Music’s powerful and enduring impact on our emotions, memories, and mental health cannot be understated.
You only need to refer to countless examples of people with dementia recalling their favourite songs to realise how effective music can be.
There are also plenty of examples from within the academic and professional world of music’s ability to trigger various psychological and physiological responses.
To explore all of these examples in more detail, we speak to Professor Jane Ginsborg of the Royal Northern College of Music.
Jane specialises in music psychology – she has a PhD in psychology and is a singer with a degree in music from the University of York. She has written various papers on singers’ memorising strategies, one of which won the British Voice Association’s Van Lawrence Award, and has been shortlisted for the Times Higher Education Best Research Project award.
As someone who holds the rare distinction of being both a musician and psychologist, Jane is perfectly placed to explain the psychological effects of music on the brain.
Here’s what she had to say.
Which areas of the brain perform the most crucial functions in eliciting certain emotional responses from music?
I should say up front that I’m a psychologist, not a neuroscientist. Therefore my understanding of the brain’s functions is not what you would call advanced, but it’s sufficient to teach undergraduate students what they need to know.
The temporal lobe is crucial in eliciting emotional responses from music because it processes what we hear across both the left and right hemispheres of the brain. Words are processed in the left hemisphere and music in the right hemisphere.
But whether or not we’re listening to songs with lyrics – and it’s often the lyrics that elicit the emotional response as much as the melody and harmony – we’re using both hemispheres. I want to drive this point home.
People often talk about right brain and left brain, but that is a metaphor. When we’re talking about neuroscience, we’re referring to the right and left hemispheres of the brain and we work across both of those hemispheres. That’s a crucial point to make.
The nucleus accumbens is very important, too, as it releases dopamine, the chemical that makes us feel good. The effect of music is to increase the amount of dopamine this part of the brain releases, and this dopamine hit makes you feel as though you’re profiting from listening to music.
When we hear music, the amygdala may be activated. In this instance, we’ll experience physiological responses such as a cold sweat, goosebumps, or the hairs standing up on the back of our neck.
I can provide a personal example here – I watched The Metropolitan Opera Centennial Gala, a televised concert that went out to millions of people across the world in 1983, and Kiri Te Kanawa was performing Dove sono from The Marriage of Figaro by Mozart.
I was learning the piece myself at the time, and Kiri Te Kanawa would likely have performed it a thousand times. And yet, she made a mistake, and it triggered a physiological response within me. The hairs on the back of my neck stood up – I knew she’d gone wrong, and it was as though I had gone wrong. I experience the same response whether I’m watching someone make a mistake or making a mistake myself whilst performing. I also get the same response if I hear something I find wonderful.
When we experience responses like this, we interpret it as an emotion. We may label it anticipation or fear, but it’s all the same thing – a form of arousal. This is a psychology-centric word for excitement.
When I teach my students about music performance anxiety, I borrow a technique I learned from a colleague who asked a group of students to write down on a piece of paper the way they felt the last time they fell in love. I say: ‘Imagine you’ve just fallen in love with somebody, and you know you’re about to meet them. Write down what it feels like.’ Then, about half an hour later, I ask them: ‘What do you feel like when you’re just about to go on stage for an important concert? Write it down and don’t tell anybody.’ Then, I ask them to compare their notes to what they wrote half an hour ago about their beloved, and they invariably say that they wrote down the same things.
I show them a Gary Larson cartoon of an elephant sitting at a grand piano on stage, and the thought bubble says ‘What am I doing here? I can’t play this thing! I’m a flutist, for crying-out loud!’ The body only has one way of being excited, and this excitement comes from the activation of the amygdala. We call this excitement stage fright or anticipation.
There are many examples of patients with Alzheimer’s disease recalling their favourite songs and responding positively to music, as the key areas of the brain which are linked to musical memory, such as the cerebellum, are relatively undamaged by the disease. Can you outline what role the cerebellum plays in processing music and just how powerful it can be?
The cerebellum is the area of the brain in which our procedural memories are stored. That is, memories of things that we find very difficult to do when we’re little, like tying shoelaces or learning to ride a bicycle. You can’t tell anyone how you do it because it’s hard to put into words.
This kind of memory is not rooted in language – rather, it’s stored in the muscles we use to perform these tasks. So, once we’ve learned how to tie a shoelace or ride a bicycle, we never forget how to do it.
You can apply this principle to learning how to play an instrument. This process involves endless repetitions of the same movements – if, for example, you’re learning to play the guitar, you’re playing the same chord formations over and over again, or if you’re learning the violin, you’re practising the same scales until they become second nature to you.
