Thursday 26 January 2012

Sixth Sense

I’m watching a friend play a Chopin Waltz.  Her eyes are kept trained on the music in front of her, but her left hand is jumping back and forth between the bass notes and the upper notes of each chord.  Her pinkie strikes a low C# and then her hand moves right off the keyboard and, unguided by her eyes, flies up to play a C# minor triad over an octave higher.  Then, in a split second, she leaps her hand down to a low D, again without looking.  How do her fingers land unerringly on the correct notes each time?  To the untrained, this seems like magic, or maybe a sixth sense.

How many senses?
Ask any school child how many senses we have, and you’ll get the answer:  five.  Sight, hearing, taste, smell, and touch.  We’re all taught this, but the answer is correct only up to a certain point.  The reality is that we have five senses that tell us about the outside world, but a number of other senses that tell us about our own body.  The two main internal senses are balance, which tells us about the orientation of our head and body with respect to gravity, and proprioception, which tells us about the positions of our various body parts.

 (Photo credit:  aarrgh via Flickr)

If you’re not convinced that you have a sense called proprioception, try this little demonstration:  Close your eyes.  Now wave your right arm all over the place.  Then touch your nose with your right forefinger.  You didn’t have any trouble finding your nose did you?  Even without looking, your brain knew exactly where in space your finger was, and exactly where your nose was.  It accomplishes this by receiving information from special receptors in your muscles and joints, which tell the brain how much stretch there is on each muscle, how much load there is on the muscle, and what angle the joints are bent at.  From this information, the brain assembles a spatial map of your body so that it knows how to move in order to accomplish a particular movement goal, like touching your nose.

Conscious and unconscious proprioception
Proprioception information from the muscles and joints travels up the spinal cord to two separate places in the brain.  One set of information goes to the somatosensory cortex (which also receives information about touch), and this provides us with conscious awareness of where our body is in space.  An identical set of information travels to the cerebellum, where it is used unconsciously to provide feedback about our movements, and to aid in motor learning.

We don’t think much about our sense of proprioception, but we are using it all the time.  This is especially true of instrumental musicians.  Consider the left hand of my friend, jumping back and forth between the bass note and upper notes of each chord.  An inexperienced (or unpracticed) pianist would need to watch her left hand to make sure she hits the correct notes, but with practice the hand “knows” exactly how far to move to land on the right keys.  This is proprioception in action.  String players similarly have to know exactly how far to move each finger to play the next note.  They don’t have to watch their fingers.  The fingers themselves are sending signals to the brain as to their positions, so we know where our fingers are without watching.  (String players, of course, also use auditory feedback to know where their fingers are, since the position of the finger on the string fine-tunes the pitch of the note)

Practice makes perfect
Like anything else the brain does, proprioception can be improved with practice.  This can start away from the instruments, by having students move particular fingers without looking.  In my beginner MYC classes we always spend time singing finger number songs like “Where is finger one?” and having the children stick up the correct finger as we sing.  After the children know which finger is which, we can try this exercise with eyes closed.  It’s trickier but very useful.

It’s common for young students to use vision instead of proprioception to guide their playing.  A little girl in one of my classes last week had exactly this problem, and had to pause and look down at the keys periodically while performing for me.  Then, as she returned her eyes to the page in front of her, she had to pause again to find her place. I recommended that her mother hold a piece of cardboard over her hands as she practices to force her to use proprioception instead of watching her hands.   

Forcing ourselves to play without looking is an excellent way to improve proprioception.  Once technical exercises are mastered, I recommend that an additional week be spent practicing them with eyes closed.  This can be quite challenging for chords and arpeggios but will pay dividends in the future.

My experience is with piano, but each instrument has its own proprioceptive demands.  For example, I also play the flute, and it’s pretty much impossible to watch your fingers while you’re playing the flute.  Flutists, therefore, are required to rely on proprioception from the very beginning of their training.  I’d love to hear suggestions for improving proprioception on any instrument.

Thursday 19 January 2012

“Fast” and “Slow” Learning

Every once in a while, I take a step back from working with my students and notice just how far they have come since they started studying piano.  That adorable five-year-old learning to harmonize her scales was (just yesterday, it seems) a chubby three-year-old who struggled to coordinate her fingers to play three different notes in a row.  The focused eight-year-old determined to get the hang of using the sustain pedal used to be an overly-bouncy five-year-old who had a difficult time playing with two hands. Part of their progress is developmental, of course, but mostly it is due to their daily practice and commitment to learning music.

