Alex's Notes on Music Education

I'm an engineer who is learning about music education. These are my notes and thoughts. I'm also building a musical training dojo. Check it out at http://www.musikata.com .

[music-ed-notes-blog]

Less is More: Multimedia Design Principles

Recently I read chapter 12 of the Cambridge Handbook of Multimedia Learning, "Principles for Reducing Extraneous Processing in Multimedia Learning: Coherence, Signaling, Redundancy, Spatial Contiguity, and Temporal Contiguity", by Richard Mayer and Logan Fiorella.

The general theme of this chapter is that less is better. The chapter presents research on how people interpret multimedia, and discusses how to design multimedia in order to avoid information overload.

Extraneous Overload

The authors define information overload in this context as extraneous overload. Extraneous overload is the overloading of a person's working memory. It happens when the amount of brain power needed to understand a concept (essential cognition), and the the amount of brain power needed to process a set of multimedia elements (external load) exceed that person's working memory capacity. The result is extraneous overload.

I think of it through the analogy of a bath tub. You could think of your memory as a bathtub. It has a certain size, and it can only hold so much water. The things you want to learn, the concepts, are the water. The media that present those concepts are also water. If you put too much water in the tub, it overflows and you lose information.

5 Principles

The chapter is organized around 5 principles:

The Coherence Principle: In general, showing only relevant information produces better learning. Too much irrelevant information can overload our working memories.

The Signaling Principle: Cues such as pointer arrows or highlighting can help people learn better, by helping them pick out relevant information.

The Redundancy Principle: It is generally better to present text in one form, rather than as both written and spoken text.

The Spatial Contiguity Principle: It is generally more effective to put written text near or within a related image, rather than far away.

The Temporal Contiguity Principle: It is generally better to present chunks of related information with less time in between.

The chapter presents research on each of these principles, and discusses how they relate to extraneous overload.

Caveats

The authors are careful to point out that these principles should be taken lightly. Much of the research behind these principles is still preliminary, and has only focused on a few specific media instances. Media varies widely. In addition, much is still unknown about how these principles relate to other factors in learning, like engagement.

Designing Multimedia for Music Education Multimedia

Despite these caveats, these principles can still help us think about how to make effective multimedia for music education.

For example, let's say we are presenting a diagram to illustrate the parts of the musical staff. The spatial contiguity principle suggests that it is better to place labels near their corresponding elements, rather than far away.

So that instead of a figure like this...

...we should use one like this instead:

I'm interested in reading more about multimedia design research, and in seeing  how it can inform music education.

[music-ed-notes-blog]

Multimedia Design for Music Education

How do we design online media that help people learn music?

To answer this question we need to consider many factors, including how our minds process multimedia. Do we learn better with pictures? Is it more effective to use written text, or spoken text? Does it help to have animations?

Questions like these are the focus of one of my new favorite books, "The Cambridge Handbook of Multimedia Learning", edited by Richard Mayer.

The Handbook consists of several chapters, each of which focuses on a particular aspect of multimedia cognition. For example, there is a chapter on the 'Split-Attention Principle', a principle that suggests that combining text and illustrations can sometimes overwhelm our visual processing systems.

In general the Handbook is wide-ranging, readable, and fascinating. In addition, it provides useful advice for multimedia design, and frameworks for evaluating multimedia. To my mind it is essential reading for anyone who produces media, or who wants to know more about how we learn.

In the future I plan to write more about what I learn from The Handbook, and how it can inform the design of multimedia for music education.

[music-ed-notes-blog]

Virtual Sensei Sketch: Song Trainer

Here is another iteration on the virtual tutor idea.

This version is a (very) rough sketch for a sensei that teaches you a song:

https://drive.google.com/file/d/0B7PuM1d5UOG8OXJWd0hyN3BvVHc/view

It’s important to learn songs, because songs give you a musical vocabulary, a gut feeling for music. This vocabulary helps you understand musical concepts later.

This draft is playing with a few ideas:

jazzing up the pace, with some background music

trying to make the sensei a little more fun

experimenting with basic animations

If you have any thoughts or comments, please do let me know. I’d love to hear what you think. Thanks for checking it out!

Media Credits: Background Music from Konami’s “Legend of the Mystical Ninja”, song from the “Jump Right In for Recorder” music series, from GIA press.

[music-ed-notes-blog]

Thoughts on the Suzuki Triangle

Edit: Updated on 2015-06-01, based on comments from A.T.