Those repetitions strengthen the connections along the neural pathways that run between the synapses in the brain. Synapses are like junction boxes – think of the electric wiring inside these boxes. Procedural memories are stored in the cerebellum’s neural pathways in the same way that wiring is stored in a junction box.
Music-making is very often preserved in people who have Alzheimer’s. I spoke at a conference on music and dementia, and I bookended it by talking about my mother’s youngest cousin. This woman was in a different generation to me but not very much older than me, and she developed Lewy body dementia when she was only in her mid-70s. She was the kin keeper of the family – she knew who everybody was, when their children were born, when they got married, and when their bar mitzvahs were. She kept everybody together. And so, to lose her memory was horrendous.
But she loved music – she had played the piano as a girl and young woman, and her son has gone on to become a very talented opera singer. The family invited me over for a Hanukkah and Christmas celebration at which we sang Jewish songs and various festive carols. When we ran out of carols, we moved on to Handel’s Messiah, which her son was singing at the time. She sang everything with us – when she wasn’t singing her part, she sang the violin part. This was a woman who barely remembered people from outside her family – it was quite remarkable.
Another recent example I can cite is Paul Harvey, a composer with dementia, whom many people will have seen on TV performing a piano improvisation based on four notes.
It’s a beautiful example of the power of music-making – Mr Harvey wasn’t relying only on his procedural memory in the cerebellum to be able to do that, but all the areas of his brain that had only recently begun to be affected by dementia. For example, he will have used the frontal lobe for planning, the Broca’s area for communicating, and the hippocampus for producing and retrieving memories and regulating emotional responses.
On a related note, a study by neurologist Gottfried Schlaug found that the cerebellums of musicians were 5% larger than those who didn’t play music. How much does music’s ability to enhance memory continue to fascinate you, and what do studies like this say about its psychological effects?
Gottfried Schlaug and his colleagues compared a section of the corpus callosum, part of the cerebellum, in 30 professional musicians and 30 non-musicians. These musicians were matched for age, sex, and handedness.
Schlaug and his colleagues did find a larger corpus callosum in musicians compared to non-musicians. However, this was attributable to a sub-group of musicians, who had started learning their instruments before they were seven.
This finding tells us that the larger size of the corpus callosum is related to the amount of practice musicians had done at an early age. The practice they’d done had laid down those procedural memories that I talked about to in response to the previous question.
Before you’re seven, your brain is at its most plastic or flexible, so anything you learn before the age of seven will strengthen the neural pathways and expand the cerebellum. But it isn’t only before you’re seven that you can achieve this. A study found that London taxi drivers had denser, or larger, parts of the brain that are responsible for route finding, and this result is based on their having learned the Knowledge of London. This effect mirrors a musician’s ability to perform certain melodies, harmonies, and so forth.
Schlaug’s study is of great interest to me because I was a professional musician before becoming a psychologist. My husband is a pianist, conductor and composer, and we brought up two children who went on to become professional musicians themselves.
Since I started working at the RNCM 15 years ago, I’ve worked explicitly with expert performers and music students seeking to pursue professional careers as singers and instrumentalists. So, I’m fascinated by the development of expertise and the role of practice in this development – we now know that practice makes permanent rather than perfect.
But whilst the size of a musician’s corpus callosum – and the reason it might be larger than those of non-musicians – is interesting, these findings don’t tell us anything about the psychological effects of music on listeners. That said, all that practice – and the motivation to do it, and to keep at it for a lifetime – certainly has a myriad of fascinating psychological effects on the musician.
Fortunately, performing has psychological effects too, and they include the release of dopamine by the nucleus accumbens.
You’ve had a successful career as a singer and have also won an award for your research on singers’ memorising strategies, which was inspired by your own experiences. Can you explain what the purpose of this research was?
Thank you for this question! The research for which I won the award was my PhD research, and you’re absolutely right to say that it was inspired by my experiences as a singer.
Singers typically perform from memory and, as well as specialising in contemporary music, I used to give a lot of song recitals. So, during my career, I spent a lot of time memorising and recalling words as well as music. I wanted to know whether the two kinds of information represented a double load on memory, so to speak, or if they interacted in such a way to be helpful. If I remembered the music, would I remember the words and vice versa?
However, the purpose of my research was not to find out what worked best for me, but what was likely to work best for all singers, and why. So, I’m not going to answer the question of which strategies were most effective based on my experience, but a different question: on the basis of the evidence from research, which strategies are most effective and why?
Based on the evidence from your research, which memorising strategies did you find to be most effective and why?
I started by reading everything I could find about the strategies that singing teachers recommend. However, there wasn’t very much because, at that time, memorisation wasn’t taught routinely in most institutions.
And I read everything I could find about the strategies recommended by instrumental teachers, going back to the end of the 19th century. Fortunately, psychologists started researching musicians’ memory in the middle of the 20th century. Yet, no one had asked my question about the relationship between words and melodies – at least not in terms of recall. I also looked at how actors memorised their lines – this was interesting because there’s a difference between memorising Shakespearean blank verse and plays in ordinary and vernacular language.
I began my own research with an interview study (as described in Aaron Williamon’s book on musical performance strategies and techniques). I asked five singers and a pianist who used to accompany himself singing cabaret songs how they went about memorising the words and melodies of songs. Unsurprisingly, I got six different answers. One respondent started with the words, another with the melody, and a third memorised them both together. The fourth respondent, my old music teacher, would speak the words to the rhythm of the melody. The fifth respondent, an opera singer, not only memorised words and melody, but also the position of his body, hands, head, and direction of eye gaze – it was a form of imaginary dance notation.
To summarise, these results weren’t very helpful. So, my next step was to conduct an observational study for which I recruited 15 singers – five amateurs, five professionals and five students. These participants learned and memorised the same song in up to six 15-minute practice sessions over two weeks and then sent me recordings of their practice sessions.
I told them to imagine they had been asked to give a concert in two weeks’ time, and they knew everything on the programme except this song, and to record all their practice sessions. I also asked them to tell me when they were and weren’t looking at the score. When they could sing the piece twice through from memory without a mistake, they could send me the tapes. Sadly, only 13 of the recordings were usable, but I had enough data to be able to write up my research.
I had thought that if I found out what the professionals did, they would have the best strategies – but this wasn’t the case. The ones who gave the most accurate performances after the shortest time weren’t all professional singers. But they all used one of two strategies: memorising the music thoroughly and then adding the words or memorising the words and music together.
How did you follow up on these findings?
I followed up with two experiments. The first one (reported in Ginsborg & Sloboda, 2007) involved sixty singers – 25 with relatively advanced musical qualifications and 35 without (I called them experts and novices).
I produced a new, unaccompanied folk song – I took the words of one folk song and put them to the melody of another. The study participants memorised the song in one of three ways during a single half-hour session:
- Words first, then melody, then both together.
- Melody first, then words, then both together.
- Words and melody together for the whole session.
I gave each participant a tape-cassette player to play the song back to themselves, and after the session was over, I asked them to sing me the song. I then interviewed them and talked to them for ten minutes about their musical qualifications and background. At the end of these conversations, I asked them to sing me the song again, and they were all caught off guard by this.
I recorded their second performances and analysed their errors and hesitations. The most accurate and fluent performances were given by the experts, as they had memorised the words and melody together. Based on the evidence I gathered, my recommendation to musicians would be not to treat the words and melody as two separate components. Providing the melody has been learned accurately in the first place, you should treat words and melody as two sides of the same coin or two facets of the same jewel. To go back to the brain, it’s processing the words and melody of the song in both hemispheres of the temporal lobe and the cerebellum.
Finally, I did a second experiment (also reported in Ginsborg, 2004a) in which I tested the hypothesis that memorising words and melody together enables singers to memorise, and perform from memory, repertoire in languages they neither speak nor understand. For example, how do opera singers sing in Czech or Russian if this isn’t their native language? I did this by creating two songs that were settings, on the one hand, of semantically meaningful words in English and, on the other, of meaningless strings of numbers. The meaningless songs took longer to learn but were recalled just as accurately by the 20 expert singers who took part in the study, confirming my hypothesis.
It’s fascinating that you co-authored a study that investigated the mental representation of musical notation. Having researched this subject in more detail, I discovered that the occipital lobe is the part of the brain which facilitates this representation. Can you explain, in a nutshell, what the occipital lobe helps musicians to do?
The concept of mental representations is crucial to research on musicians’ memory. Way back at the beginning of the 20th century, the pedagogues Edwin Hughes and Theodor Leschetizky wrote about different ways of memorising: using visual, auditory, kinaesthetic, and analytic strategies.
Visual could be for the score, the fingers on the keyboard or any other instrument, or conductor’s cues. Auditory is for the sound of the music just sung or played, or about to be sung or played. Kinaesthetic is procedural, motor memory. But they’re all held together by an understanding of the structure of the music that is to be memorised.
Visual representations and imagery are processed in the occipital cortex. The most effective memorising strategies are those that make use of all four types of representation: visual, auditory, kinaesthetic, and analytic – and they may also make use of narrative, whereby the performer recalls the programme or the story of the music. The story may or may not be the same as that intended by the composer (see Ginsborg 2004b and Ginsborg, in press).
