Don't you shiver?: Pleasurable Responses to Music and the Brain

by Isabel Acosta, 2007-49035

(If you love music, I really encourage you to read this seemingly intimidating article!)

Didn’t you just shiver after hearing that song? All its elements –the melody, the tone, the rhythm, the pitch, the singer’s voice, the buildup of instruments –all that abstract jumble of notes and sounds makes my heart soar every time I listen to that song. And apparently, my brain is going crazy as well.

“Music expresses that which cannot be said and on which it is impossible to be silent.” Such powerful words by Victor Hugo, a French poet, novelist and playwright. Music has always been a profound aspect of human life. It affects us in unexplainable and incomparable ways. Those ‘chills’ we feel when we listen to music, those ‘shivers-down-the-spine’ –those occurrences are universally felt and experienced by all of humanity. And two scientists sought to explore and explain the neural mechanisms behind that experience.

Blood and Zatorre (2001) sought to explore what people’s brains looked like while listening to ‘intensely pleasurable music.’ The participants of the study consisted of 10 students from McGill University (five female, and five male) with at least 8 years of music training. Although music training is not needed to experience ‘chills,’ musicians were selected as participants, because the researchers assumed that this group of people would be the exact kind of people to respond emotionally and strongly to music. An important qualifier was that all the participants reported frequent and reproducible experiences of chills. This quality is definitely needed for the experiment to be a success.

After the participants were properly oriented and given consent forms, Positron emission tomography (PET) scans were then performed and registered with Magnetic resonance imaging (MRI) scans. PET is a technique for visualizing brain activity, usually by measuring the accumulation of radioactive 2-deoxyglucose (2-DG) or radioactive water in the various areas of the brain. MRI is a procedure in which high-resolution images of the structures of the living brain are constructed from the measurement of waves that hydrogen atoms emit when they are activated by radio-frequency waves in a magnetic field. So, brain activity was registered with the MRI, so that the researchers can easily locate which part of the brain is activated during listening to intensely pleasurable music. Each participant chose a musical selection which consistently made them experience strong emotional responses (which includes chills). The researchers made the participants choose their own musical selection because music preference varies greatly between individuals, and because this would be the most efficient and direct way to produce the necessary intense emotional responses. All chosen musical selections belonged to the classical genre –there were no words associated with the piece, and they were all of intrinsic value to the participant. The latter assures that the selection does not evoke emotions due to personal associations and/or memories. 90-sec excerpts of the musical piece (which should include the part that specifically gives the person chills) were used as ‘subject-selected music’ for that participant. Each of the excerpts was used as another participant’s emotionally neutral control. This was done to control for confounding variables brought about by varying stimuli. This assures that the analysis of the group-averaged data involves the comparison of identical sets of stimuli (same tempo, same pitch, so no confounding variables --when the data from the subject-selected music condition and the data from the control condition were averaged separately and subtracted from each other during analysis, only the pure emotional valence remained). The participants rated the emotional intensity of their responses to the other nine musical selections of the other participants. A rating of less than or equal to 3 (≤3) out of a 10 was required in order for the selection to qualify as that rater’s control. The participants were then asked to familiarize themselves with their control musical selections in order to control for the confounding effects of novelty.

Each PET scan lasted for about 60 seconds. During these scans, the participant listened passively to one of four stimuli (subject-selected music, control music, and two baseline conditions: amplitude-matched noise and silence). These baseline conditions are needed in order for the researchers to establish the participant’s default state and in order to examine the participant’s full sensory experience. The scan began around 15 seconds after the stimulus was played in order to ensure the stabilization of the participant’s response to the stimulus. Each of the conditions were repeated thrice, and the order of the stimulus representation was pseudorandomized. Heart rate (HR), electromyogram (EMG, a measure of the electrical activity of muscles), respiratory depth (RESP), electrodermal response, and skin temperature were measured during the PET scans using a machine called a F1000 polygraph. After the scans, each participant rated their emotional reactions to each stimulus. Three items were assessed: chills intensity (0 to 10), emotional intensity (0 to 10) and unpleasant vs pleasant (-5 to +5). The pleasantness and unpleasantness scale was needed because the researchers’ previous study was about the brain parts correlated with unpleasant or mildly pleasant emotions elicited by musical dissonance (lacking in harmony of musical notes). The participants were asked to listen to a range of unpleasant musical selections (mildly unpleasant to very unpleasant) and observed the negative/unpleasant emotions evoked by that piece through PET scans. The researchers wish to see if the same neural circuits recruited by the mildly pleasant music/emotion would be recruited by the chills.

The researchers used regression maps to assess the significance of the relationships between chills intensity rating and cerebral blood flow (CBF, as indicated by the PET scans).

In case you fell asleep during this part in your Math class or you simply don’t remember, regression analysis describes an unchanging relationship between two phenomena. The line representing a simple linear regression is expressed through a basic equation: Y = a₀ + a₁ X. X is the independent variable, and Y is the dependent variable (or vice versa). A₀ is the y-intercept (the value of Y when X is zero) and a₁ is the slope of the line. The slope of the line characterizes the relationship between the two variables. Using two more equations, a line can be drawn. The typical distance between the line and all the points (sometimes called the “standard error”) indicates whether the regression analysis has captured a relationship that is strong or weak. The closer a line is to the data points, overall, the stronger the relationship. (Dizikes, 2010)

The researchers performed a number of corrective computations and subtraction analyses in order to control for confounding factors. Within-subject differences in familiarity was accounted for, control music conditions were subtracted from subject-selected music conditions, baseline conditions were subtracted from music conditions to confirm that the CBF decreases/increases in the brain were from the baseline level and not merely differences from the subject-selected and control music conditions. Regression maps were also used to analyze the relationship between psychophysiological activity (HR, EMG, RESP, etcetera) and CBF. Afterwards, regression with chills were recalculated without the effects of these psychophysiological activities.

So what did these researchers find in those scans? They found that the subjects experienced chills during 77% of scans when their selected piece was played. Heart rate, respiratory depth, and electromyogram responses increased significantly during the highest rated chills music condition compared to the control condition. Skin temperature and electrodermal measurements did not differ significantly. No chills were reported during the control condition, and the two baseline conditions. The average rating for chills in the subject-selected music condition was 4.5 out of 10 (responses ranging from 1 to 9). Comparing this to the average rating for emotional intensity (7.4 out of 10) and pleasantness (4.4 out of 5), the researchers suppose that perhaps pleasantness and emotional intensity must reach a certain level before chills can be experienced. I think that this certainly makes a lot of sense. How can one feel that surging chill if pleasantness and emotional intensity are low and weak? The regression analysis correlating CBF and chills intensity ratings point out that:

a) with increasing chills intensity ratings, regional CBF increases (positive correlation) in the:

- Left ventral striatum (VStr)

o role in reward-related behavior, neurons in this region are sensitive to expectation of upcoming reward⁴

- Dorsomedial midbrain (Mb)

o relay station for auditory and visual information, it is also part of a dopaminergic pathway

- Bilateral insula (In)

o integrates sensory and autonomic information from the viscera, damage in this area leads to auditory agnosia or inability to recognize or differentiate between sounds

- Right orbitofrontal cortex (Of)

o receives information from sensory cortices and assigns them a reward value, related to motivational and emotional behavior⁵, and pleasure responses

- Thalamus (Th)

o sensory relay station

- Anterior cingulate cortex (AC)

o autonomic functions, such as regulating blood pressure and heart rate, as well as rational cognitive functions, such as reward anticipation, decision-making, empathy and emotion; role in reward-based decision making

- Supplementary motor area (SMA)

o Involved in planning of motor actions and bimanual control; also in retrieving sequences (and music is thought to be a kind of sequence)

