Turning Fight Reading into Sight Reading
I taught music for decades before I started coaching practice exclusively over 10 years ago.
In the 13 years prior I taught high school music to all levels in classes and rehearsals.
Underdeveloped, or non-existent, sight reading can exist on any instrument/voice, but is a particular problem on mine.
Well into my university performance degree I couldn’t fight read my way out of a wet paper bag.
Not that I didn’t want to get better at it, but the advice I heard from all kinds of people, including teachers and performers was, “Just do it.” I heard it so much I thought they worked for Nike.
Oh, and they would also say, “Use music that is several levels below your current ability.” Now how am I supposed to find that if I can’t read it well? 🙄
Shortly after I became a teacher I came up with a sight reading protocol for myself which worked.
I then turned my attention to designing a sight reading method for the high school classes I was teaching. There were none in existence for my instrument!
It worked year in and year out with hundreds of students.
It was then I first applied CLRP1, but I had no idea what that was all about yet. My first contact with cognitive science was years away, and I was just beginning to figure out the real principles of learning as I grew in my teaching.
S-L-O-W-L-Y V-E-E-E-R-R-R-Y-Y S-L-O-W-L-Y
If only someone, somewhere had told me what was available in the learning science. If you are anything like me then finding real, actionable advice out there about the science is hard.
I was able to develop a technical protocol for sight reading instruction from the beginning and moving forward. In my opinion, a significant reason for the disparity in sight reading ability out there, even among players who are otherwise similarly skilled, is the lack of such a protocol.
There is a skill underneath the skill that can, and should, be taught in and of itself from the first stages that I’ve never heard addressed specifically.
I was reminded of this recently, and was able to dig up some interesting things on sight reading that should help us understand it better.
More than a little research
Have you heard of Noa Kageyama, the Bulletproof Musician? If you have not, you should. His work can be a good jumping off point for further study.
One of his weekly emails, highlighting some research on sight reading, put me through the learning process of teaching sight reading again. I got the same outcome, but this time more quickly and with better and more detailed information.
Let’s see what he found out about that mysterious black hole – learning to sight read.
The blog post focuses on a meta-study.
A meta-study2 is one in which an expert reviews other studies on a particular area, and looks for consensus and other insights. They are a great place to start for research.
“An exhaustive survey of the available research literature was conducted resulting in 92 research studies that reported correlations between sight-reading and another variable.” (Mishra 2014)
According to the blog it, “sheds some light on (a) the age-old debate about whether sight reading is an innate ability or if it is a skill that can be developed, and (b) what factors may be related to being good at sight-reading.” (Kageyama, 2024)
You probably already know my thoughts on that – if it is a skill it can be developed to any level one would like with the proper kind of work.
This is where things get a little sideways. I think we are looking a little too far downstream to solve the sight reading mystery.
What did they find?
There were 17 general areas identified (You can read all about them in the Bulletproof blog). The top four constructs were, “…ability to improvise, ear-training ability, technical ability, and music knowledge” (Mishra, 2014)
The good news is that all of these are learnable skills. The bad news is that is the end of the information road. What do we do with it?
More, please
It takes time to learn those skills. Not a week or a month, but a lot of time until we move from studying them to being able do do them competently.
Were all of those who were good at sight reading, as well as the other areas, exceptionally advanced in improvisation, ear training, technical ability and musical knowledge as beginners, or any time close to that? That data was not reported, but I suspect not.
What role would those constructs play if we began teaching sight reading during the first phases of development, while we are still slowly learning them?
Something doesn’t seem right with all of this. It does not seem to make sense with what I know about the learning process as found in cognitive science.
Why aren’t we teaching sight reading as a technique from the ground up like we do with note reading, theory, ear training? Do we have to wait for other ancillary skills to be learned?
Can only those who have developed these skills hope to sight read?
What, and how much, should we do of these things to teach the specific skill of sight reading?
The article tells us – they don’t know.
At the same time the general public takes in this information as beginners to the science and can form haphazard ideas how this stuff works.
I suspect studies about scale playing would show similar things. Do those who play scales well in their repertoire show greater proficiency in music knowledge etc.? My teaching experience has shown me those areas do affect our skill in any musical endeavor. Further, anyone who has taken their overall music study seriously enough to get good at those skills is probably practicing a good amount.
I don’t know anyone who would look to that and decide to teach scale proficiency by getting good at ear training or theory.
They’d want you to practice the technique of scales in which the most basic elements of the skill can be addressed. This would reduce the cognitive load to the basic skill.