I always say to my students – when you listen to a piece of music, don’t worry about how the composer would have analysed it or how an analyst would analyse it, just consider what makes sense to you.
Another piece of research that we found interesting was your most recent paper on Developing Familiarity in a New Duo. Can you talk us through the methods you used to explore the relationship between practice, rehearsal, and performance?
This piece of research was the third I’ve carried out in a series of what we call longitudinal case studies – longitudinal in the sense that they take a long time!
The first one involved repeated tests of recall over six years – and it’s called a case study because it involves only one or two musicians at a time, unlike my PhD research which involved more than 100 singers.
These three studies were also practice-led, in that – again, unlike my PhD research – they focused on my preparation for performance from memory, which I analysed myself. My colleague Roger Chaffin, a British cognitive scientist who has been based in the United States since the late 1960s, has conducted a series of longitudinal studies – principally with a pianist and a cellist – in which the musician participated in the research and he, as the cognitive scientist, conducted the analyses.
Roger has developed a theory of what he calls performance cues. He argues that musicians rely on landmarks in their mental representations – or maps – of the pieces they’re memorising to recall them when they perform from memory. He’d only worked with individual performers, however, whereas I was interested in joint representations, i.e., what happens when two musicians such as a singer and pianist are rehearsing together to produce a performance.
Roger and I collaborated on the design, data analysis, and interpretation of the findings of my first two studies. This collaboration involved recording my individual practice sessions, the pianist’s individual practice sessions, all our joint rehearsals, and the final performance. I then transcribed everything we sang, played, and said to each other, and we identified specific features of the words and music we thought we were likely to use as landmarks for retrieval when we performed from memory. After the performance, we noted the features that we had actually used as landmarks for retrieval, or performance cues.
In the first study, we were able to show a relationship between what we rehearsed and how we rehearsed it and particular kinds of performance cues – especially those that we refer to as ‘expressive’. By that, I mean the locations where we wanted to convey a specific feeling or emotion to the audience.
I also carried out six free recalls over the course of the next five years – writing out the vocal line from memory – so that we could look at:
- How my memory for the piece faded over time.
- What I remembered.
- What I had forgotten.
I had always been a bit critical of one part of the theory, which proposed that performance cues are a subset of the landmarks or features of the music identified during rehearsal.
First of all, this would imply that performances are equivalent to rehearsals except that they take place in public. From my own experience, I know this isn’t so – unexpected things happen in performances, sometimes bad (such as memory lapses) but much more often good – and these may be remembered in subsequent performances of the same piece.
I also pointed out that you couldn’t call them performance cues unless they actually functioned as retrieval cues in performance. We had only looked at one performance – you wouldn’t know if something was a performance cue unless it cues performance next time.
So, in our second study, my pianist and I learned and performed another piece. We noted the features of the music we thought would function as performance cues. We gave the performance and identified the features we were actually thinking about.
That’s a very interesting approach. What were the results of this second study?
A few months later, I gave not another performance, as such, but a reconstruction of the piece from memory. I tried to sing as much as I could without the piano – not very much – and then, with the piano accompaniment, reconstructed it phrase by phrase until I could sing it all the way through from beginning to end without any mistakes. We recorded this session and transcribed it to see how I had built up the piece, and once again, the reconstruction process reflected the performance cues. But, importantly, I recalled the new thoughts I’d had while performing, as well as the features I’d prepared in rehearsal all those months before (Ginsborg et al., 2012).
This brings us to the third – and most recent – study. One potential criticism of the previous two studies is my very close relationship with my pianist – we’ve been together for 45 years and are married to each other. So, the question we asked in this most recent study was how does a new duo develop, and make use of, performance cues?
I travelled to Australia and stayed with a fellow musician, academic and researcher named Dawn Bennett, whom I didn’t know well – we had read each other’s work and met each other at conferences – and had never performed with before. We worked together every day and used the same method:
- Recording our individual practice and joint rehearsals every day for a week.
- Transcribing everything we played and sang, and everything we said.
- Analysing them in the same ways we had done in previous studies.
We noted the features of the music that we identified as potential performance cues at the end of the rehearsal period, and those that we had actually used as cues during public performance, on two occasions at the end of the rehearsal period in March 2014, and again in a third public performance in January 2015.
The piece was new to us both and consisted of two songs for voice and viola. We both memorised one of the songs and sang or played the other using the score.
The paper outlines how certain cognitive processes are triggered by practice, rehearsal, and performance. What are these processes and, of all the results you collected from your research, which did you find most interesting?
