Tuesday, 22 November 2011

Mixing It Up (Part 2)

Imagine 41 beginner clarinet students, about 11 years old.  Their task:  learning three new short melodies over three days.  Half of the kids practice one melody every day, turning to a new melody the next day (“blocked practice”).  The other half of the kids practice all three melodies every day, in a random fashion.  The researcher records their practice sessions, and also has them come back 24 hours later to test how well they have learned the melodies.  What do you think?  Which group learns the melodies better?

Contextual Interference
As I discussed in my last post, studies have shown that practicing motor tasks in a blocked fashion leads to better performance during the practice sessions, compared to random practice.  However, retention is better when tasks are practiced in a random fashion.  What this means is that our long-term memory is better for tasks practiced in a random fashion.  This is known as the contextual interference effect.

However, most of the research that has looked at this effect is based on simple laboratory tasks.  In real life, tasks involve complex sensorimotor feedback and have multiple components that we have to control at the same time.  This is particularly true in music practice.  Practicing music involves visual cues from the music we’re reading, somatosensory feedback from the positions and actions of our joints and muscles, and auditory feedback from the sounds we’re creating.  We have to pay attention to pitch, rhythm, tone quality, evenness, dynamics and articulation.  In a complex task like this, does the contextual interference effect still apply?  Is it stronger or weaker?

The Clarinetist Study:  Stambaugh (2011)
For the clarinetists discussed above, the contextual interference effect seemed to be slightly weaker.  The researcher, Laura Stambaugh, found that both groups made about the same amount of mistakes, but the speed at which the two groups played the melodies differed.  During the first half of the trials of each melody, the group practicing in a blocked fashion was able to play faster than those practicing in a random fashion.  For the second half of the trials of each melody, the random group was able to play faster.  And when tested 24 hours after the end of practice, the random practice group played significantly faster than the blocked group.

From Stambaugh (2011). Journal of Research in Music Education 58:368

In other words, the blocked group initially seemed to learn better, since they were able to play faster, but the advantages of practicing in a random fashion became apparent even before the end of the practice sessions.  In the classic contextual interference studies, random practice led to worse performance throughout practice, so Stambaugh's result is a little different.  There are a couple of possibilities for this difference:  it could be just because the practice was split over three days, or it could be because in this study the participants are children rather than adults.  Other studies using children have found that the contextual interference effect is not necessarily the same for children. Some studies have found that children do not necessarily perform better while practicing in a blocked fashion compared to a random fashion.  It's a good reminder:  We need to be cautious in applying the results of psychological studies done on adults.  Because children’s brains are still developing, they may not operate in the same way, and so the lessons learned from adults may not apply.

Worth the Effort
In any case:  Stambaugh’s clarinetists confirm that mixing up the order of our practice sessions may be worth the effort.  I think this is particularly true for certain types of practice.  When students are preparing for exams and have to be able to play scales in a number of different keys, practicing these in a random fashion is definitely effective.  It can be hard to convince our brains to change between key signatures, so practicing making this switch would certainly help. 

When students play for me during lessons or classes, they (or their parents) will often say that they “played it better at home”.  And I’m sure it’s true.  Partly this is due to nerves at the lesson, but I also think a large part of the reason students play worse in class is because at home, they play the piece several times, not just once like they do in the lesson.  Probably the first time they play the piece in a practice session at home, it’s not a lot better than what they play for me.  By the third or fourth time playing the song through at home, it’s pretty good.  But at the lesson, they only get to play once.  And this is true for exams and performances as well.  If students practiced more in a random fashion, they would play better the first time.  Performance, whether at an exam, recital or a lesson, is always in a random fashion, and we need to be ready for that.

1 comment: