CHAPTER 4

Repetition, Surprises, and Goose Bumps

Repetition in music

I hope you enjoy this joke—even though you may have heard it before:

Question: What do you call a fly without wings?

Answer: A walk

I hope you enjoy this joke—even though you may have heard it before:

Question: What do you call a fly without wings?

Answer: A walk

The previous eight lines tell us something very important about repetition. The fact is that there is no such thing as true repetition as far as a human being is concerned. When you read the joke for the first time, it created a particular response, which might in this case have been a smile of amusement or recognition. When you read it a second time, you may have thought “This must be a printing error” or “Is this any different from the first version?” but you certainly didn’t experience a repeat of your earlier smile/recognition response. The print on the page is an exact copy of the earlier rendition, but your attitude toward the lines has changed—simply because they were repeated.

In most contexts (reading, conversation, jokes, etc.) we don’t usually value or enjoy repetition. But there is one exception. We love repetition in music, which is why nearly all music is extremely repetitive.

Pop songs are repetitive, classical music is repetitive, rap is repetitive—I could go on…

Of course some “repeats” in music are not really identical. There may be a change of harmony under a repeated melody, or the same melody might be played by a different instrument. But even if the repeat is an electronically produced exact copy, our response to it is adjusted by the fact that we heard it earlier.

This is one of the reasons why musical repeats are not boring. If they were boring we’d have a problem, because there is an extraordinary amount of repetition in almost every genre of music.

You’ll have noticed that I said “almost” every genre. The exceptions are rare and usually involve intellectual modern classical composers deliberately trying to avoid repetition at all costs. To investigate our addiction to musical repetition, music psychologist Elizabeth Margulis played recorded excerpts from little-known pieces of this sort of non-repeating music to a roomful of professional music theorists (people who could be expected to be well disposed to new musical ideas such as the avoidance of repetition). Dr. Margulis played some of the excerpts in their original “no repetition” versions, but without telling the audience, she also tampered with some of the pieces, adding repeats of short sections of the music. For example, if the composer had written a piece made up of three segments, intended to be played in the order 1, 2, 3, the audience heard the music with the first and third segments repeated, like this: 1, 1, 2, 3, 3.

The audience members were then asked how much they enjoyed the various pieces of music, and the results showed that they much preferred the pieces that included the repeats.

As Dr. Margulis says in her fascinating book On Repeat:

This is a stunning finding, particularly as the original versions were crafted by internationally renowned composers and the (preferred) repeated versions were created by brute stimulus manipulation without regard to artistic quality.1

So not only do we love repetition in music, but also we get less enjoyment out of music that doesn’t involve repetition.

One reason why musical repetition is enjoyable and useful to a listener is that music cannot be summarized. If I say to you, “Steve just rang me to say his brother borrowed his motorbike without asking so he’s getting the bus. He says he’ll be here about an hour late. He sounded furious,” you know that the gist of this message is “Steve will be an hour late.” If you are a particularly supportive friend, you might extend the summary to “Steve will be an hour late and he’ll need cheering up.” But you’ll probably discard the information about Steve’s transport arrangements. You can’t do this summary thing with music. A tune has no gist. If you want to hear, or even remember, some part of a melody, you have to experience it at the speed at which it is normally played. Also, although you can start and finish your memory (or listening experience) wherever you choose in a piece, you can’t take shortcuts from one part of a tune to another without ruining the flow of the music.

Music flows through your brain in a stream you can’t easily think about while it’s happening. So it’s very useful if that great melody or bass line you’ve just heard is repeated a few times. After several repeats you start to understand how the music goes and begin to follow what’s going on. As psychologist Peter Kivy puts it, musical repeats “allow us to grope so that we can grasp.”²

