I See You Shiver — The Neurological Basis of Frisson (2017)

Abstract:

Many people experience emotional and physical changes in response to musical stimuli. This can include musically induced laughter, awe, and chills. The term “frisson,” handily borrowed from French, is applied to these chills (or, memorably, “skin orgasms” (Panksepp, 1995)) in response to music. This paper aims to discuss multiple suggested neurobiological and evolutionary roots of this phenomenon. Note that there are many competing theories about frisson — sources vehemently differ about the certain aspects of the phenomenon, especially the origin. Because of this I have had to focus more on certain views more than the others in this paper, while attempting to cover all of them.

Introduction:

Have you ever been listening to a piece of music and had a tingle run up your spine? Or felt goosebumps when you reached a certain “emotional climax” of a song? You are experiencing “frisson”, a term coined by John Sloboda in 1991. The term comes from French meaning “aesthetic chills.” The effect often manifests itself in the physical appearance of piloerection (goosebumps). There isn’t a huge amount of scientific understanding about the phenomenon–science is still trying to catch up (through some guesswork) about how and why music invokes a physical reaction often associated with cold or fear.

Two important aspects to understand even before music is introduced are piloerection and the “theory of expectation”, both of which undeniably have large amounts to do with the effect of frisson (Goldstein, 1980; Panksepp, 1995; Sloboda 1991, Huron, 2006).

Goosebumps are remnants of a fear response that evolved when humans were much hairier. Literature has agreed that animals choose from fight, flight or freeze when presented with a stressor. Goosebumps comes from the fight response. When this response is chosen the first step is to present an aggressive display. If a threatening enough display is shown it’s possible that the animal won’t have to fight, as its opponent will back down. This could include baring teeth, growling or eye contact as well as efforts to make the creature appear bigger such as standing up on hind legs or bristling hairs. This is what we see in cats and dogs when they arch their backs and raise their hackles. Now that humans have lost their hair, the only remnants we have of this effect is piloerection — the raising of empty hair follicles.

Goosebumps also occur when we are cold though. The James-Lange Theory that suggests that the physiological response precedes the feeling could explain this; “fear doesn’t make us cold … fear causes piloerection and this response evokes the feeling of coldness.” (Huron, 2006). There is data to show that people are more likely to experience frisson when cold though, taking advantage of this connection.

The other important area to understand before we add music to the equation is a general theory of expectation. David Huron’s book Sweet Anticipation, provides a clear visual explanation. He uses the abbreviation ITPRA to describe the model: Imagination, Tension, Prediction, Reaction, and Appraisal. The process starts as the mind begins to imagine different possibilities. As the event comes nearer, tension builds as physiological arousal increases in preparation. Then the event happens and immediately our system compares the outcome to our predictions, as well as a very fast cursory reaction that is processed mainly through the amygdala. Finally, the frontal lobe has time to give an appraisal of the event, given the entire context. This final stage is called valence.

Manifestation of the phenomenon and its measurement

As it is an emotionally-based response, frisson can be tricky to measure empirically. It also presents a challenge in that not everyone feels it: studies have shown anywhere between only 55% (Grewe, 2007) and 86% (Panksepp, 1995) actually experience this effect — an issue brought up later in the paper. Most studies have used the Galvanic Skin Response (hitherto referred to as GSR) as an acceptable measure of frisson and emotional involvement (Colver, El-Alayli, 2015, Rickard 2004, Steinbeis, Koelsch & Sloboda, 2006).

Evolutionary Basis — Exploitation of the Fear Response

We have already been over some basics of how the goosebumps phenomenon works, but not in terms of sound and music. Most literature agrees that frisson comes from the exploitation of the fear response and violation of expectation (Panksepp, 1995, Huron, 2006). Surprise, evolutionarily, is undeniably bad. It means that the body has not successfully prepared for the event to come. So when music evokes chills, laughter or awe, some research suggests that this is in fact exploiting this fear response. What causes the experience to be enjoyable in the end is the valence. Our fast-track brain run by the amygdala always assumes the worst, but when the positive, harmless valence comes through the frontal cortex (“it’s just music it’s not going to cause me bodily harm”), we take enjoyment from the scare (from the limbic contrast). “In effect, when music evokes one of these strong emotions, the brain is simply realizing that the situation is much better than first impressions suggest” (Huron, 2006). Especially because this taps into our evolutionary survival instinct, it works every time–our brain never lowers its guard.

Measuring Musical Expectation

The first step to understanding “violation of expectation” is to quantify what is expected. To this end, there have been many studies done over the last forty years about how to empirically measure expectation. The first study of measuring expectation was by Greenberg and Larkin in 1968. It showed that people were able to hear a sound more reliably amongst constant loud noise if they were expecting it either in time or frequency (“accurate expectation facilitates perception” (Greenberg, 1968)).

The most famous experiment for measuring musical expectation was the Shepard/Krumhansl probe-tone method in 1979. This had listeners judge the “fit” of different probe-tones given a constant melodic/harmonic context. A revision of this technique (that was less tedious) gave each participant a ‘grub stake’ of poker chips to bet on which note would come next. Correct bets won tenfold, incorrect bets lost completely. This also helped participants consider the likelihood of all possibilities, rather than just the first to come to mind.

Other experiments to measure musical expectation used reactions to quantify expectation. These included head-turning (orienting response), bradycardic response (heart rate) (Guhn, 2007), or reaction time (the ability to process a sound you expect more rapidly). In each case, the experiment relies on dishabituation (adjusting to consistent stimuli and measuring the physiological effect of suddenly changing the stimulus)

Of course all of these experiments are affected by conditional probability; it’s important to remember that contextual size/probability order is independent of the contextual distance. That is, sometimes a single action proceeds a single action in the far future (someone getting a work bonus for overtime and going on holiday), sometimes lots of small events lead up to a single immediate event (individual BINGO numbers until the overall win) and all the other iteration in between (Huron, 2006). As we will see, music exploits each one of these types of these conditions and each affects our reaction.

Types of Surprise

There are also three different types of musical surprise that are important to frisson: schematic surprise, dynamic surprise, veridical surprise and garden path surprise.

In schematic surprise the music is composed so that it will violate a general schema that the listener has. Musical examples of this would be deceptive cadences or chromatic mediant chords. Deceptive cadences are schematic because a normal V-I cadence is expected in the schema for all genres of music but the deceptive leads to an unexpected (and unstable) chord, the VI. Chromatic median chords are chords whose roots are a major or minor third apart, share triad qualities (both chords are major or minor), and share one common tone.

Dynamic surprise relates to unexpected changes in volume level. Veridical surprise has become more common; “since the advent of sound recording, it has become commonplace for listeners to acquire an intimate familiarity with particular recordings” (Huron, 2006). Using a different tempo, adding fermatas, changing keys or other musical changes from what we get used to can create veridical change.

The final kind of surprise is called “garden path surprise” — a term borrowed from linguistics. It references sentences where you gain information later in the sentence that invalidates your interpretation of the previous words. Examples include: “The old man the boats”, “The man who hunts ducks out on weekends”, and “The cotton clothing is usually made of grows in Mississippi” (Pinker, 1994). Musically, an example of this could be the beginning of Beethoven’s Piano Sonata op. 14 which starts with agogic stress on the third, sixth and ninth notes, suggested a 2/8 (essentially a hemiola) despite the piece being written in 3/8.

Common Themes In Frisson-Inducing Music

Now that we have discussed the various types of expectation/probability and different kinds of surprises, as well as understood the basis of physiological basis of “chills”, it’s time to introduce music. The seminal study for this was by John Sloboda from Keele University in 1991. In his study he distributed a questionnaire to 83 music-lovers, asking for information on their favourite emotional music. From this data, Sloboda analysed the musical traits of each trigger stimuli to try and find a connection. He found that frisson was most closely correlated with abrupt changes in harmony and dynamics;

Other explanations of and relations to frisson:

Since frisson is so related to emotion in music, understanding how it works has become the focus of many different areas of science including not only neuroscience and psychology, but also anthropology, ethnomusicology and even philosophy.

A recent paradigm-shifting study by Matthew Sachs at Cambridge University observed significant neural differences in the white matter between the sensory processing areas in the temporal gyrus and the emotional area in the insula between those that reported experiencing frisson and those that didn’t (Sachs, 2016). This was hugely important as it “provides the first evidence for a neural basis of individual differences in sensory access to the reward system, and suggests that social–emotional communication through the auditory channel may offer an evolutionary basis for music making as an aesthetically rewarding function in humans.” (Sachs, 2016).

There have also been connections inferred between the personality trait of “openness” and susceptibility to the experience of frisson (Colver & El-Alayi, 2015), which could explain another reason why some people experience frisson more readily than others. A more involved model suggests from Juslin (2013) an eight-pronged model of emotions including social, autobiographical, psychophysiological, and psychological factors. These eight “mechanisms” are (1) brainstem reflexes, (2) rhythmic entrainment, (3) evaluative conditioning, (4) contagion, (5) visual imagery, (6) episodic memory, (7) musical expectancy, and (8) aesthetic judgment, but these were not covered in my review.

Additionally, some ethnomusicologists have said that certain papers I quoted in this review “overly reductive” (Harrison, 2014) for not considering other cultures. Becker (2011) emphasizes that many cultures conceive of music as an integrative, full-body phenomenon, even to the point that many regional West African languages don’t have a word for music as a solely auditory experience (Agawu, 1995).

Future and Reflection:

The future research of frisson should focus more on broadening the context of musical stimuli used, but once we “acknowledge and respect individual differences in subjective experience” (Harrison, 2014) we may unlock a more general understanding of the social and cultural effects of music, instead of empirical, western focused case studies. Doing this could increase our understanding of how music became such an integral part of the human aesthetic/emotional experience.

References:

Agawu, V. K. (1995). African Rhythm: A Northern Ewe Perspective. Cambridge: Cambridge University Press. https://academiccommons.columbia.edu/catalog/ac:181269

Becker, J. (2004). Deep Listeners: Music, Emotion, and Trancing. Bloomington, IN: Indiana University Press.

Colver, Mitchell & El-Alayli, Amani, (2016) Getting aesthetic chills from music: The connection between openness to experience and frisson, SEMPRE, pp. 413–425.

http://journals.sagepub.com.proxy.lib.duke.edu/doi/abs/10.1177/0305735615572358

Grewe, O., Nagel, F., Kopiez, R., & Altenmüller, E. (2007). Listening to Music as a Re-Creative Process: Physiological, Psychological, and Psychoacoustical Correlates of Chills and Strong Emotions. Music Perception, 24(3), 297–314.

http://dx.doi.org/10.1525/mp.2007.24.3.297

Harrison, L & Loui, P (2014) Thrills, chills, frissons, and skin orgasms: toward an integrative model of transcendent psychophysiological experiences in music. Department of Psychology, Wesleyan University. https://doi.org/10.3389/fpsyg.2014.00790

Huron, David (2006) Sweet Anticipation: Music and the Psychology of Expectation, Boston, MA, MIT Press.

Juslin, P. N. (2013). From everyday emotions to aesthetic emotions: towards a unified theory of musical emotions. Physical Life Review 10, 235–266. http://www.sciencedirect.com/science/article/pii/S1571064513000638?via%3Dihub

Panksepp, Jaak. (1995). The emotional sources of “chills” induced by music. Music Perception: An Interdisciplinary Journal, Vol. 13 №2, Winter, 1995.

http://mp.ucpress.edu/content/13/2/171

Sachs, Matthew E., Ellis R. J., Schlaug G., Loui, P., (2016) Brain connectivity reflects human aesthetic responses to music. Social Cognitive and Affective Neuroscience, Volume 11, Issue 6, Pages 884–891. https://doi.org/10.1093/scan/nsw009

Sloboda J. A. (1991). Music structure and emotional response: some empirical findings. Psychology of Music Vol 19, Issue 2, pp. 110–120.

http://journals.sagepub.com/doi/pdf/10.1177/0305735691192002#articleCitationDownloadContainer

Martin Guhn, Alfons Hamm, Marcel Zentner (2007) Music Perception: An Interdisciplinary Journal, Vol. 24 №5 (pp. 473–484) http://mp.ucpress.edu/content/24/5/473