We have stand-alone technique practice for scales and a lot of other technical components of music making, but not sight reading.
Why?
More importantly how would that ‘magical’ gateway work?
There are answers.
We’ll look at some more research you may not have encountered in Part 2.
Gregg
P.S. here are some more quotes from the study with brief thoughts added.
“Musicians differ in their ability to sight-read, and hundreds of researchers have explored reasons why this might be the case,” p. 453
We have no idea how to teach the technique of sight reading.
“In general, music constructs that can be improved with practice (i.e., ability to improvise, ear-training ability, technical ability, and music knowledge) correlated more strongly with sight-reading.” p. 460
I can’t imagine this is exclusive to sight reading.
“In a meta-analysis of experimental sight-reading research, Mishra (2013) found that the greatest improvements in sight-reading were associated with treatments focused on aural skill training or creative activities, such as improvisation.” P. 461
When does one acquire enough skill in those things to effect sight reading!
Kageyama, N. (2024). Are Great Sight-Readers Born or Made? Retrieved from https://bulletproofmusician.com/are-great-sight-readers-born-or-made/
Mishra, J., (2014). Factors Related to Sight-Reading Accuracy: A Meta-Analysis. The Journal of Research in Music Education, 61(4), 452-465.
1CLRP™, Cognitive Load Reduction Protocol™
The process I’ve developed to apply the learning science to develop instructional models that bring complex skills to an appropriate level to begin understanding and progress quickly. The resultant methods can be used by any level from beginner to virtuoso (and have).
2 A narrative synthesis of the various studies correlating sight-reading with other variables must conclude that results are contradictory, and the synthesist would be required to choose a handful of studies to represent the research as a whole. A meta- analysis, on the other hand, is a statistical procedure capable of synthesizing a large body of research (Borenstein & Hedges, 2009) and allows for the inclusion of all relevant studies in the analysis. This procedure allows the synthesist to look systematically at the body of research as a whole rather than at research findings in isolation to identify patterns that might not be apparent when examining only a handful of studies. A meta-analysis also weights the results based on population size, weighting the results of studies using larger populations more heavily than those with smaller populations. A meta-analysis also allows design elements of the primary studies (e.g., population characteristics) to be used as variables to investigate whether there are differences between studies with varying designs. An increasing number of meta-analyses have been conducted in the field of music, especially in music therapy (e.g., Standley, 1996, 2008), music psychology (Kämpfe, Sedlmeier, & Renkewitz, 2010) and have been used to investigate the effects of music instruction, especially as concerns the Mozart effect (e.g., Pietschnig, Voracek, & Formann, 2010).
(Mishra, 2014)
I’m eagerly awaiting your lightbulb masterclass on this subject. I’ve got so much out of the previous masterclasses, this will definitely be one not to miss.
I try to do some sight reading every day but I’ve not seen much improvement, so what I’m doing is not helping much if at all. My method is to take an easy few bars from a piece of music I’ve never played before. Look through it for 30 secs (that’s the time allowed in ABRSM exams) to check the key signature, time signature, tempo, style, possible era, musical patterns, bass and treble clef changes, rhythm, accidentals and 101 other things! Play it through once and once only, how are we supposed to learn a skill from that? I then take the liberty to play it through once more to address problems thinking it might help me in the future. I don’t have a problem reading notation but my brain and fingers are not capable of the cognitive load.
I need your help Gregg.
“Play it through once and once only, how are we supposed to learn a skill from that?”
Reminds me of something I wrote as part of a class description;
Practicing sight reading is weird because it is all retrieval.
Once we get a bit of a clue about practicing, and know we need to work on hard sections, we use different learning interventions like contextual interference. We spend a lot of time, and we should, on encoding then, if we know what we are doing, we work on retrieval which requires less time.
Sight reading is always retrieval. The encoding already happened elsewhere. You know where the notes are given enough time.
That is a retrieval issue.
If you’ve been told, like I was years ago, to learn to sight read by ‘just doing it’ they were not wrong.
That is kind of like telling someone to learn to bench press 300 pounds and then letting them loose in a gym. It would work…sometimes, but the person would have to figure out a lot of important details on their own.
It is no wonder that some think sight reading skill may be an inborn talent. Perhaps flown in from the Valley of Echos by Tinkerbell on a Unicorn sprinkling pixie dust over the lucky few.
We can instead use the massive amount of research on retrieval practice to design efficient protocols to get it done comprehensively, and fast.
Along the way I’ve designed some of those protocols based on that research. This is pretty unique stuff.