenarios—this is part of the imagination’s functional design. The imagination allows for “intellectual simulations and forecasting, the working out of solutions to problems without high-cost experimentation in actual practice,” as Denis Dutton puts it (2009, 105–106). Such simulations are not merely intellectual and cognitive, like the kinds of scenarios we might imagine an advanced computer coldly and rationally running through. When we entertain hypothetical scenarios, when we imagine alternative pasts or possible futures, we feel the emotional import of such scenarios (Tooby and Cosmides 2001, Boyer 2007). We don’t have to have personal experience with sticking our hands into a terrarium full of tarantulas, stripping naked in school, or jumping off the Grand Canyon Skywalk to realize that such actions are, generally speaking, unwise. Just imagining doing so elicits an emotional response that tells us: “No. Just no.”

  Horror fiction typically is designed to draw us in and keep us engaged. It does so by drawing up a recognizable fictional universe (the setting tends to be fairly naturalistic even in stories that feature supernatural monsters); giving us an “anchor” in the fictional world (one or more characters from whose perspective we experience story events and/or with whom we can empathize); and exposing that anchor to nasty events. This structure allows for audience transportation, that is, it allows audiences to project themselves into the fictional world and feel with and for the protagonists. We mirror Danny’s terror in The Shining (Kubrick 1980) and hope he escapes his crazy dad; we feel sorry for—and grossed out by—demon-possessed Regan in The Exorcist (Friedkin 1973); and we are anxious with Agent Starling as she nervously investigates a dark storage facility with an unsteady flashlight in The Silence of the Lambs (Demme 1990). Moreover, horror fiction is particularly well-equipped to grab and hold our attention because of the genre’s typical subject matter. Evolution has designed us to pay close attention to cues of danger in the environment—even fictional environments—and our attention is riveted by the kinds of situations represented in horror.

  A genre-typical scene from the 2007 horror film Cloverfield (Reeves 2007) will illustrate the idea that horror fiction exploits evolved defensive psychological machinery. In this film, a giant monster from space is wreaking havoc on Manhattan. A group of protagonists attempts to escape the island. They travel poorly-lit subway tunnels, carrying along a camcorder with an attached and ineffective light source (see Figure 2.1). Visibility in the tunnels is exceptionally poor. Suddenly, the characters notice scores of rats running in the same direction as them. “It’s like they’re running away,” says one character. “From what?” asks another. An unidentified, distant source then emits a jarring, disconcerting sound, apparently from behind the characters. They, and the audience, know that something is approaching and frightening the rats, but the poor visibility prevents identification of the potential threat. When the characters think to enable night vision functionality on the camcorder, they and the audience see several dog-sized, spider-like organisms with sharp teeth approaching in the tunnel. The monsters attack the characters who scramble for safety. One character is severely injured.

  Figure 2.1: Protagonists fleeing an enormous monster in Cloverfield (Reeves 2007). The spectator is visually aligned with the characters through point-of-view cinematography. It feels as if we’re in the dark subway tunnel with the characters, anxiously looking out for danger. This claustrophobic scene exploits evolved fears of the dark and of predation.

  The mise-en-scène tells us that we are in a confined location with few possibilities for escape. Adding the knowledge that a dangerous monster is loose, we realize that this is a potentially catastrophic situation and mirror the characters’ anxiety. Cinematography (the use of a hand-held camera, operated by a character) and narration (restricting the audience’s knowledge to what the characters know) keep the audience in the dark, along with the protagonists. Darkness and visually obstructed scenery are mainstays of horror fiction. Humans, with our poor night vision, are especially vulnerable to ambush in the dark, and research has demonstrated that darkness increases anxiety. This is an adaptive response to a potentially dangerous situation, given that darkness makes threat-detection much more difficult. The startle reflex—a defensive, hardwired reaction—is enhanced by fear and aversive states, a phenomenon known as “fear-potentiated startle.” In one study (Grillon and Davis 1997), subjects in a dark environment reacted more violently—with greater startle—to aversive acoustic stimuli (brief, loud bursts of white noise) than did subjects in a well-lit environment. Darkness sets the fear system on edge; when people talk about fear of the dark, they usually mean fear of what dangers may hide in the dark. As Lovecraft observed, “the oldest and strongest kind of fear is fear of the unknown” (1973, 12). Nowhere is the unknown more forcefully present than in the dark. In the Cloverfield example, characters are confined in dark tunnels, and cues of approaching danger (the jarring noise) engage the spectator’s fear system. We are kept on the edge of our seat, anxiously scanning the virtual environment for signs of danger along with the characters in the film. The horror cliché of depicting characters anxiously investigating a visually obscure and threatening environment is an efficient way of creating cognitive and emotional engagement in the audience because of the design features of evolved neurocognitive defense mechanisms.

