An Alarm pattern is when you create sounds that mimic the vocalizations humans and animals produce when they are afraid. These fearful sounds have consistent, distinctive auditory cues: non-linearities (distortion) resulting from vocal strain, high amplitude modulation rates (acoustic roughness), arcing sweeping pitch contours, and high fundamental frequencies. The more abruptly you introduce alarm cues into your music, and the more you intensify the acoustic properties that distinguish them, the more likely audiences are to experience chills

The Alarm pattern can trigger a frisson-inducing fear response by creating stimuli that activate our amygdala – a brain region that helps us detect and respond to threats – and fit with a “fear” schema – a stereotype learned from past experience – in our memory.

When we hear alarm cues, our brain puts them on a “fast track” relative to other sensory information. The amygdala quickly prompts the release of chemicals in our brain (transmitters like dopamine, serotonin, and norepinephrine) and body (hormones such as adrenalin and cortisol) that signal something important is happening and prompt us to search our environment for an explanation of our sudden, heightened arousal. The meaning of an aggression cue is added by schemas in our memory that help us more efficiently respond to stimuli we have previously associated with danger. Recent research confirms also that humans can reliably distinguish aggression cues from similar auditory cues denoting fear or other forms of heightened arousal.

It’s theorized that alarm cues evolved to accomplish three objectives. First, to alert any nearby friends and family that a threat is close and make it as difficult as possible for any listeners to ignore or habituate to our signal. Second, to induce fear in any nearby listeners in order to prepare them to respond to the threat (when they hopefully come to our aid). And third, to alert predators that we have been spotted them.

A common element across alarm cues is that they produce unusual acoustics. Whereas normal speech has amplitude modulation rates between 4 and 5 Hz, for example, screams modulate at a rate between 30 and 150 Hz. To ensure that alarm cues are honest signals and avoid the boy-who-cried-wolf problem, our evolutionary ancestors developed atypical, unnatural, and difficult to fake sounds. Recent research confirms that alarm cues are so specific that we can reliably distinguish them from similar cues denoting aggression or other forms of heightened arousal.

Our data indicates than when you create high-pitch sounds with unusual frequency features called non-linearities (chaotic broadband energy, sidebands, and warbles, all between and outside the normal harmonics of a voice or instrument), this can often induce frisson in listeners. You know non-linearities when you hear them; these are harsh, hard-to-ignore sounds. This method is effective because non-linearities are a key component of human and animal alarm calls; we strongly associate them with fear. Given that it takes significant energy to strain a sound system and produce spectral non-linearities, our brain generally trusts that non-linearities are an honest signal of fear and danger.

In chills-inducing passages, some of the most reliable techniques we see composers and performers using to achieve this method include:

• Overblown or strained sound sources, including atonal screams and shrieks by talented lead vocalists, trumpet squeals, screeching strings, and distorted electric guitar
• Clashing sound sources, or when multiple instruments combine to sound dissonant chords, often doubled and with added reverb to create a very unstable sound
• Timbres that produce inharmonic noise, especially waterphone, theremin, whistles, metallic percussion (gongs, cymbals, anvil, chimes, etc.), and guitar-amp feedback

Don’t interpret this to mean that any vocal scream will give listeners chills. Our data suggests that when using this technique, it helps to:

• Place screams at points when listeners least expect them, for example at the end of a cadence or climax of a phrase when listeners are distracted by the melodic movement
• Minimize orchestration and any other melodic or harmonic movement during the scream to make it as conspicuous as possible for listeners
• Sustain the scream to re-assure listeners it is intentional and then re-start the melodic flow to create space for a positive listener appraisal response
• It’s easy for these sounds to quickly become unpleasant and disruptive if not handled properly.

Anecdote: Biologist Daniel Blumstein, an expert on animal alarm signals, collaborated with film composers to study the effects of added non-linearities on listeners. Blumstein not only found that film soundtracks already heavily incorporate non-linearities, but even more importantly that adding more non-linearities increases and intensifies audience emotional response.

Our data indicates than when you create high-pitch sounds with unusually rapid, non-linear amplitude variations referred to as acoustic roughness (a measure of how rapidly a sound’s loudness fluctuates that we perceive as beating or rattling), this can often induce frisson in listeners. Roughness is like a strobe light effect for sound, producing a pulsing effect that can vary in rate and height. This method is effective for frisson because rapid, non-linear amplitude modulation is a key component of human and animal alarm calls. We already heavily associate it with fear; it’s the distinguishing characteristic of screams and alarm clocks. Given that high-roughness sounds are unusual and difficult to produce, they are highly attention-grabbing.