Recently I met AT, a Suzuki piano teacher here in Camberville. She told me about the Suzuki teacher-parent-student triangle.

I'm writing about the triangle because it can inform how we teach music online.

What Is the Triangle?

The triangle is a set of jobs and roles. It breaks down the process of learning into jobs like 'taking notes during a lesson', or 'playing a recording'. It then assigns these jobs to three roles:

The Teacher

The Parent

The Student

Each role has different strengths, weaknesses, and requirements. The triangle assigns jobs to the most appropriate role. For example, the job of 'daily coach'* ‘cheering squad’ goes to the parent rather than the teacher, because the parent sees the student every day.

* A.T. correctly points out that ‘coach’ is not really the right word. The parent should not be in the habit of correcting her child’s playing, because this constant correction can make the child resistant to practicing. In addition, often the parent lacks the expertise to know how to give the proper feedback.

The Role of the Parent

I want to highlight the role of the parent, because to my mind, it is what makes the triangle special.

I should first point out that the parent is an active participant in the learning process. She acts as a scribe, coordinator, scheduler, and coach cheering squad. She does things like help the student play recordings, or keep track of the assignments the teacher gives. These things may seem small, but they can have a big effect. Many younger students struggle with discipline, and the very youngest students can't take notes yet. The parent actively helps with these jobs during lessons, and between lessons.

Contrast the parent's active role in the triangle to the passive role that parents typically take in private lessons (in the USA at least). In most private lessons, the parent leaves the student with the teacher, and then leaves or sits in the corner. At the end of the lessons the parent comes back, and maybe gets a few notes from the teacher about what piece the student should practice. And that's it. The parent is not a part of the lessons. The parent does not study the music with the student. Furthermore, the parent often does not really understand what the student should be doing, how the student should be learning.

The triangle explicitly involves the parent, and this brings many benefits. One of the benefits is that the parent can ease the load on the student and teacher. They don't have to worry as much about the little things, and can focus on listening, practicing, and teaching.

Another benefit is that everyone becomes more aware of the mechanics of learning. The process of learning can sometimes seem mysterious and overwhelming, especially for beginners. But when we start to think about learning as a set of tasks and relationships, it becomes feasible, it seems like something we can do. When we involve the parent, we have to communicate what happens during learning. And this helps all of us understand it better.

But perhaps the most important benefit is that the parent has deeper relationships with the student and teacher. The student now has a regular partner in her journey, someone to talk with about music, someone to listen with. The teacher now has an co-teacher observer, someone to talk with about how the student is doing. someone to talk with about practice plans.

How the Triangle Can Inform Online Education

An understanding of the triangle can help us design better tools for online music education.

The triangle helps us see what jobs need to be done, what roles are needed. We can then build tools that help people with these jobs. For example, the triangle suggests that making a simple 'listening log' could be useful. The teacher could load up the listening log with the student's songs for the week, and give it to the parent. The parent and student could then make an entry together in the log each time the student listens. This way, all three people would be able to keep track of how much the student is listening. And it would help the student see that he is making progress.

The triangle also help us understand what tools we should not build. When we think about the different strengths and weaknesses of the roles of the triangle, we can see where technology and media may be insufficient. For example, one of the teacher's jobs is to give specific, targeted feedback during lessons. It would be very hard (perhaps impossible) to design media that can give this kind of feedback. So we should instead leave this to the teacher, or design tools to support the teacher.

Other Resources

These are just a few of my thoughts about the triangle. If you are interested in learning more, I highly recommend checking out some other resources.

Here are a few:

http://www.philasuzukipiano.com/2011/Parent_expectations__The_Suzuki_Triangle.html

http://www.suzukiviolinsb.com/article/4230-on-the-suzuki-triangle

Better yet, find a local Suzuki teacher!

[music-ed-notes-blog]

Lessons from SimCalc

Recently I have been exploring MIT's course 11.132x, "Design and Development of Educational Technology" on EdX.

Today I watched the videos in which Drs. Erik Klopfer and Jeremy Roschelle talk about the development of SimCalc, a system that teaches math concepts through graphs.

Dr. Roschelle talks about the concept of SimCalc, but also about the development process, and what he thinks made it work.

A few things jump out me.

Multiple Representations

One is related to content. Dr. Roschelle mentioned a theory that multiple representations can help more people learn. For example, presenting a math concept as an animation and a graph can be more effective than presenting it as a graph alone.

