The Universal Sense (15 page)

Fear is one of the most studied emotional states, possibly because it is sort of an emotional primitive—even frogs can feel fear (if you hang out with them long enough to understand their highly non-verbal body language). It is also one of the few emotions with a well-characterized anatomical and physiological basis, and it is usually studied with sound using a very old technique called classical or Pavlovian conditioning. The basis of classical conditioning is relatively simple: you have an unconditioned stimulus, something to which you would have a reflexive response, such as the sight of a steak making a hungry dog (or grad student) salivate. The steak is the unconditioned stimulus and the salivation is the unconditioned response. The trick Ivan Pavlov came up with is stimulus substitution. Right before showing the
hungry dog the steak, he rang a bell (the conditioned stimulus). After a few repetitions in which the experimenter rang the bell within half a second of presenting the steak, the dog’s brain created an association between the ringing of the bell and the presence of the tasty steak and actually rewired its reflexive response to salivate in response to the bell, even if there was no steak.
²⁴

Scientists who study fear rely on classical conditioning and other techniques using sound, in large part because there are traceable neural pathways leading from the ears to a region of the brain that modulates fear and fear-like responses, the amygdala. The amygdala is one of those rare brain nuclei that gets a lot of press—though most of the information out there in the popular media is wrong. If you get your science from news websites, you will often hear about new and exciting findings showing that anything having to do with control of emotion or the failure of such control—criminal behavior, political orientation, the perfect diet, the ultimate date, why chocolate is the same as sex—is associated with the amygdala somehow. The popular press view is that the amygdala is the brain’s “emotional center” and that if you do something socially unacceptable, it’s because your amygdala is screwed up.

The truth about the amygdala is much more interesting. It gets input from both the fast (thalamic) and slow (cortical) pathways and provides output throughout the cortex. The fast pathway provides a quick and immediate reaction as well as a basis for learning the difference between dangerous and non-dangerous input; the cortical pathway, which wends its way
through the memory and associational regions of the brain, is slower but much more accurate in determining if an emotional response to a sound is valid. Was that a scream in the film you’re watching or was someone just butchered behind you? (A really good film will blur that boundary because of good sound editing; we’ll talk more about that later.) But a real and complex fear response is triggered rapidly, then maintained and strengthened due to a network of connections in your brain, not just the amygdala. Sounds associated with fearful events get looped forward and backward through the amygdala, the auditory cortex, and the hippocampus, the last of these being the gateway to long-term memory storage. Early in these loops, information is fed to deep cortical regions and hypothalamic areas associated with control of autonomic systems such as blood pressure and heart rate, modulating the rapid release of adrenaline, making your heart beat faster, constricting your pupils, and drying out your mouth. A frightening sound, such as a very loud lion’s roar, causes body-wide physiological responses that feed back to your brain, building on the early emotional response and simultaneously comparing older memories of frightening events with new input to see if what frightened you is still really scary. If the roaring sound has faded, meaning you outran the lion, your sympathetic nervous response starts to diminish and you calm down, still aroused and wary but with the time to look around and confirm that you are no longer in danger.