In chills-inducing passages, some of the most reliable techniques we see composers and performers using to achieve this method include:

• Tonal screams by talented lead vocalists, flutter-tongue on winds, tremolo string bowing, vocal ululation, instrumental trills, and synthetic roughness
• Narrow dissonant harmonic intervals (e.g., minor seconds) whose tones are close enough in frequency to interfere and creating a pronounced rattling
• Steady state sound sources such as singing voice, bowed stings, brass, and winds, which result in more salient roughness than impulse sources like percussion and plucked strings
• Ratchet

Don’t interpret this to mean that any random scream with heavy distortion will automatically give listeners chills. Our data suggests that when using this technique, it helps to:

• Place dissonances at moments when listeners are distracted (e.g. at the end of a cadence, on the upbeat, etc.) to make them as jarring as possible
• Continue the consonant, melodic flow over the top of the dissonances so that they are not overly disruptive for listeners
• Sustain or repeat the dissonant intervals, and leave a rest after, to re-assure listeners they are intentional and create space for a positive listener appraisal response

Our data indicates that when you feature siren-like acoustics – rough sounds with a gliding, arcing pitch contour and pronounced peak – this can often induce frisson in listeners. This method is likely effective for two reasons. First, arced pitch contours paired with acoustic roughness produce significant non-linearities. Researchers have recently found that arced pitch contours are one of the key defining characteristics of effective screams. Second, arced pitch contours trigger learned associations with alarm due to our repeated exposure to ambulances, storm alerts, and other sirens. Together, this makes siren-like sounds highly effective for frisson.

Note: This method is always combined with one or multiple of Techniques 1-3 above.

In chills-inducing passages, some of the most reliable techniques we see composers and performers using to achieve this method include:

• Single, slow arced sounds that glide up and down over several seconds (asymmetric)
• Rapid, repeated arched sounds, typically one or more per second
• Recordings of real-life sirens (e.g. ambulances, storm alarms, etc.)

Don’t interpret this to mean that any high-pitched note with a gliding contour will automatically give listeners chills. Our data suggests that when using this technique, it helps to:

• Introduce siren sounds on instruments that differ dramatically from the preceding timbre palette of a song to make them as jarring as possible
• Use reduced texture and increased dynamics to focus listeners on the sound of the siren and prevent non-acoustic distractions
• Prolong or repeat the siren to re-assure listeners it is intentional and create space for a positive listener appraisal response

Anecdote: The trailer for the film Star Wars: Rogue One was a frequent submission to our dataset of popular listener frisson moments. The video features a prominent, repeated siren throughout the final 30 seconds of the trailer. Sure enough, the top comment on the Youtube video, with 1500 upvotes, is: “That siren is bone-chilling”.

UPDATE

Our data indicates than when you create high-pitch sounds with unusually rapid, non-linear amplitude variations referred to as acoustic roughness (a measure of how rapidly a sound’s loudness fluctuates that we perceive as beating or rattling), this can often induce frisson in listeners. Roughness is like a strobe light effect for sound, producing a pulsing effect that can vary in rate and height. This method is effective for frisson because rapid, non-linear amplitude modulation is a key component of human and animal alarm calls. We already heavily associate it with fear; it’s the distinguishing characteristic of screams and alarm clocks. Given that high-roughness sounds are unusual and difficult to produce, they are highly attention-grabbing.

In chills-inducing passages, some of the most reliable techniques we see composers and performers using to achieve this method include:

• Tonal screams by talented lead vocalists, flutter-tongue on winds, tremolo string bowing, vocal ululation, instrumental trills, and synthetic roughness
• Narrow dissonant harmonic intervals (e.g., minor seconds) whose tones are close enough in frequency to interfere and creating a pronounced rattling
• Steady state sound sources such as singing voice, bowed stings, brass, and winds, which result in more salient roughness than impulse sources like percussion and plucked strings
• Ratchet

Don’t interpret this to mean that any random scream with heavy distortion will automatically give listeners chills. Our data suggests that when using this technique, it helps to:

• Place dissonances at moments when listeners are distracted (e.g. at the end of a cadence, on the upbeat, etc.) to make them as jarring as possible
• Continue the consonant, melodic flow over the top of the dissonances so that they are not overly disruptive for listeners
• Sustain or repeat the dissonant intervals, and leave a rest after, to re-assure listeners they are intentional and create space for a positive listener appraisal response

Our data indicates that when you feature pitch jumps to high fundamental frequencies with acoustic energy concentrated in a certain range, this can often induce frisson in listeners. Such leaps are likely effective for two reasons. The first is that they concentrate acoustic energy in the part of the spectrum to which we are most sensitive (1-4 kHz and especially 2-3 kHz range that mirrors the resonant cavity of our ear). We experience sounds in this range as especially piercing and gripping. The second is that a rapid change in fundamental frequency is a drastic non-linear sound that, in and of itself, is highly attention-grabbing. Together, this makes these frequency leaps effective for frisson.

Note: This method is always combined with one or multiple of Techniques 1-3 above.

In chills-inducing passages, some of the most reliable techniques we see composers and performers using to achieve this method include:

• Vocal leaps with accentuated formants, especially the squillo technique in opera singing that brings out the “singers formant” at 3 kHz
• Leaps on “bright” instruments that bring out upper frequencies (e.g. trumpet, electric guitar, synths)

Don’t interpret this to mean that any sudden high-pitched note will automatically give listeners chills. Our data suggests that when using this technique, it helps to:

• Place these notes at the beginning of a section or end of a cadence to make them as conspicuous as possible for listeners
• Precede leaps with subtle, rubato slow-downs or pauses and change the texture or rhythm after the leap to make it as jarring as possible for listeners
• Hold the high note to create space for a positive listener appraisal response