- Bilateral cerebellum (Cb)

o processes input from other areas of the brain, spinal cord and sensory receptors to provide precise timing for coordinated, smooth movements of the skeletal muscular system; also proposed to be involved in some emotional functions such as regulating fear and pleasure responses⁶

b) with increasing chills intensity, regional CBF decreases (negative correlation) in the:

- Right amygdala (Am)

o processing and memory of emotional reactions

- Left hippocampus/amygdala (H/Am)

o functions in consolidation of new memories, emotions, navigation, and spatial orientation

- Ventral medial prefrontal cortex (VMPF)

o involved in processing of risk and fear, and in decision making, deals with emotion regulation

- Bilateral posterior neocortical regions, including occipital, parietal, and temporal cortices (widespread, particularly the cuneus/precuneus regions --region of the medial aspect of the occipital lobe of each cerebral hemisphere). The visual cortex (Vc) is highlighted in the image below.

Pictures a, b and c represent the scans of areas that show rCBF positively correlated with increasing chills intensity, while pictures d, e and f represent the scans of areas that show rCBF negatively correlated with increasing chills intensity.

The researchers also found that increases in psychophysiological activity correlated with increases in rCBF in that thalamus, anterior cingulate cortex, orbitofrontal cortex, insula, cerebellum and supplementary motor area. This basically affirms the relationship of chills and rCBF changes, since changes in autonomic activity and physiology are markers of chills.

What do these all mean? Changing activity was observed in some brain parts associated with the brain’s reward circuitry (left ventral striatum and dorsomedial midbrain, right amygdala, left hippocampus/amygdala and ventromedial prefrontal cortex). Activity was also observed in the paralimbic/emotional regions (bilateral insula, right orbitofrontal cortex) and regions associated with arousal (thalamus and anterior cingulate cortex) and motor processes (supplementary motor area and cerebellum). Apparently, this kind of brain activity is comparable to those observed in people experiencing euphoria, cocaine addiction (Th, In, AC, Am, and VMPF), and eating chocolate. Animal studies confirm that the structures activated in this study (such as VStr, Mb, Am, H, and VMPF) are crucial in the circuitry of reward processes, hedonic (pleasant sensations) impact, reward learning and motivation. Dopaminergic activity is crucial in this kind reward circuitry. Just so that you guys can understand and appreciate these results more, here is a short explanation of how the reward circuit works:

“The natural function of the reward circuit is to provide a reward and associated pleasurable feelings in response to life sustaining functions, such as eating, to encourage repetition of that function. The reward circuit functions through the use of neurotransmitters, mainly dopamine.” When the reward circuit is activated, it triggers a release of dopamine, which induces pleasurable and euphoric feelings or highs in the ventral tegmental area (VTA). Addictive drugs work the same way, they activate the circuit and cause the system to release more than normal levels of dopamine. Some drugs even block dopamine receptors, which are needed in order to control the dopamine available in the synapse. Blocking of receptors results in a build-up of dopamine. Addiction results when the brain gets used to the high levels of dopamine. The user feels the need to keep taking the drug in order to restore ‘normalcy.’ (Quraishi, 2002)

No wonder listening to music is such a rewarding and powerful experience!

The amygdala was of particular interest to the researchers. The amygdala and hippocampus interact with the midbrain in both reward and emotion systems. In cocaine administration studies, decreases in the left amygdala have been correlated with ratings of cravings rather than rush, so the researchers speculate that the decreases of rCBF in the amygdala in this study might be due to the anticipation of the chills rather than the chills response itself. Interesting right? In addition to this, amygdala decreases plus ventral striatum increases indicate that the “approach” and “withdrawal” system might be at work in this study. See, the amygdala is involved in the processing of fear, other aversive emotions and biologically relevant emotions, while the ventral striatum mediates processes associated with reward and motivation/approach behavior. Activity is both structures are also said to be negatively correlated (amygdala activity decreases as a consequence of increased ventral striatum activity) So, relating this to the study, perhaps music is pleasurable because it activates the reward system and because it simultaneously decreases brain activity structures that are associated with negative or aversive emotions. I find this really interesting! Isn’t it astounding how physiology explains and makes sense of something as abstract as music? Our brain really is such a beautiful aspect of humanity.