I wonder if multiple representations can also help people learn music. For example, in presenting the concept of articulation, we could represent articulations with pictures. Long, smooth articulations would be depicted as long, abutting rectangles. Short articulations would be depicted as short, separated rectangles.

On one hand, it might help people understand the concept of articulation. But on the other hand, it might not help them learn to hear articulation. They may get the symbolic concept, but not acquire the actual skill.

I don't know if learning the concept actually distracts from learning the skill. It's something I wrangle with.

Development Process: Team

Dr. Roschelle also spoke about the SimCalc team, the skills they brought. The initial SimCalc team had a domain expert with vision, an educational technologist, and a learning science researcher. He said that this team was crucial to making SimCalc work.

I am coming to learn that team is essential. And I'm not yet sure how to build teams for music education. This is something I need to learn.

Development Process: Change

Dr. Roschelle also mentions that SimCalc changed greatly over the course of its development. This is also something I wrangle with. I want to get things right early, but I need to accept that I won't. I need to learn to accept that things will change, and that we can't know everything in the beginning.

Distribution: Packaging

Dr. Roschelle spoke about how packaging and format helped SimCalc succeed. The SimCalc team made it easy for teachers to integrate into existing curricula. They made it a turn-key product, so that it would be easy to install and use.

And, crucially, they made it modular. So that teachers wouldn't have to throw out the existing curricula they knew. Instead, they could take bits and pieces from the SimCalc curriculum. So that they could integrate it gradually.

Gradual integration also helped SimCalc spread, because individual teachers could take little bits, test them, and then tell their colleagues about them.

I think of this as the bite-size sample model. You might not want to try a new food, so instead you just try a sample. You find out that it's good, so you bring back some samples for your friends.

[music-ed-notes-blog]

Communication Blockers in Music Education

Earlier I wrote about how music educators have few opportunities for casual communication. ([The Water Cooler Problem in Music Education])(http://music-ed-notes.tumblr.com/post/118942342048/the-water-cooler-problem-in-music-education)).

In this post I write about why this is the case. I'm thinking like an engineer. I want to understand the constraints, the blockers.

The biggest blocker is time. Most music teachers I know are busy. If they are full-time school teachers, they get up early for long teaching days, and then go home to recharge for the next day. Part-time teachers often have multiple jobs and teach on the side. Teachers in general do a lot of prep, and a lot of juggling. So for many teachers just finding time to do anything can be tricky.

Another blocker that relates to time is scheduling. Music teachers can be constrained by their schedules. Some teach after school, some teach at night, some teach on the weekends. But they often have fixed schedules with their students that can't be adjusted.

A third blocker is the effort needed to organize events. A good organizer (or team of organizers) needs to get a venue, spread the word, arrange for speakers. These things take time, organization, communication, and persistence. These things are not easy. And they are often thankless and invisible.

A fourth blocker is communication. Spreading the word is not a trivial task. In some ways, we have a bit of a chicken-and-egg problem. We want to help music educators get together, so that they can form stronger networks. But we need a network to get them together.

So with all these blockers, is it still possible get music educators together? I think so. In fact, I think it's very possible.

In a later post, I may write about some possible solutions.

[music-ed-notes-blog]

Constructionism in Music Education

Recently I learned about Papert's "Constructionism" theory in MIT's "Design and Development of Educational Technologies" EdX class.

As I understand it, the main idea of constructionism is that we learn through making things, through constructing. Sometimes we learn by making physical things. We develop a sense of mass and gravity when we build a tower out of books. We can also learn by making abstract things. When we write a story, we develop a feel for pacing and dialogue. We could not learn these things from passive reading or listening. Words alone are insufficient. We have to do things to understand.

Can the theory of constructionism apply to music education?

Do people learn music better when they can build music, play with it, experiment with it? Can we make tools that help people experiment with music?

I think the answer to all of these questions is 'yes'. I say this because there are many things in music which are difficult or even impossible to describe in words. Often the only way to understand something in music is to hear it or play it. There is no way to tell a beginning music student in words what the difference is between a major and minor chord. But when you play a chord, then it becomes clear. When you have the student play the chord herself it becomes even clearer.

I think digital tools can help people play with music. For example, there are online 'chord explorers' that let people play different chords with a click of button.

It's important to note that constructionism has limits, in music and in general. We can't learn everything from making or building. Sometimes it's better learn from rote memorization or drilling. For example, it's probably faster and more effective to learn multiplication tables through memorization. Similarly, sometimes it's best to learn a melody by rote first, before trying to understand it and build new melodies from it.