  Taken alone, such scenes are rarely that scary because horror stories tend to be structured to build increasingly strong emotional responses over the course of the story—again, by exploiting evolved aspects of the human fear system. When we are exposed to stimuli that produce fear or anxiety, we become sensitized to such stimuli and react more strongly to subsequent stimuli. That is why many horror stories begin with a strong opening that
suggests the presence of some kind of more or less vaguely defined threat. The first footage we see in Cloverfield is text on a black background, suggesting an amateur (and thus authentic) video recording. The first shot has text that says “PROPERTY OF THE UNITED STATES GOVERNMENT.” This is replaced by a shot that says “DOCUMENT #USGX-8810-B467 DIGITAL SD CARD MULTIPLE SIGHTINGS OF CASE DESIGNATE ‘CLOVERFIELD’.” The third and final shot of the opening has text that says “CAMERA RETRIEVED AT INCIDENT SITE ‘US-447’ AREA FORMERLY KNOWN AS ‘CENTRAL PARK’.” Within a few seconds, the film lets us know that we are about to be told the story of some hugely destructive event, something that destroys Central Park and requires serious military intervention. But we don’t know what that might be; our interest is piqued. Similarly, The Exorcist grabs its audience by portraying weird disturbances—noises from the attic, strange omens—a few minutes into the film. Less overtly, Halloween (Carpenter 1978) suggests menace in its opening title sequence, which features a slow track-in on a pumpkin against a pitch-black background. The pumpkin’s eerily glowing eyes and mouth and the unsettling music let the audience know that something nasty is going to happen in this film (see Figure 2.2).

  Figure 2.2: A pumpkin with glowing eyes from the title sequence of Carpenter’s Halloween (1978). The slow track-in and eerie music establish an unsettling atmosphere of apprehension. Many horror films open with the suggestion of some more or less vaguely defined threat, thus priming viewers’ fear system through sensitization.

  These films, and countless films like them, attempt to sensitize the audience to danger early on. Literary fiction uses the same tactic, early in the story introducing clear or ambiguous signs of danger, and horror video games tend to open with some backstory or background information about the gameplay universe which suggests a world fraught with danger. It is an effective narrative strategy because sensitization lowers the threshold for fear (Dozier 1998). Fear generates more fear and increases our attention to threats in the environment (Öhman 2008). This extends to fictional environments. The well-constructed horror tale has us anxiously scanning the fictional environment for threats, and our emotional reactions increase in intensity as the story progresses. Seeing a YouTube clip of the infamous spider-walk scene from The Exorcist (Friedkin 1973) is much less horrifying than seeing the sequence in context, primarily because of the cascading effect generated by horror fiction. (A secondary reason is that we don’t really empathize with the characters if we haven’t seen the bits that came before, and lack of empathy means lack of emotional investment.) The film has sensitized us to fear, it has gradually built up an atmosphere conducive to anxiety (we know that something is wrong in the fictional world, that something nasty is likely to happen soon). An even more striking example is the final scene of The Blair Witch Project (Sánchez and Myrick 1999), where two of the protagonists enter a dilapidated house in the woods. It’s dark, they are afraid, and the film has given us reason to believe that something nastily dangerous is in the house. When seen in context the scene is utterly terrifying; when seen in isolation, much less so. Suspense, expectations, and empathy have been built up during the film—by the time we reach the end, we are invested in the characters, anxious about whatever is hunting them, unsure of what will happen but expecting the worst.

  Horror entertainment is designed to scare us, unsettle us, disturb us. Given that our evolved constitution provides only a limited number of ways in which to scare and unsettle us—the things that scare us are distributed in a possibility space constrained by natural dispositions—the monsters and scary scenarios of horror entertainment are likewise nonrandomly distributed and tend to be exaggerations or embellishments of historically prevalent dangers. The big, attacking spiders in the Cloverfield subway scene are a case in point. Spiders play no real role in present-day mortality statistics in the industrialized west, so why would they figure so prominently as horror monsters in our scary entertainment? Only by looking at the deep history of human evolution can we begin to answer that question.

