We all have a song that makes the hairs stand up on the back of our neck, sends shivers down our spine or transforms our mood. When we choose what we want to listen to, we could make a selection based on artist or genre, but we might also make a choice based on how we want to feel. For example, a track to energise you during a workout, or a track to calm you after a stressful day (mine is The Blue Bird by Stanford, the choral harmonies at one point always make my spine tingle – it’s my go to track after a challenging day).
Or you might choose music to set the mood for an event. Maybe a reflective piece for a memorial, or a joyous piece to celebrate a wedding? But how does singing make us happy or sad, and is this something we can analyse and measure? In this article we delve into some of the research behind music, mood and emotion.
Is there a difference between mood and emotion?
Emotions might be intensely experienced, for only a brief time. Basic evolutionary emotions such as anger, fear or joy play a major role in preparing us for actions that might ensure our physical or social wellbeing, or even survival. For example, fight or flight in response to fear. ‘Refined’ emotions are those that are felt rather than acted on, they don’t necessarily lead directly to action.
Where emotions are generally a response to something specific, moods tend to be more diffuse, and aren’t necessarily linked to a specific event, person or object. They are less intense but can last for longer, hours or days. A mood is a general feeling, rather than a reaction to a particular situation.
Music might stimulate an unexpected emotion (that spine tingling moment) but we also use music to set or change our mood. Being able to determine the mood evoked by a piece is hugely useful to brands like Spotify or Apple Music that want to be able to suggest music and playlists to help us work-out or relax. Given the huge volume of music available in the world, which is being added to every day, automated ways to classify music into mood categories are being developed using technology such as machine learning.
Measuring emotional response to music
Modern brain scanning techniques mean it is possible to examine how the brain responds to music as we are actively listening. Positron emission tomography has been used to study the neural mechanisms underlying emotional responses to music. In one study, each participant was asked to selected a piece of classical music without lyrics that consistently sent shivers down their spine, or gave them the ‘chills’. The emotional responses in this study were chosen as intrinsic to the music itself, rather than personal associations and/or memories. Memories linked to music can trigger a powerful emotional response, but this study was designed to focus purely on the brain response to the music itself.
When a participant’s brain response was monitored, whilst listening to their chosen piece, it was found that the same parts of the brain were stimulated that also respond to other sources of euphoria such as food or sex. Although music may not be imperative for survival of the human species, the existence of this response suggests it has intrinsic value, not least to our mental and physical well-being.
Music and emotions are personal
An important feature of the study above was that each participant chose their own specific piece of music. What triggers an intense emotional response in one person, might not have the same effect for another. Musical preference is highly individual – a track can only relax you if you enjoy listening to it (not easy if you really don’t like the genre or artist). Our personal preference is unique to us; based on our experiences, the musical heritage we grew up with, memories associated with when we first heard the piece and our individual taste.
This makes how we experience emotions hard to quantify, making studies in this area very challenging. We instinctively know that music can stimulate an emotional response and change our mood, but how do we go about classifying and categorising it, so that we can study and understand it?
Mapping music by emotion
Researchers often use two terms to map emotions:
- Valence – a measure of whether a felt emotion is negative or positive.
- Arousal – the intensity to which the emotion is experienced.
For example, feelings of joy are positive – joy has a positive affect on our emotional state, whilst anger is a negative emotion, it has negative emotional valence. Anger is also a high-arousal emotion, whereas peacefulness is low-arousal. The reality of emotions is very complex, but categorising them is helpful in order to study them.
Valence and arousal can be mapped visually. Valence mapped on one dimension of a continuum from negative or positive, whilst arousal is mapped on another from high to low. Showing these dimensions as axes on a simple graph reveals quadrants (see example below). For example, joy is typically characterised by positive valence and relatively high arousal, whereas sadness or depression is typically characterized by negative valence and relatively low arousal. Anger is usually characterised as negative valence and high arousal.
A piece of music can be mapped onto these axes depending on the valence and arousal reported by listeners. For example, the grid above, taken from a research study mapping people’s assessment of songs and emotion words.
Listening to sad music can be good for you
Emotions can be felt (internal) or expressed (external). The way music makes you feel (felt emotion) may be more important to your enjoyment than noticing the emotion the music is trying to convey (expressed emotion). It is important not to confuse the emotional intent of a song – e.g. a ‘happy song’ with how something we hear might make us feel. A ‘happy’ song might make us feel sad if it triggers the memory of a loved one who has died. A song might feel important to us but we can’t listen to it often as it triggers difficult memories. Or we might deliberately listen to a song with negative associations because it has a healing effect, it is cathartic and helps us to reflect and make sense of life-events. For example, when going through a process of bereavement.  We might deliberately choose to listen to a piece of music that makes us feel sad, motivation by a desire to evoke memories of valued past event.