Certain elements of sounds act as basic triggers, even without previous experience. Sudden loud sounds can cause you to startle, but if you add in very low pitch, your brain starts making subconscious associations. As with frogs selecting mates, loud and low-pitched means large. Large sources of sounds might be
desirable if you are a female frog looking for a mate, but in a normal human day, large is often scary. One evolutionary argument has been made that we have been biologically hardwired to interpret loud sounds at the very low end of human hearing and lower, in the realm of infrasound, as signaling “predator.” Analyses of the roars of big cats such as lions and tigers have demonstrated the presence of a high-amplitude infrasonic element. Evolutionary neuroethologists, who study the ways our brain-driven behaviors have changed through evolution, have argued that those animals that did not automatically run away when they heard such sounds were the ones that got eaten and didn’t pass on their genes. But another hypothesis, which combines non-auditory physiological acoustics with a tie-in to autonomic sensitivity, points out that loud infrasonic sound not only is heard but is felt throughout the body, vibrating the organ-filled abdomen, the air-filled lungs, and even the bones themselves, similar to the non-tympanic pathway in frogs. Vibrating the rest of the body at low frequencies can induce nausea and a feeling of sickness via the enteric nervous system, a poorly understood subsection of the autonomic nervous system that provides sensory feedback from your internal organs. This low-frequency non-auditory pathway underlies things such as vibroacoustic disease, which affects some construction workers with excessive exposure to jackhammers and other construction-based vibration. So a sudden, loud low-frequency sound is not merely triggering basic auditory connections but taking input from your whole body and telling you to
run
. Interactions with memory and attention regions come later (and more slowly) but guarantee that you will remember that the sound was associated with danger, and help you respond even faster next time you hear it, thanks to Hebbian plasticity, the brain’s ability to
rewire itself to speed up responses to previously encountered events. The scary sound has become a survival tool—next time you hear that roaring sound under the same circumstances, you’ll probably start running faster and not waste time yelling, “What the hell was that?”

What about sudden loud sounds that are not particularly low-pitched and hence probably not life-threatening? Then the context provided by memory and previous associations becomes more important. For example, you’re cruising the web and you hit a web page that suddenly screams “Congratulations—you just won an iPad” or blares some band’s latest thirty-second low-sample-rate offering. Irritation closely follows the startle, and you almost instantly close that page (or take the more drastic measure of turning off sound when browsing the web). Sounds that are intrusive but not immediately associated with danger assume what is called “negative valence”—feelings like annoyance or anger. They are a response to a false alarm: you got all ramped up because of a sudden sound, but it’s just another familiar irritation. This is one reason the use of complex sound in technology is problematic. Remember the talking-car warnings of the 1980s? A sudden poorly synthesized human voice with no one in sight telling you “Your door is ajar” or “Your seatbelt is not fastened” was startling and rapidly categorized as irritating, especially when it would preempt the stereo.
²⁵
More recently, most web pages (at least those from smart designers) no longer use complex sound as soon as you land on them—the sudden sound is an irritant, and if it lasts longer than the few
seconds it takes for you to visually recognize the image as something you find interesting or desirable, you close the window.

One of the most widely known negative-valence sounds is one you probably tortured classmates with back in elementary school. Back in 1986, Lynn Halpern, Randolph Blake, and James Hillenbrand wrote a wonderful paper called “Psychoacoustics of a Chilling Sound” that did something that far too few scientific papers do: it addressed a really basic question. Why do we hate the sound of fingernails on a blackboard? (Not to mention metal, such as a rake, being dragged across slate, which was rated as an even worse sound.) They hypothesized that the spectrum of the sound closely matched that of the warning cries of the macaque monkey, and so fingernails on a blackboard was the sensorineural equivalent of a primate alarm call. This was a cool idea, and it was widely cited, but, as with most non-mainstream scientific studies, it didn’t get much further testing beyond a paper by Josh McDermott and Marc Hauser in 2004 showing that cotton-top tamarins (very small, very cute monkeys) didn’t show the same response, possibly due to the fact that they rarely interact with blackboards or metal rakes. So the “Chilling Sound” study’s results remained unconfirmed. But the psychological effect of this sound is nearly universal among humans—fingernails on a blackboard and metal on concrete both create sounds that make you want to plug your ears and throw sharp objects at the person making the noise, no matter your age, sex, occupation, or cultural upbringing.