Fast vs. slow learning
When we practice a musical instrument (or other motor skill) over a long period of time, there are actually two types of motor learning going on.  The first is “fast” learning, which happens on a timescale of minutes to days, and is characterized by automatization of specific motor tasks.  This is the learning that we see when a student gets better at playing a particular piece of music.  It is likely due to changes in synaptic strength in particular brain regions. Scientists see a decrease in brain activity as fast learning occurs. 

In contrast “slow” learning occurs over weeks to years and is seen as gradual improvements in performance.  This is the learning that we see over the long-term, when students get better at playing overall, so they’re able to master more difficult repertoire.  It depends on reorganization of areas in the brain, and the recruitment of additional brain regions for certain tasks.

We can illustrate the difference between these two types of learning by comparing a beginner musician with an experienced musician.  Imagine the two musicians are given the same piece of music to learn, a piece of intermediate difficulty.  For the beginner, the piece would be extremely hard to play, because she would not have the basic motor skills needed to perform it.  For the advanced musician, the piece would be less difficult, because she would already have learned “how” to play.  The difference between the two musicians is in the amount of slow learning that they have done.  The experienced musician has motor patterns laid down in her brain that allow her to more-or-less effortlessly convert the written music to movements which cause the correct sounds to be produced.

If both musicians practiced this piece, they would both get better at playing it, through repetition of the specific movements required to play the piece.  This is fast learning.

Brain changes in fast and slow learning
Studies have shown that fast and slow learning involve changes in different parts of the brain.  Susan Landau and Mark D’Esposito at UC Berkeley performed a fascinating study that was published in 2006.  They compared the brains of pianists and non-pianists performing a finger-movement task while undergoing fMRI scans, to see what parts of the brain were being used.  This study looked at slow learning that had already happened in the brains of the pianists, by seeing how their brains were different from non-pianists.  And it also looked at fast learning, by seeing how the brain of each individual changed as the participants improved at the task.

What they found was that the pianists used several areas of the brain that were not used by non-pianists, most notably the caudate nucleus (part of the basal ganglia, thought to be involved in motor planning) and parts of the parietal lobe, which is known to be involved in integration of sensory information, and has been previously shown to be activated during music playing.  During their years of musical training, the pianists have recruited these extra regions of the brain to be used during finger movements.

For fast learning, the study showed that brain activity decreased as participants learned the task.  This was particularly true for activity in the premotor and supplementary motor cortex.  What this means is that our brains have to work hard at a new task.  We can feel this, because new tasks seem effortful.  As we repeat new tasks, the areas of our brain responsible for planning and organizing our movements don’t have to work as hard – some of their burden has been shifted to automatic memory processes.

Memory representations
If we learn new but similar tasks over a period of weeks to years, we learn to use different strategies, involving different parts of our brain.  A lot of this involves “chunking” – grouping thoughts or movements into chunks.  For example, instead of  playing a C and an E and a G, we just play a C triad.  Our brain thinks of it as one entity instead of three separate notes, and this representation is much more efficient.  There’s also a motor representation of this: when I think of a C triad, I can imagine playing it and it’s a one-step hand movement.  My brain has a motor pattern already set up to play a C triad, and this uses a lot less brain power than reading each note separately and converting it into movements of each finger.

Teachers guide slow learning
Students (especially children) tend to focus on fast learning:  they just want to get better at playing each particular song.  They realize that their playing is improving generally, but they are not thinking about what path their slow learning should take.  This is what I, as a teacher, need to be aware of:  where is their slow learning taking them?  The teacher’s role is to guide the student's overall motor development.  A given student might need to practice playing in a particular key, or using four-note chords, or playing lots of octaves, so that these hand positions and characteristic movements will become part of their repertoire of motor patterns.

My students are gradually learning motor strategies and memory representations for musical paradigms, and this slow learning is what is turning them into better pianists.

Thursday 12 January 2012

Earworms: What Makes a Tune Sticky?

 (ear photo by jemsweb, worm photo by daz smith; from Flickr Creative Commons)

Last weekend I had the pleasure of attending a reading of an amazing new musical, One plus One, written by my friends Gil and Sarah Jaysmith.  The music from this show has been bouncing around in my head ever since.  You’ve certainly had this experience:  a fragment of a tune “stuck in your head”, looping over and over.  It seems the only possible way to shake this experience is to replace the tune with another, and even that’s only successful some of the time.  We call these tunes “earworms” and music that tends to get caught as earworms is known as “sticky”.

But are there really characteristics of music that make it more or less sticky, more or less likely to be an earworm?  There is not a lot of research yet about earworms, but it’s starting to crop up in journals about the psychology of music, so I had a quick look to see what’s known.

To start with, scientists tend to call this phenomenon “Involuntary Musical Imagery”, INMI for short.  It’s kind of hard to study something that can only be based on people’s verbal reports of its existence, so the studies that are out there all use the technique of simply asking people about their earworm experience.  And the studies have shown something you may have guessed:  earworms are very common. Over 90% of people reported that they experience earworms at least once a week.  About 60% have them at least once a day, and 26% of people have them more than once a day (these numbers are from Liikkanen, 2008).
Recent vs. Sticky
Another interesting study from Finland looked at whether there was a recency effect in earworms.  The recency effect is a well-known memory characteristic, in which we’re more likely to remember the last item in a list, rather than an item in the middle, because we’ve seen it more recently and also, no other items have been introduced to interfere with the memory.  In this 2009 study, Liikkanen proposed that tunes that we’ve heard most recently are most likely to be stuck in our head.  This makes sense to me based on my own experiences:  I’ve often had an old pop song stuck in my head and wondered, “Where did that come from?”, but when I thought a little, I remembered that that song had been playing in the grocery store I had visited 10 minutes earlier.  I hadn’t even really noticed the song when it was playing, but just hearing it caused it to be stuck in my head almost right away.

Liikkanen performed an internet study using thousands of Finnish internet users, and he triggered earworms by having the particpants complete lyrics to a number of songs.  The order in which the songs were presented was varied, so that Liikkanen could test whether the song presented last was more likely to trigger an earworm.  After completing the lyrics, the participants did a “filler” exercise for four minutes, and then they were asked if they had experienced any earworms.  About 50% had, and the results showed two things.  First of all, some songs were just more “sticky” than others, no matter what order they were presented in.  But also, the song presented last was more likely to get stuck in the participants’ heads.  So yes, there’s a recency effect.  But the characteristics of the songs themselves also have an effect on whether they’re likely to get stuck in your head. 

On the other hand, a diary study by Beaman and Williams (2010), in which 12 people reported in detail on over 250 earworm episodes, found that only 10 different tunes were experienced as earworms by more than one person.  You’d think that if certain tunes were intrinsically very sticky, then these tunes would get stuck in everyone’s head, especially if they were current pop hits or TV themes, or current advertising jingles.
The most recent paper on the topic of earworms, by Williamson and colleagues, looked at the circumstances under which they occur, and amassed a large pool of earworm data from internet users.  The researchers were trying to answer the question:  “Under what circumstances do earworms occur?”, and their results were interesting.  The most common circumstance in which a song got stuck in someone’s head was when it had been heard recently, or heard repeatedly.  This jibes with Liikkanen’s 2009 study.

Another common circumstance that triggered an earworm was through association – for example, seeing a picture of Elvis might trigger Blue Suede Shoes, or remembering a high school dance might start Stairway to Heaven playing in your head.  Obviously these associations are highly individual.  Mood plays a factor too, since certain moods like sadness or stress can trigger certain melodies for some people.  In fact, just thinking of whether sadness triggers any melodies for me brought into my head the song I’m so lonesome I could cry, as performed by Holly Cole.

Another experience that was identified as leading to earworms was boredom, when the mind starts to wander.  And this I find particularly interesting because I think this is a clue to what parts of the brain might be involved in generating earworms.

The Default-Mode Network
In general music imagery activates the same parts of the brain that are activated when we listen to music.  If you imagine hearing a song, it actually activates the auditory cortex just as if you were hearing it.  However, involuntary music imagery is probably slightly different, because it tends to occur when we’re not actively thinking about the music.  There’s actually a brain network that’s activated when we’re not consciously thinking about anything: the default-mode network, which consists of the medial prefrontal cortex, precuneus, posterior cingulate cortex, inferior parietal cortex, and lateral temporal cortex.

Figure from Buckner et al., 2008

Activity in the neurons in this network goes down when we’re thinking about something, and increases when we’re doing nothing.  Activity in the default-mode network is believed to be the neural correlate of mind-wandering.  There hasn’t been any research on this yet, but I’d be willing to bet money that your default mode network is active when you’ve got an earworm.

Characteristics of Sticky Songs?
The research on earworms certainly is a start, but I was surprised that I couldn’t find any research that looked at musical characteristics of songs to try to determine what would make a song “sticky” or not.  There is a lot of speculation about it, but nothing even close to conclusive. 

Here’s my own take on earworms:  I have music stuck in my head pretty much all the time, except for when there’s music actually playing, or if I’m busy teaching, talking, or thinking about something.  In other words, if my mental effort is directed somewhere, no earworms, at least not that I’m aware of.  If I’m doing something that doesn’t require my conscious thoughts, then there’s an internal soundtrack, usually a repeated fragment of a song.  What this means is that the earworms are generated when my default mode network is active:  when I’m cooking meals or washing dishes, during my 30 minute bike ride to pick up the kids from school, when I’m waiting for a bus, when I’m lying in bed trying to get to sleep.

The default mode network seems to be responsible for working through background thoughts, making connections, making sense of stuff that’s happened recently.  And what’s happened to me recently?  Well, lots of music.  The music that gets stuck in my head the most is music that I’m trying to learn, or trying to make sense of:  songs from a new musical I’ve heard, for example.  In the last few days, when my earworms haven’t been Jaysmith earworms, my mind’s ear has been working away at the “Mah na Mah na” song from the Muppets.  I’ve been teaching this song to my students with the ambitious plan that they will sing it as a combined group at their upcoming piano recital.  I still haven’t quite figured out the best way to teach them the rhythms, which are a little complicated, and I also haven’t figured out who will sing which bits, and whether the verses will be improvised scat or sung as written.  I’ve also been practicing the piano accompaniment.  With all this focus on the song, it’s not at all surprising that it’s the one stuck in my head:  I’ve heard it lots recently, and also my conscious mind has several problems to solve with respect to it.

The research on earworms is admittedly pretty limited so far, but a lot of it points to the idea that earworms are all about memory:  the tunes that turn into earworms for us are those that are important, heard recently, heard repeatedly, or are triggered by an association with something else.  None of this has much to do with the characteristics of the music itself, but anecdotally we certainly find that some music is “stickier” than others.  I think this is simply because some music is more memorable than others.  Sticky music is tuneful, entertaining, accessible without being predictable, and contains repetition of the themes to help you remember it.  The Jaysmith’s music easily has these characteristics, and it’s not surprising that it has a strong tendency for earworm formation.

Memorable music gets remembered, and our default-mode network has us thinking about it when we’re not doing other things.  The really interesting question about earworms is why we can’t stop them.  Some people have related them to obsessive-compulsive disorder (OCD), in which people have thoughts that they are unable to stop.  But no one has shown a relationship between OCD and earworms, so it’s completely unclear whether there is any neurological similarity.  Why can’t we stop these bits of tunes from bouncing around in our heads?  No one knows.  I guess the researchers still have some work to do. 


  • Beaman CP, Williams TI. Earworms (stuck song syndrome): Towards a natural history of intrusive thoughts. British Journal of Psychology. 2010;101(4):637-653.
  • Liikkanen LA. How the mind is easily hooked on musical imagery. Proceedings of the 7th Triennial Conference of European Socity for the Cognitive Sciences of Music. 2009.
  • Liikkanen LA.  Music in Everymind:  Commonality of Involuntary Musical Imagery.  Proceedings of the 10th International Conference on Music Perception and Cognition. 2008
  • Schürmann M, Raij T, Fujiki N, Hari R. Mind’s Ear in a Musician: Where and When in the Brain. NeuroImage. 2002;16(2):434-440.
  • Williamson VJ, Jilka SR, Fry J, et al. How do “earworms” start? Classifying the everyday circumstances of Involuntary Musical Imagery. Psychology of Music. 2011. Available at: Accessed January 12, 2012.

Wednesday 4 January 2012

This is my Brain on Books about the Brain

Ahhhh… The kids are back at school and I finally have a little time to myself.  As I take a few days to catch up, I thought I’d give you an informal post about what neuroscience and music books I’ve been reading lately (and what I think about them), what I’m reading right now, and also what I plan to read soon.

I should preface my reviews of these books by saying that I generally have little patience for books written about music and the brain.  This sounds like I’m being snotty, but it’s just a topic that I already know a lot about, so I have to sift through a lot of information to find something that catches my interest.  Daniel Levitin’s bestseller This is Your Brain on Music was particularly mind-numbing for me (although I do recommend it), because it assumes the reader knows nothing about music and nothing about the brain.  It’s a good book, but I am certainly not the intended audience.