To summarise the cognitive processes – musicians identify features of the music during individual practice and, if working with other musicians, joint rehearsal. Some of these are retained as performance cues in one, two, or more subsequent performances.
Others are forgotten or assimilated. For example, taking a breath or using two down-bows instead of a down-bow and an up-bow becomes automatic, so the musician doesn’t have to think about it while performing. And some performance cues are thoughts the musician has in one performance that recur in a second performance.
For example, in one of the songs I sang in my second study, there were two canons in the piano part. I’d registered the first one during rehearsal, but I only registered the second one during the first performance. A few months later, when I sang the song again, I was consciously listening out for that second canon, and I sang my own entry differently as a result.
In the third study, when we compared Dawn’s and my use of performance cues, I had far more ‘core’ performance cues than she did. The things I prepared in rehearsal stuck during all three performances. But we think that this is because I was much more used to thinking about using them as a memorising strategy than she was. After all, I’d been thinking about them and teaching students at college to use them for nearly ten years by the time we carried out this project, so performance cues are second nature to me. Dawn is also a viola player and singers are more used to memorising than viola players.
A lot of memorising is implicit – it comes with learning, which is why it’s so important to learn accurately. But implicit memory isn’t reliable in performance. It needs to be carried out deliberately, and you need to employ multiple strategies so that if one fails, another will kick in.
If, for example, you’re a singer and you get the words in the wrong order, it doesn’t matter if you know what it ought to sound like and can use an auditory memorising strategy to get the next set of words right.
More generally speaking, you might suddenly forget the sound of what comes next, but you can visualise the score. You might lose your place in the music, but if you have landmarks to remind you where the last section began or the next section starts, you can jump backwards or forwards as you need to, and the audience will never notice.
So, what did I find most interesting from my research? Well, one of the analyses we reported in the paper you read was of our rehearsal talk, what we said to each other in rehearsal. This turns out not to have reflected what we did in rehearsal. And it’s what we did that determined how we memorised and what we remembered. So, if you want to know how musicians prepare for performance, it’s better to watch and listen to what they play and sing than to rely on what they say!
Generally speaking, how can music have a positive impact on our cognitive ability on a day-to-day basis?
I don’t think anyone should listen to music or make music solely in order to improve their cognitive ability. I’m very distrustful of anyone who tells you that music makes you smarter, and I certainly don’t think music should be taught in schools to improve mathematical ability or social skills, even though there’s evidence that it does. Why not? Because music is important in its own right, and we should enjoy it for its own sake.
Having said that, singing is essential to communication between parents and small children, and it can foster a love of singing and music more generally. Learning to play an instrument requires discipline and practice and allows for the development of creativity through improvisation and composition. All of these characteristics can, of course, be transferred to other kinds of learned skills.
Playing music with other people develops the cognitive skills of attending, perceiving, listening, and adapting, as well as social skills to do with leadership and active membership of groups. Learning about music can provide windows into different periods of time and different cultures. All of that has got to have a positive impact on our cognitive ability.
What about music’s impact on mental health – how can music help alleviate various mental health conditions and act as a medium for processing emotions?
I’m currently editing an article on music in prison and how prisoners express difficult emotions through music, so this is a subject I’m well acquainted with.
Music can be a comfort, an escape, a way of remembering, and looking forward to happier times. Making music with other people is a way of communicating with them without having to use words and can often be a way to process emotions.
But I think we must differentiate between the mental health of musicians, and that of people who don’t identify as musicians. The way our profession has evolved, musicians have extremely stressful lives, and in some ways never more so than since the beginning of the pandemic.
That said, my colleague Susanna Cohen and I conducted a study for which we interviewed 24 freelance orchestral musicians about their experiences of the first lockdown. Some of the musicians we interviewed (Cohen & Ginsborg, 2021) said it was a great relief not to have to be away on tour all the time and working for such long hours. They can experience musculoskeletal disorders, repetitive strain anxiety, and – thinking particularly of mental health – stage fright or music performance anxiety.
They sometimes have to cope with difficult bandmates, desk partners, managers, and conductors. They have a lot to contend with, yet they’re still expected to turn out wonderful performances. Meanwhile, the wonderful music they produce is being ‘socially prescribed’ for people with mental health conditions and elderly people, for example. I gave a talk at Liverpool Institute for Performing Arts, and I began by saying that if you Google ‘music and health’, you will get a million results on how music is good for you, but if you Google ‘musician’s health’, you will get a million results on all of the bad things.
But, to end on a positive note, growing evidence shows that, for many musicians, it’s the amazing experience of performing that keeps them going. They couldn’t imagine their lives without it – and neither can I.