Sleep and Memory Consolidation

One year, when I was a teenager, my high school put on the musical Cabaret and I was volunteered into the role of pianist for the show.  I was handed a huge binder containing a piano reduction of the orchestral score and told to go to it.  I accompanied rehearsals, helped weak singers learn their parts, and played in the pit orchestra.  And what was extremely clear to me was that I was not actually a good enough pianist for the job.  That music was hard!  I wasn’t experienced enough to be able to figure out what bits to play and what to leave out, and so struggled along trying to play every single note.

At the end of each day, I would fall into bed exhausted, and every night I would dream I was playing the piano.  Not just some random dream where I was playing: in every dream I was sitting at the piano, looking at the music for Cabaret, actually expending mental energy to play the notes in front of me. 

Sleep and Learning

I hadn’t thought much about that time-period in years, but when I read a recent paper, it all came back to me.  The paper, written by Simmons and Duke in 2006, looked at the role of sleep in solidifying memories.  The researchers taught a simple piano melody to two groups of volunteers.  One group learned the melody in the morning, and was tested to see how well they played it in the evening.  The second group learned the melody in the evening and was tested in the morning.  The results were striking:  the second group (tested in the morning) showed a huge improvement in performance compared to the night before, while the first group showed no improvement.  The only real difference between the groups was that the second group had had a night’s sleep. 

Data from Simmons and Duke (2006).  PM/AM refers to the group that practiced in the evening and was retested in the morning (after sleep).  AM/PM refers to the group that practiced in the morning and was retested in the evening.

Sleep enhances memory consolidation

This study is just one of a number of recent studies showing the role of sleep in memory consolidation.  Consolidation is an important stage of memory formation, in which recently formed memory traces are made resistant to interference, strengthened, and anatomically rearranged in the brain.  Studies have shown that for motor memory, there is definitely consolidation happening during non-REM sleep (the deep sleep when we’re not dreaming).  Sleep deprivation blocks memory consolidation and so sleep-deprived people do not show an improvement in performance the next day.

Role of REM vs. non-REM sleep

But let’s step back a second… I just said that consolidation definitely happens during non-REM sleep, when we’re not dreaming.  But at the beginning of this post I was talking about how much I was dreaming about piano practice.  So what’s the relationship?  Well, non-REM sleep has been shown to be critical for motor sequence learning, when we’re learning a pattern of movements, like learning a particular new song.  So sure, when I was struggling with the piano score for Cabaret, I was learning new songs, and that required non-REM sleep.  But there was so much new music that I was also learning how to sight-read better.  That’s not a new motor memory, that’s a new skill.  Skill learning like that is a form of sensorimotor learning, where we have to take sensory input and translate it into the appropriate motor response.  And that type of learning requires REM sleep, the sleep in which we're dreaming.  When people are intensely learning a new skill or immersed in a new language, the proportion of time they spend in REM sleep has been shown to increase.  And depriving people specifically of REM sleep inhibits their learning.

Playing difficult music day in and day out, like I did with Cabaret, is an intensive learning situation, and probably the repeated dreams I experienced were related to memory consolidation during REM sleep.  I literally was practicing in my sleep.

All of this is to say that sleep (both REM and non-REM) is absolutely critical for stabilizing and improving memories.  Just by sleeping, you can improve your music performance. Getting a good night’s sleep on a regular basis is probably the best thing you can do to improve your ability to learn.

Deliberate Practice and the Role of Attention

Deliberate Practice is most effective

Thinking about playing music while you’re playing music is a good start, but what really makes practice most effective is to always be striving for improvement.  Our goal should be that each practice session consists of deliberate practice.  Deliberate practice is an idea from K. Anders Ericsson, a professor at Florida State University, who studies how people become experts at something: 

In contrast to play, deliberate practice is a highly structured activity, the explicit goal of which is to improve performance. Specific tasks are invented to overcome weaknesses, and performance is carefully monitored to provide cues for ways to improve it further. We claim that deliberate practice requires effort and is not inherently enjoyable. Individuals are motivated to practice because practice improves performance.

                        -Ericsson, Krampe & Tesch-Romer (1993)

This idea of deliberate practice is relevant because sometimes students (or their parents) assume that it would be better just to play for fun most of the time, instead of spending time practicing seriously.  A parent might say to me, “Little Matilda is so musical and just loves playing the piano, so I don’t want to squelch her enthusiasm by making her play scales”. And of course enjoyment of playing is important, and there is value in spending time improvising or just “noodling” on the piano, but if students really want to improve as pianists, deliberate practice is the most effective way. 

Playing “just for fun” does not make you a better musician

A 1996 study by John Sloboda and his colleagues compared five groups of children who were or had been taking music lessons.  The highest-achieving group consisted of children who were at a special music school, the second group included children who had applied but not been admitted to music school, the third group was children who had considered applying to the school, the fourth group were “regular” children taking music lessons, and the lowest-achieving group was children who had played an instrument in the past but had quit.  When Sloboda looked at how much time children in the different groups had spent practicing, there was a clear correlation between the hours they had practiced over their lifetime, and the quality of their playing.  By age 13, children in the high-achieving group had practiced twice as many hours as the children in the second group, and more than five times as many hours as the children who subsequently gave up playing their instrument.  