Anyway, I’ve recently finished two books “for the layman” about music and science:  Healing at the Speed of Sound, by Don Campbell and Alex Doman, and The Power of Music by Elena Mannes.  Although the two books cover very similar topics, they are very different.

When I read an interview with the Don Campbell on, I immediately requested Healing at the Speed of Sound from the library.  The book purports to discuss how pervasive music has become in our society, especially through the use of ipods, and the effect that this is having on our brains.  Sounds interesting, no?  Unfortunately, the book did not live up to its potential.

I was a more than a little disappointed by the fluffiness of this book.  In part, it tries to be a self-help book, starting out with recommendations for music to start the day with.  Depending on how easy it is for you to wake up, the book recommends nature sounds, classical music, or rock ‘n’ roll.  And it goes on from there, discussing the whole soundtrack of your day.  Music to listen to in the car on the way to work, music to listen to at work.  Listen to Bach to increase your creativity, listen to driving rock while you work out to keep you energized.  You get the idea, I’m sure.  I was unimpressed. I can find my own playlists, thank you.

Later chapters read like an infomercial for the benefits of music.  This is not a story or a serious discussion, it’s more like a list of all the great things music can do for you.  Music is amazing!  Music can heal!  Music can make you smart!  There is very little detail about the research.  Although some of the claims the book makes are intriguing, I found that most of the references are newspaper or on-line new articles rather than original research or conversations with scientists, leaving me unsure what to believe.  Many of the topics are ones I am familiar with, so I have an idea of what research has been done, but the book didn’t offer me any of the caveats of the research, or tell me that some of the conclusions are only tentative.  And the repeated warnings throughout the book to turn down our music so as not to harm our ears left me feeling like this book was written by someone’s grandpa.  As my kids would say, “I know, that; I’m not a dumb-head!”

In contrast to Campbell, Elena Mannes describes specific experiments and has interviewed a number of scientists and musicians, mostly as preparation for her documentary The Music Instinct:  Science and Song, upon which the book is based.  This lends credibility to her writing; her fluid story-telling prose helps too.  She loses some of that credibility (to my mind) in the chapter “Music of the Spheres”, where she compares music to the vibrations of the stars and planets.  I was also bored by the discussions of whether birdsong and whalesong should be considered music – it’s a little too “if a tree falls in a forest…” for my liking.  Still, her explanations of the role and potential roles of music in healing are much more scientifically based than Campbell’s, and though she discusses therapies that are outside of the mainstream, such as psychoacoustic therapy, and brain-wave entrainment, she at least admits that the scientific support for these areas is weak, and the jury is still out on their usefulness.  Overall, this book is worth reading, although I suspect watching the documentary would be a lot more entertaining.

On my bookshelf right now is Thinking Fast and Slow by Daniel Kahneman.  It’s not music-related, and I’ll admit that I’m not very far into the book yet.  But so far, it’s a fascinating account of how we use two very different systems for thinking – one automatic, quick and often highly biased, and a second one that is effortful, slow, calculating, and also very lazy.  Kahneman, a renowned psychologist and Nobel laureate who has written a number of books, has a style that is dense but gripping.  The book is full of quick self-tests to demonstrate the fallibility of our reasoning systems, followed by engaging explanations.

And here are my future reading plans:

I’m itching to read Michael Gazzaniga’s Who’s in Charge? Free will and the Science of the Brain, as soon as it actually gets onto the Vancouver Public Library bookshelves; it’s still listed as “on order”.  It’s based on an idea that I’ve been interested in for some time:  our actions are all controlled by our brains, which are collections of cells, governed by the same deterministic and probabilistic rules as the rest of biology.  So when we choose to do something, isn’t it really just our neurons responding to the action potentials of other neurons, which are triggered somewhere upstream by external stimuli?  Where does free will come into this?  Does it actually exist or is it a figment of our lively imaginations?  My interest in this topic was stimulated by another book, The Mind and the Brain by Jeffrey Schwartz, who brings quantum mechanics into the mix.  I’m not sure what I believe about all this, but I definitely want to read more.

Also on my library hold list is Music Cognition: A Science of Listening by Henjan Honing, a professor of music cognition at the University of Amsterdam.  The book sounds promising, even though it is written for a general audience.  I’ll keep you posted.