Lastly (phew, I know right, they found a lot of things in this study), remember that the researchers wanted to integrate this study with their previous study of music unpleasantness and pleasantness? Well, the researchers found that brain structures associated with intensely pleasant emotion elicited by music were different from the brains structures associated with unpleasant responses to musical dissonance (right parahippocampus, if you were curious). This is important because it suggests that negative emotions are specifically-related to this brain region. Structures that are involved in reward/motivation processing were not related to the structures activated by pleasant responses to musical consonance (harmonious songs), except for the ventromedial prefrontal cortex and the orbitofrontal cortex. This suggests that these brain parts probably respond to multiple emotions.

It's pretty obvious that I am really, really interested in this topic, and that the study really drew me in. I was really impressed with how the study was planned and conducted. The researchers really controlled for every possible confound, and such foresight is not present in a lot of researches. They handled such an abstract concept so intellectually and so efficiently, that it is no wonder why this study has been cited numerous times. I was really drawn into the essence of the study. I was really amazed at how physiology makes sense of an immense feeling I thought was too abstract and too grand to be explained. It is really astounding how physiology makes sense of music, and at the same time, the study of music sheds some light on the workings of the very complex human brain. There is logic and meaning in all life occurrences! The dynamics of it all makes me appreciate even deeper music and the beautiful plan or blueprint of the human body. I also think that is just interesting that listening to music evokes the same response as biologically relevant stimuli. Unlike eating and sexual intercourse, music has no adaptive or reproductive function, yet it activates and recruits the same neural systems of emotion and reward. The reward circuitry, in particular, exists to reinforce the repetition of actions that ensure our survival and self-preservation. Why then is there a need for constant and consistent music exposure? I do not know, but the researchers propose that appreciation of music may just be an emergent property of the complexity of human cognition. Perhaps as our primitive brain systems started to evolve into more complicated structures, human beings gained the capacity to assign meanings to abstract stimuli and derive pleasure from them. Whatever the reason is, this study only proves that music, despite lacking in survival and life-sustaining significance, is probably significant to our psychological well-being. Music must be contributory in this sense for it to affect our brain systems so immensely. I am sure that most of us, if not all of us, can agree that music affects us and benefits us in much deeper ways than even our most primary needs. Indeed, Nietzsche was right, “without music, life would be a mistake.”

Didn’t that article just make you shiver? Here’s another song to shake up your brain. Ooh and aah, feel your brain slowly lighting up. Enjoy that glorious pleasure, it’s good for you ;)

Sources:

1) Blood, A. J., & Zatorre, R. J. (2001). Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. (M. E. Raichle, Ed.) Proceedings of the National Academy of Sciences , 98 (20), 11818-11823.

2) Dizikes, P. (2010, March 16). Explained: Regression analysis. Retrieved July 29, 2010, from MIT News: http://web.mit.edu/newsoffice/2010/explained-reg-analysis-0316.html

3) Pinel, J. P. (2008). Biopsychology (7th Edition ed.). Jurong, Singapore: Pearson Education South Asia Pte. Ltd.

4) Ventral striatum plays important role in circuitry of reward-oriented behavior: Researchers. (2010, May 5). Retrieved July 29, 2010, from The Medical News : http://www.news-medical.net/news/20100505/Ventral-striatum-plays-important-role-in-circuitry-of-reward-oriented-behavior-Researchers.aspx

5) Cerebellum. (2010, July 15). Retrieved July 29, 2010, from Wikipedia: http://en.wikipedia.org/wiki/Cerebellum

6) Rolls, E. T. (2004). The Functions of the Orbitofrontal Cortex. Brain and Cognition , 55 (1), 11-29.

7) Quraishi, S. (2002, January 2). Addiction and Reward Circuit. Retrieved July 29, 2010, from Serendip: http://serendip.brynmawr.edu/bb/neuro/neuro01/web2/Quaraishi.html

8) Wikipedia.com