Listening to music with a negative valence can be a positive experience. We can enjoy and appreciate experiencing the beauty of a sad piece of music through dissociation. It is the successful evocation or experience of emotion that is pleasurable. One study examined why this might be so in more detail  and found four different types of ‘pleasure’ associated with listening to sad music:
- The pleasure derived from engaging in an imaginative processes.
- Emotion regulation, such as mood enhancement and venting.
- Empathy – the pleasurable effects associated with sharing the sadness portrayed by the music as an expression of another’s emotion, such as the composer.
- Taking pleasure in music-evoked sadness, by savouring and better understanding its emotional aspects per se, without necessarily experiencing negative ‘‘real-life’’ consequences such as a lost love.
The study also found that sad music evoked not only sadness, but a wide range of complex and partially positive emotions, such as nostalgia, peacefulness, tenderness, transcendence, and wonder. Nostalgia has been characterized as a ‘‘bittersweet’’ emotion because it includes both positive and negative facets simultaneously, such as joy and sadness.
How does the brain respond to music?
In another study, identical musical excerpts were used for all participants, to investigate brain response to music, independent of personal preference. The study used ‘pleasant‘ music (joyful instrumental dance-tunes from the last four centuries) as well as ‘unpleasant’ music (electronically manipulated counterparts of the original tunes with unpleasant pitch and tunings) to investigate not only positive, but negative, responses. None of the participants had any special musical expertise or musical education; none had learned an instrument or taken singing lessons. Functional magnetic resonance imaging (fMRI) was used to analyse the neural correlates of emotion processing as the participants listened to the musical excerpts.
Different parts of the brain responded to the pleasant excerpts (the amygdala, hippocampus, parahippocampal gyrus, and temporal poles), compared to unpleasant excerpts (ventral striatum, the anterior superior insula, and in the Rolandic operculum). In addition, the pleasant musical excerpts stimulated the same part of the brain involved in making vocal sounds (similar to a phenomena known as ‘mirror neurons’ – a type of brain cell that responds equally when we perform an action and when we witness someone else perform the same action). In simple terms, the brain enacted the physical act of singing the pleasant excerpts, even though no actual sound was made.
In another study, the response to pleasant and unpleasant environmental sounds were compared to the response to pleasant and unpleasant music, whilst monitoring respiration, skin conductance level and heart rate. The physiological response to the negative and positive sounds was different to that experienced for negative and positive music. This suggests that emotional valence of music is processed differently by the brain, compared to other non-musical sounds. 
What makes you move?
Whilst current brain imaging technology enables us to see how the brain responds to different musical emotional stimuli, it is still a long way from being able to predict the specific response that a piece of music might evoke in an individual. Studies can attempt to disassociate our emotional response to music from our memories and experiences, but in reality I suspect we all have reasons (conscious and sub-conscious) for why we love the music that moves us. I’m off for a run in the sunshine to Happy…
Author: Dr. Sam Duffy
 Blood, A.J. and Zatorre, R.J. 2001. Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. Proceedings of the National Academy of Sciences of the United States of America. 98, 20 (Sep. 2001), 11818–23.
 Charland, L.C. 2005. The heat of emotion valence and the demarcation problem. Journal of Consciousness Studies. 12, 8-10 SPEC. ISS. (2005), 82–102.
 Collier, G.L. 2007. Beyond valence and activity in the emotional connotations of music. Psychology of Music. 35, 1 (Jan. 2007), 110–131.
 Emotional Valence – APA Dictionary of Psychology.
 Garrido, S. and Schubert, E. 2011. Negative Emotion in Music: What is the Attraction? A Qualitative Study. Empirical Musicology Review. 6, 4 (2011), 214–230.
 Gomez, P. and Danuser, B. 2004. Affective and physiological responses to environmental noises and music. International Journal of Psychophysiology. 53, 2 (Jul. 2004), 91–103.
 Koelsch, S. et al. 2006. Investigating emotion with music: An fMRI study. Human Brain Mapping. 27, 3 (Mar. 2006), 239–250.
 Schubert, E. 2007. The influence of emotion, locus of emotion and familiarity upon preference in music. Psychology of Music. 35, 3 (2007), 499–515.
 Schuller, B. et al. 2010. Determination of non-prototypical valence and arousal in popular music: Features and performances. EURASIP Journal on Audio, Speech, and Music Processing. (2010).
 Taruffi, L. and Koelsch, S. 2014. The paradox of music-evoked sadness: An online survey. PLoS ONE. 9, 10 (Oct. 2014), 110490.
 What Are Moods? Paul Thagard, Psychology Today, (2018).
 The Mind’s Mirror: Lea Winerman, American Psychological Association (2005).
 Yang, Y.H. et al. 2006. Music emotion classification: a fuzzy approach. Proceedings of the 14th Annual ACM International Conference on Multimedia, MM 2006 (New York, New York, USA, 2006), 81–84.
 Zentner, M. et al. 2008. Emotions Evoked by the Sound of Music: Characterization, Classification, and Measurement. Emotion. 8, 4 (2008), 494–521.