 

From Sloboda et al. (1996)

There is a very clear correlation between hours spent in serious practice and the level of playing.  In contrast, when Sloboda looked at how much time had been spent playing “for fun”, there was no correlation.  The high-achieving group did not spend significantly more time playing previously-learned pieces, improvising, or fooling around on their instruments.  It was hard practice, with their attention focused on improving their playing, that led those children to be better musicians, not playing for the sheer enjoyment of it.

Why is deliberate practice better than just playing music?

You might think (and we often do think this) that simple repetition is enough to make us learn.  If you play a tune enough times, you’ll eventually know it well, whether or not you were paying full attention while you were playing it, right?  Well, yes and no.  You can learn something up to a certain level of proficiency without giving your full attention, but it takes a lot longer, and you won’t master it to the same level you could if you were really focusing on learning. 

The difference at the level of the brain seems to be based on the neurotransmitter acetylcholine, which plays an important role in learning.  When we are paying attention to something, a part of the brain called the basal forebrain is activated, and releases acetylcholine into the appropriate parts of the brain for learning that particular task.  Neurons that are active and have acetylcholine released onto them will be more plastic and this helps us lock in our memories of whatever we’re paying attention to.

Deliberate practice, that is, paying attention to your practice and focusing on improvement, is the most effective way of learning to play music.

References:

Ericsson KA, Krampe RT  Tesch-Römer C (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363-406.

Sloboda JA, Davidson JW, Howe MJA, Moore DG (1996).  The role of practice in the development of performing musicicans British Journal of Psychology 87:287-309.

Explicit and Implicit Memory in Learning Music

Welcome to “Training the Musical Brain”, a blog combining my interests in neuroscience and music pedagogy.  I believe that neuroscience, psychology and related areas of research have a lot to tell us about the affects of music on the brain, and about the best ways to optimize our musical training. 

I’m going to start off the blog by giving a summary of a talk I gave this fall at the Music for Young Children (MYC) Conference, in Princeton, B.C.  The topic was “The Neuroscience of Practicing”, a subject dear to my heart because my passion for neuroscience began with an interest in memory and learning, and that lead me to pursue a degree in neuroscience.  Because the talk was long, I’ll split it into a number of blog posts.

Practicing is the way we learn, the way we store facts, events or skills in memory.  You might think that your music teacher is doing all the teaching, and that the learning that you do is passive (you just sit there and absorb what she says, right?) but it doesn’t work that way.  Learning is an active process on the part of the learner, and the teacher’s main job is as a guide.  It is during your practice time that most of the learning happens.

There are two main types of memory systems:  explicit memory and implicit memory, and studying music uses both of them.  Explicit memory is memory for facts and events:  when we learn that the key signature of G major is F#, this is stored in explicit memory.  Implicit memory, on the other hand, is the memory for skills and habits.  We use a lot of implicit memory in playing the piano.  When we learn a new scale or a new piece of music, we practice it until we can play it without too much conscious effort.  We don’t have to think about every single note every time we play it.  Implicit memory is sometimes called motor memory or muscle memory.  I don’t like this last term because it implies that the memories are stored in your muscles, which they are certainly not.  Motor memory involves a part of the brain called the cerebellum, which has a role in linking sensory input to motor commands.  In this way, the cerebellum takes the sensory feedback that you get from playing the piano (like the way the piano keys feel, how you have just moved your fingers and arms, and the sound that you have just produced) and links it to the motor command that was just sent out to the muscles.  It also links it to the next command that will be sent, so that a whole sequence of motor commands and sensory inputs is linked together in memory.

When we perform music, we need to use both implicit memory and explicit memory.  Many pianists have had the experience of playing the piano using implicit memory only; it seems like the hands can play the piece without any input from the brain!  That’s actually not true: unconscious parts of the brain like the cerebellum and basal ganglia are telling the motor cortex what commands to send to the muscles, but it feels like the brain is not involved because these are unconscious processes.  The downside of this automatic type of playing is that often pianists find that if they start thinking about what they’re playing, they make a mistake and are unable to continue playing the song.  What’s happening here is that the conscious parts of our brain are sending commands to the motor cortex that interfere with the commands coming from the cerebellum, and so we get mixed up.  The solution to this problem is that we should not let the performance of a piece get too automatic.  How can we accomplish this?  The best way is to form explicit memories of the piece alongside the implicit memories.  For instance, you could analyze the chord structure of the piece and memorize that, so you would know what chord you should be playing at each moment.  Or you could form an explicit memory of what notes you should be playing at the beginning of every fourth bar (or each phrase).  Or whatever works for you.  The key thing is to at least have some explicit memory of the song, even if it is just every few bars, that way if you lose your automatic train of thought, you have an explicit landmark to go back to, so you can get back into the song and continue playing.

This was my first main point of the talk:  When learning music, it is important to form explicit memories along with implicit memories.