Having lots of sound equipment about and being an academic, I had access to blackboards and fingernails, as well as a metal rake and a concrete driveway. So, after putting in earplugs and sampling both these sounds, I noticed something very interesting: the fine timing structure of both sounds was actually what’s
termed “pseudo-random”—that is, the underlying waveforms were
almost
periodic, repeating their fine structure over time, but there was enough random variation to make it a temporally messy sound. It’s sort of like looking at a badly Photoshopped image, with the ragged edges, out-of-place colors, and pixels making it obvious that something is wrong. The only other time I’d seen anything like that was in a recording of people shrieking at the top of their lungs. And then it snapped together in my head: our reaction to these sounds is probably based not on the frequency content of some ancestral warning or alarm call but on the pseudo-random variations in the fine time structure, just like those that show up when someone screeches uncontrollably as if in pain or a panic and the normally harmonic structure of the voice becomes ragged and out of control. It isn’t the whole sound that causes the reaction, just a piece of it, but such a basic piece that it elicits a profound and unpleasant reaction.

So what does this mean? If we hate pseudo-random sounds, do we like regular periodic ones? Well, sometimes. As mentioned earlier, living things tend to make harmonic sounds, with regular mathematical relationships between frequency bands and very periodic timing. But sometimes a fine acoustic detail can shift the valence of a sound, turning our associations with it either negative or positive. One of my favorite demos is one I discovered by accident when playing with a sound editing program. I ask students to rate two sounds for various factors, basically ranging from comforting to alarming. I start off by playing a sound that is almost universally frightening: the sound of angry bees. Even though it’s not low-pitched or even necessarily very loud, this sound is definitely fear-inducing, and not just in humans—one study showed that elephants will actually move far away and emit specific warning vocalizations when they hear
this sound. My students all rate this as an alarming sound. Then I play a sound that is almost universally calming: the sound of a cat purring. This is almost always rated as a pleasant sound. But then I add a twist: when I take the sample of the cat purr and speed up the amplitude modulation rate from a few cycles per second to several hundred per second, it begins to sound just like alarming bees. Taking the calming respiratory-like sound and speeding up its repetition rate (and throwing in a little randomness) changes the valence of the sound radically, mostly likely based on previous experience and associations of the two sounds.
²⁶
In another demo that shows how powerful previous associations can be, I put on a sound that has sort of a sizzling, ratchety repetitive quality. On half a sliding blackboard, the sound is identified as “rain on a sidewalk.” I ask students to rate it: is it a nice sound, is it pleasing, does it make them comfortable or uncomfortable? Most of them say it’s soothing, reminding them of rainy afternoons with not much going on. And then I shift the board over to reveal the line “it was actually mealworms devouring a bat carcass.” You can almost see neurons melting in their brains as they go from neutral to “ewwwwwwww” in three-tenths of a second.

But what about a lack of sound? Silence, especially in a normally noisy context, can be an extraordinarily powerful emotional acoustic event. Since we are always presented with subconsciously monitored background noise, a sudden lack of
outside sound leaves an awful lot of attentional and arousal control bandwidth available. A perfect example is the old movie cliché: two explorers are wandering through a jungle when one stops and says, “Do you hear that?” The other responds, “I don’t hear anything,” to which the first one responds, “Exactly. It’s too quiet.” The detection of the absence of sound, while slower than the detection of a sound, triggers its own set of responses, increasing attention and arousal, which can lead to internal mechanisms of increasing your ear’s gain or sensitivity. This increased arousal from silence has the same effect as increased arousal from a startle or a frightening sound—it heightens your emotional preparedness. A study by Denis Paré and Dawn Collins examining conditioned responses to a series of tones followed by a silent period showed increased blood pressure and synchronization of cells building up steadily during the silent period. This suggests that silent anticipation before something unpleasant was critical for learning about unpleasant or scary stimuli. It may be that inappropriate silence is not so much frightening in itself but sends signals throughout your brain that something is missing, something is
wrong
, preparing you for something bad. It may be as basic as the lack of crickets in a forest at night, making your hindbrain wonder if they heard something padding about that you didn’t, or something as complex as the utter silence of UC Davis students as the school president walked by after an incident in which a police officer pepper-sprayed some non-violent protesters on campus—a social warning by denial of sound.