Repetition allows us to concentrate on different aspects of the music and enjoy the experience more deeply. In the case of a song, we might relax our concentration away from the melody (which is repeating) and start listening to the words. In an instrumental piece, we might listen more carefully to the rhythm of the bass notes or the soulful timbre of the sax solo. Elizabeth Margulis suggests that repetition gives us pleasure as the result of “a growing sense of inhabiting the music.”3 There is a parallel here with how small children experience stories. Little kids like stories to be repeated—and they like the repetition to be as exact as possible. If you leave out a detail or give the rabbit the wrong sort of voice, you’ll find yourself being treated with withering contempt by a three-year-old. One reason children like the repetition is that they are processing a lot of information, and repetition allows them enough mental space to do so with pleasure. Almost any story for small children will include some words they haven’t used before. Psychologists Jessica Horst, Kelly Parsons, and Natasha Bryan have discovered that if children hear a new word as part of a repeated story, they start to use it much faster than if the same word is presented in lots of different tales.⁴ This is not an obvious result. You might have thought that hearing a new word in several different contexts would be the best way for a child to get a sense of when to use it. Instead it seems that simple repetition in a single context gives the child a solid base from which to begin using the new word.

Repetition and music are so interwoven that in some cases you can turn a non-musical sound into a musical one simply by repeating it a few times. For a great example of this, check out Diana Deutsch’s speech-to-song illusion (you can hear it on her website,^* or on YouTube if you search for “speech to song illusion”). Ironically, Professor Deutsch discovered the illusion by accident while she was editing the spoken part of her CD Musical Illusions and Paradoxes. At the time of the discovery she was editing out a hesitation in her delivery of this sentence:

The sounds, as they appear to you, are not only different from those that are really present, but they sometimes behave so strangely as to seem quite impossible.

In order to fine-tune the timing of the sentence, Diana was listening to the phrase “sometimes behave so strangely” on a repeating loop when suddenly the words started to behave strangely…

Although the phrase was spoken normally, after a few repeats it seemed to turn into a little song with a clear melody.

The recording on the website (and on YouTube) consists of the whole sentence followed by repeats of the “sometimes behave so strangely” part. If you listen to it you’ll hear the full sentence and the first couple of repetitions of the shorter phrase as normal speech. But after a few repeats, for most people the phrase will start to sound like singing. The really weird thing is that if you go back to the beginning of the recording and listen to the full sentence again, it all sounds like perfectly normal speech except for “sometimes behave so strangely,” which sounds like part of a song, at least to the vast majority of us (about 85 percent of listeners). So some normal speech can be turned into song simply as a result of repetition. Professor Deutsch was so impressed with the strength of the effect that she put it on her next CD of aural conundrums, Phantom Words and Other Curiosities. (Those of you who are interested in how fallible our hearing systems are will find these CDs mind-boggling.)

Further investigation of the phenomenon confirmed that the pitch jumps of the spoken version were a long way off the pitch jumps of the song it apparently produces.5 The spoken version does go up and down in pitch, so it has a contour like a tune, but the pitch jumps are not as big as those in the tune that appears after repetition.

During Diana’s investigation, people were asked to imitate the phrase after they had heard it only once, when they were still experiencing it as normal speech. When they did so, they followed the pitch jumps of the normal speech version fairly accurately—as you’d expect. Later, after they’d heard several repeats of the spoken phrase and it had turned into a song in their heads, they were again asked to imitate it. This time nearly all of them followed the same contour of up-and-down jumps in pitch, but they increased the sizes of the jumps until they matched real musical intervals: in other words, they were singing. Not only were they singing, but they all sang the same tune. Their brains had taken the real information (the moderate ups and downs of speech) and changed it, magnifying the jumps until they matched musical intervals. And all of this happened simply because the phrase had been repeated.

A much simpler example of repetition transforming sounds into music happens every day, whenever we hear a car’s turn signal, or a clock. Most turn indicators and clocks make the same ticking noise all the time—they go tick, tick, tick, tick. But for reasons that no one has figured out yet, we experience this ticking as a repeating cycle of tick tock, tick tock. Try listening to a clock or watch tick. You will probably hear this tick tock, tick tock pattern, with the tock seeming a little lower in pitch, or quieter, or louder than the tick. Now concentrate and try to group the sounds in threes rather than twos—tick tock tick, tick tock tick. Say this pattern of ticks and tocks aloud (in time with clock) to get yourself into the swing of it, then stop saying it and you will probably continue to hear the new pattern. If you manage this, you will notice that you’ve turned some of your tocks into ticks, and indeed some of your ticks have transmogrified into tocks. This proves that the ticking sounds must all be identical, and you can now, if you like, make yourself hear what’s actually happening—tick tick tick tick tick. But don’t get too proud of your newfound skill. This clarity will quickly collapse and you’ll soon be on the slippery slope to ticktockitude again. It seems that our brain finds it easier to deal with similar repeating sounds if it divides them into little groups.