  CHAPTER 3

  How Horror Works, II

  Spooky Monsters, Scary Scenarios, and Terrified Characters

  The most effective monsters of horror fiction mirror ancestral dangers to exploit evolved human fears. Some fears are universal, some are near-universal, and some are local. The local fears—the idiosyncratic phobias such as the phobia of moths, say—tend to be avoided by horror writers, directors, and programmers. Horror artists typically want to target the greatest possible audience and that means targeting the most common fears. As the writer Thomas F. Monteleone has observed, “a horror writer has to have an unconscious sense or knowledge of what’s going to be a universal ‘trigger’ ” (qtd. in Wiater 1997, 119). All common fears can be located within a few biologically constrained categories or domains. Over evolutionary time, humans and their ancestors have faced potentially lethal danger in the domains of predation, intraspecific violence, contamination-contagion, status loss, and in the domain of dangerous nonliving environmental features (Barrett 2005, Boyer and Bergstrom 2011, Buss 2012, 71–103, Marks and Nesse 1994). In other words, they faced danger from predatory animals (ranging from mammalian carnivores to venomous animals such as spiders and snakes); from hostile members of their own species; from invisible pathogens, bacteria and viruses; from loss of status, ostracization, and ultimately social exclusion, which in ancestral environments could mean death; and they faced the risk of lethal injury following dangerous weather events such as violent thunder storms, falls from cliffs, and other potentially hazardous topographical features. The selection pressures from these types of danger have resulted in domain-specificity in the reactivity of the fear system, meaning that the system has evolved special sensitivity toward such dangers. Sometimes such sensitivity allows the fear system to unreasonably expand a category and target an innocuous object, such as expanding the category of “dangerous animals” to include moths. As I mentioned earlier, in the domain of survival the golden rule is “better safe than sorry.”

  The most basic, universal, genetically hardwired fears are the fears of sudden, loud noises and of looming objects—those are the fears that we aim to evoke when we hide behind a door, waiting to spook an unsuspecting friend by jumping at them with a roar. Sudden, loud noises and looming objects will cause an involuntary startle response in humans and in many other species as well. You can sneak up behind a rat and yell at it, and its reaction will be similar to your own, if somebody sneaks up behind you and yells at you. You can also try the experiment with a dog or a squirrel or a human infant; it’s guaranteed to work. The startle reflex is primitive and swift, and very effective in orienting the organism toward, and preparing it for, danger. Horror video games and horror films, in particular, exploit this innate fear when they resort to “jump scares,” such as having a monster jump out of a closet without any warning and frightening the viewer or player.

  The jump scare arguably achieved its purest mediated form in the phenomenon known as the “Internet screamer,” which is an animated file—typically a short film—that shows a static peaceful scene for a while and concludes with an unexpected, shocking element, often a disturbing image and a loud noise. Screamers started circulating the Internet in the mid- to late-1990s. One of the most well-known screamers, called “What’s Wrong with this Picture,” was an image of a peaceful dining room. Viewers would scrutinize the image, looking for whatever was supposed to be “wrong” with it in the manner of a “spot the difference” game. After about thirty seconds, the image would abruptly shift to briefly showing a black-and-white close-up of a screaming woman with black holes for eyes, accompanied by the sound of a scream. Those viewers who were caught unawares—and there were plenty of us back in the early 2000s, when this prank started circulating the Internet—were in for a very nasty scare. Nowadays, YouTube overflows with “reaction videos” showing footage of people being scared witless by screamers. Apparently, there is a peculiar pleasure in watching other people react with vi
olent negative emotion to a basically innocuous stimulus. It may be a form of Schadenfreude, a pleasure in seeing others humiliated and reduced to the most basic behavioral response by their primitive biological hardwiring. Even the high and mighty, the urbane and sophisticated, the meek and innocent—even they jump and scream when they play the “Scary Maze Game,” a simple piece of online software that has players focused on moving a dot through a maze on the screen, only to be unpleasantly surprised by a flash of diseased-looking Regan McNeil from The Exorcist accompanied by a loud scream. One 52-second-long YouTube video, showing a boy reacting with abject terror to the Scary Maze Game, has been viewed more than twenty-seven million times at the time of writing (“Scary Maze prank—The Original” 2006).

  In critical computer-game parlance, the strategy of using the sudden, unmotivated appearance of a monster to scare players has become known as the “Monster Closet” tactic. A narratively unmotivated Monster Closet is by many reviewers and gamers perceived as insultingly gratuitous and cheap, but there is no denying that it can be highly effective. The tactic achieved notoriety in the 2004 survival horror first-person shooter game Doom 3 (Willits), which featured monsters literally jumping out of closets, spooking the player. Gratuitous jump scares also tend to be frowned upon by horror film fans, who see it as a primitive, artless way of scaring them. One of the most famous jump scares in horror film history, however, is not only surprisingly effective in generating a startle response but is narratively and thematically motivated—artful, in other words. It is from the final scene of Brian De Palma’s Carrie (1976). The film’s protagonist, abused and telekinetic Carrie, has killed herself after wreaking havoc on her school and killing most of her tormentors. The only survivor of Carrie’s high school massacre, Sue, dreams that she is visiting Carrie’s grave. As Sue is placing flowers on the grave, a bloody hand shoots up from the soil and grabs Sue’s arm. The scene has terrified countless viewers—the movie is almost over, Carrie is dead, the scene has a slow, peaceful, dreamlike quality which is supported by the calm musical overlay . . . and then that bloody arm erupts. More than giving viewers a jump scare, however, the scene serves a function in suggesting Sue’s guilt and the psychological trauma she herself has suffered. Carrie may be dead and the mayhem may be over, but Sue will forever be haunted by her role in pushing Carrie over the edge.