A team of French psychologists working at the University of Burgundy has confirmed this. They wired people up to EEG machines and observed their brain activity as they listened to a series of identical tones. They found that the brain automatically organized the tones into pairs and experienced more activity on the first tone in each pair.6 The sounds were identical, but listeners were subconsciously accenting them into the equivalent of TICK tock TICK tock.

A large number of psychological tests have shown that we experience our most intense emotional responses from familiar music, and repetition is of course a great aid to familiarity, whether it’s repetition of the whole piece, the chorus of a song, or even a single note—as in the beginning of “Flash” by Queen.

Even if you don’t realize that something in the music is being repeated, it still casts its “repeat” spell over you. The music theorists I mentioned earlier may not have realized that the repeats were increasing their pleasure, and most of us don’t listen to music with some sort of conscious repeat barometer, so we don’t generally notice all the repetition going on in the music we listen to. In some cases only the harmony is repeating, or the bass line—but whatever it is, it helps to draw us into the music.

The subconscious aspect of all this is very interesting. Back in the 1960s the psychologist Robert Zajonc identified a phenomenon he called the “mere exposure effect.”⁷ Zajonc’s investigations showed that we have more positive reactions to things we have seen or heard before, even if we don’t realize that we have experienced them previously. This phenomenon sounds similar to the old adage “I like what I know and I know what I like,” but it’s deeper and weirder than that.

In one experiment, Zajonc showed non-Chinese students twelve meaningless squiggles like these, which look like Chinese characters:

The students were divided into several groups, and although each group was shown the same twelve images, they weren’t shown them the same number of times. During the course of the test some images were shown once, some twice, some five times, some ten times, and some appeared twenty-five times. But the images were shuffled around so an image that had been seen ten times by one group was seen five times by the next and twenty-five times by another, and so on. The images were shown to the students for only about two seconds each time, so they had very little chance of remembering any of them.

Later, the students were told that the images were Chinese ideographs for various adjectives, some pleasant (perhaps “beautiful” or “sweet”), and some unpleasant (perhaps “rotten” or “difficult”). They were then asked to rate them just by looking at them—to assign each one a “goodness” score, from zero (if they thought the ideograph might mean something really bad) up to a maximum of six (if they thought it might mean something very pleasant). The students probably thought they were rating the images on the basis of their appearance (“this one looks spiky and aggressive so I’ll give it a 1,” etc.), but what they actually did was rate the ideographs they had seen more often as the most pleasant.

Although the students hadn’t seen the images often enough or for long enough to remember them, in each group the images that had been shown most frequently were rated as having the most positive meaning, and vice versa. Each group of students had different “most positive” and “most negative” symbols because the frequency of appearance of the different images had been changed from group to group. The overall effect was quite clear and has been reproduced many times since in lots of different contexts. Generally, for images and sounds, we respond more positively to what we have been exposed to before and more negatively to the unfamiliar.

The really surprising thing about the mere-exposure effect is that it doesn’t seem to have anything to do with conscious thinking or choice-making. It’s an unconscious process. The effect has even been shown to work on images that are flashed for only four one thousandths of a second, too fast for the brain to consciously “see” the image at all.

The unconscious mere-exposure effect is probably one of the reasons we enjoy repetition in music, even when we aren’t conscious of the repetition. The effect also works to some extent when we hear variations on a theme, as opposed to verbatim repeats. This makes sense, given that a lot of repetition in music involves slight changes of rhythm, pitch, or instrumentation.

So repetition gives us a baseline of pleasure. But for the truly beautiful musical moments, we need the repetition to be spiced with surprise.

Surprises in music

We don’t usually pay attention to everything that’s going on around us. Attention requires energy, and as biological systems, we like to minimize the energy we use. On the other hand, we need to pay attention whenever something important or interesting is going on. In bygone days “something important” might have been the sudden appearance of a wolf pack, or a thunderstorm. Nowadays it’s more likely to be the sudden approach of dessert, or the realization in a boring meeting that someone just mentioned your name. When this sort of thing happens, both your arousal and attention systems kick in together. The arousal system gets your body prepared to deal with the situation by increasing your pulse, glucose uptake, and breathing rates (among other things), and the attention system focuses your brain.8 This response kicks in whenever the situation we’re in starts to feel unpredictable. We have an in-built capacity to judge whether or not the immediate future is likely to be dangerous, and, if everything looks predictable and calm, we switch off our energy-consuming attention and arousal systems. Whenever something surprising happens, however, our attention and arousal states rise very quickly—and initially, we always expect the worst. As David Huron puts it in his excellent book Sweet Anticipation:

Since surprise represents a biological failure to anticipate the future, all surprises are initially assessed as threatening or dangerous.⁹

But surely some surprises are pleasant. An unexpected gift? A phone call from a good friend?

Yes, thankfully there are pleasant as well as unpleasant surprises—but the nice ones are judged to be pleasant only after an initial “threat” response, which is usually so short-lived that it doesn’t register with your consciousness.

Imagine you’re a child reading in bed. The bedroom door clicks open. It’s your mother coming to kiss you good night. The click comes as a surprise, even though your mother comes in every evening at about this time. For the first one sixth of a second or so your mind is running around ringing alarm bells and insisting that the wolves are just around the corner. And then, before you’ve even noticed that you are thinking negative thoughts, you realize that the whole click/kiss thing is a positive event. And this is not just a childish response. Twenty-five years later you’re sitting in bed reading the Sunday papers and the bedroom door clicks—and it’s your partner with a tray of tea and toast. Once again, for the first sixth of a second it’s all wolves, earthquakes, and Viking raiders, until these negative unconscious responses are washed away by a proper realization of what’s going on.

At the moment the click happens, an immediate “action stations” signal is sent off to the fear center of the brain (the amygdala). At the same time, a “what’s going on?” message is sent off that reaches the amygdala about one sixth of a second later because, on the way, it passes through the main section of the brain, where decisions are made about what’s happening: Does the fear signal need to be escalated or should it be turned off?

No matter how often the door clicks, we always have an initial negative response that is then either backed up by a conscious negative response (This really is an earthquake!) or swept away by a positive one (Ah! Tea and toast!). The reason for our inability to learn that a bedroom door clicking in Manchester is unlikely to be a danger signal is that the human race would never have survived if it could un-learn the fight-or-flight response to unexpected sounds (or unexpected touches or tastes or visual events). We have to put up with having brains that always overreact initially because the alternative would mean underreacting on the rare occasions when something dangerous really is going on. Our over-responsive brain might make you a little too jumpy, but an under-responsive brain would eventually kill you.

The fact that our brain responds subconsciously to any unexpected change in what we are listening to has been confirmed by (EEG) brain sensors. For example, even the brains of four-month-old infants give off a surprise signal if they are listening to two alternating notes and the pattern is changed. (If the notes are going ping pong ping pong ping pong, the babies register surprise if they hear two pings next to each other.)10 Tests like this on adults and infants have shown that we are alerted by any oddity in a pattern of sounds. Unexpected changes in pitch, timbre, loudness, and even the direction the sound is coming from surprise us to some extent. And the fact that we are equally surprised by increases or reductions in loudness (or even sudden silences) shows that it’s the change in pattern which is important. A pattern of sounds makes you expect a continuation of that pattern, and if your expectations are violated, your brain starts paying attention. Obviously, small deviations generate minor responses and big pattern changes are more startling.

This surprise response is central to our enjoyment of music. When we listen to a familiar type of music, we are presented with a stream of sounds that are generally fairly predictable and only occasionally surprising. The surprises can be based on note choice or note timing (among other things), in either the melody or the accompaniment. When the music is predictable, we get a warm sense of satisfaction from understanding what’s going on—which is one source of musical pleasure. If something surprising happens, we always experience an initial negative response that will almost always be followed by a wash of relief that the surprise was merely musical. This negative-positive change magnifies the positive effect you feel. The magnification effect can be illustrated by the following analogy.

Every month you and your family visit your grandmother. Understanding what it’s like to be a perpetually broke teenager, your grandma has started to secretly slip a $20 bill into your inside jacket pocket before she hands it back to you as you leave. After one visit you feel for the $20, and it’s not there. Damn! Grandma must have forgotten.

But five minutes later you find that she’s put the money into a different pocket—and it’s not $20, it’s $40. Excellent! In this case your expectation was $20, and simply finding $40 in the usual pocket would have been great. But finding $40 after thinking you’d “lost” $20 feels even better. The initial negative has amplified the eventual positive.

The subconscious relief you feel as your brain kicks in to calm down its fear center is one reason why we feel pleasure—sometimes extreme pleasure—when something unexpected happens in a piece of music. If you are listening to a discordant, jarring section, you will be predicting more of the same in the near future. But if the music surprises you by turning all tuneful and agreeable, the pleasant effect will be amplified by the surprise involved.11 The fact that the initial negative response is biological and can’t be turned off is the reason why a certain passage of music can have the same “goose-bumpy” effect even when we know it well. Another contributory factor to our pleasure is the expectation that the marvelous goose-bumpy bit is just about to happen.^*

The tingle factor

Sometimes your emotional reaction to music can produce a strong physical reaction. Professor John Sloboda has identified three main physical reactions to music—which can be experienced singly or in combination.¹² They are:

1. A “lump in the throat” feeling, sometimes accompanied by tears (often these are tears of happiness or relief).

2. A tingling skin sensation—goose bumps as the hair on your skin stands on end—sometimes accompanied by shivers down the spine.

3. An increased heart rate, sometimes combined with a sinking feeling in your abdomen.

This level of emotional engagement happens only occasionally, and it’s more likely to happen with music you are familiar with and when you are listening carefully—not just listening as you fill the dishwasher. Although it’s difficult to pin down what causes this type of thrill reaction, there is evidence that it’s often linked to an unusual change in the music—a violation of your expectations of what this sort of music would normally do at that point. Obviously if you are familiar with the piece of music involved, you know that the unusual change is coming, but it still has the capacity to thrill you as long as you’re in the right mood. Of course, some of us are more prone to emotional musical thrills than others. Looking into this, several psychological studies have suggested that people with the personality trait “openness to experience”* are more likely than others to feel intense emotions as a result of listening to music.¹³

Although there are no firm and fast rules (as usual with music), the three types of physical/emotional reaction have been linked to the following musical techniques.¹⁴

The “lump in the throat” feeling with occasional tearfulness can be caused by various musical effects including something musicians call a “melodic appoggiatura.” In plain language, this involves singing or playing a melody note that doesn’t fit with the accompanying harmony, and then shifting up or down a step in pitch to a much more suitable note. This release of tension is the musical equivalent of going “phew” after almost missing your train.

Shivers and goose pimples are usually caused by changes in the harmony rather than the melody. The music may be tootling along normally and then, without warning, the composer starts using different notes for the chords that provide the harmony. It’s a bit like watching an outdoor sporting event through orange-tinted sunglasses and suddenly changing to green-tinted ones. A lot of what’s going on hasn’t changed—but something fundamental is different.

Increases in heart rate are, as you might expect, linked to the timing of the music. For example, your heart rate can increase if the music you’re listening to goes from a steady, straightforward rhythm to a syncopated one. “Syncopated” just means that the rhythm puts emphasis in unusual places. Instead of the usual:

One and Two and Three and Four and…

we might get, for example:

One and Two and Three and Four and…

Another way for the composer to accelerate your heart rate is to build up your expectations for a climax and then provide it slightly early.

You might have noticed that all of these big emotional effects are based on building up expectations and then violating them (but not too much), a principle that also works for storytelling and jokes. We like things to take a path that makes sense, but we don’t want it to be too predictable. One big difference between music and storytelling or jokes (as far as adults are concerned) is that repetition improves our attachment to a piece of music. We generally experience the strongest emotions from music we are familiar with.15 In the case of jokes and stories, although we all have our favorite film quotes and snippets from TV sitcoms, we usually get the maximum impact the first time we hear them.

Not very James Bond, are we? We like musical surprises—but only if they’re not too surprising—and we prefer surprises we’ve heard before.