It's also important to realize that building effective tools is not easy. A good tool has to hit a sweet spot. It must be expressive enough to allow for a range of exploration. But it must not be too difficult for beginners, otherwise few people will use it. This is one of the problems with musical instruments: they are hard to control at first!

A good tool also needs to be integrated with guidance. A tool alone will be insufficient for most people. If we give an abacus to an average math student, it's unlikely that he'll learn multiplication. Instead, we need to combine the tool with content and coaching. We tell the student about multiplication, we do a few demonstrations on the abacus, we coach him on how to use it. Even better, we coach a group of students, and then have them coach each other.

Constructionist education can't do everything, and it's not easy to do well. But despite these limits and challenges, I think the theory of constructionism can give us useful ideas about how to learn music.

I want to explore ways in which it musical tools can be combined with virtual coaches. One idea is to have the coach introduce a new concept, and then use a tool to allow students to play with that concept. The coach says "Here's a toy, go play!"

[music-ed-notes-blog]

The Water Cooler Problem in Music Education

This post is about a theory I'll call "The Water Cooler Problem". It's mostly speculation about teachers in the U.S., based on just a few observations, not hard data. So read with skepticism. But I'm curious to see if it rings true.

Music educators rarely talk shop with each other. I call this the water cooler problem.

Many workplaces have a spot where people can bump each other. A spot like the water cooler:

When people bump into each other around the water cooler, they can jam. They can tell jokes, they can get support, they can ask each other how they're doing, they can tell each other about the great band that's playing tomorrow night. 

Ideas flow, relationships develop, and people have more fun together. You get people who feel better, and who do better work.

For a deeper dive into idea flow and networks, check out Alex Pentland’s “Social Physics”.

For many music educators there is no water cooler. They rarely get to bump into each other and jam.

Most private music teachers I have met say they don't talk with other teachers about the craft of teaching. If a teacher is lucky, maybe she can go to a workshop every few years.

The situation is similar for in-school music teachers. Usually the only time they get to meet is at the yearly all-state conference.

The water coolers are few and far between.

I think things would be better if there were more water coolers, if music educators got together more. 

I know things would be better for me at least. I'm an engineer who is interested in developing tools and services to help people learn music. But it's tricky to find time to jam with teachers and hear what they think. I've been lucky to meet many amazing teachers, and I'm always blown away by what they say. But those opportunities can be hard to come by.

I suspect that things would also be better for teachers. Currently music education networks are weaker than they should be because there is less communication. Ideas don't flow, mentor-apprentice relationships don't develop, support groups don't form, and people probably don't have as much fun. When we don't hang out, we miss out. But if we get together more, communication would improve. Networks would improve.

Why does the water cooler problem exist? Can we create more water coolers? I I may write some thoughts about these things in a future post.

For now I just want to put the water cooler problem out there, to see what people think.

But for the meantime, stay hydrated ;)

[music-ed-notes-blog]

A Sketch About Music Ed Collaborations

A kind of quasi-poem/essay sketch about music ed jamming.

People in Boston should talk about music education.

Not just music teachers and musicians, but also technologists and scientists. Designers and artists. You and me. We've got a lot happening in this town.

We should get together. Draw things, share things. Brainstorm, jam.

Some good things might happen if we get together.

We could share teaching tips and favorite tools. We could build cool new tools. We could tell each other about projects and gigs and grants.

We could learn, innovate, collaborate.

We could have a lot of fun.

All of this would help people who want to learn music. Which ultimately means there would be more people digging on music.

Which would be very good indeed.

What's not to like? Why aren't we doing this?

I have a few theories. Maybe I'll write about them in the near future.

[music-ed-notes-blog]

Virtual Musical Tutors

I am thinking about making a virtual musical tutor. Like Kurikko here:

The goal of the tutor is to help people build audiation skills. It would give people musical examples and help them practice.

The idea of a virtual tutor is not new. In fact, virtual tutors have already been used in many fields: http://en.wikipedia.org/wiki/Intelligent_tutoring_system

A virtual tutor can provide some useful features for learning music. I will list a few here.

One feature is sending practice reminders. For many people, the hardest part about practicing is just getting started. Sometimes all it takes is a nudge. A virtual tutor can give you that nudge in the form of a short email or phone ping.

A virtual tutor can also customize practice sessions for your current level. Many people struggle with knowing what to practice, and when to practice it. A virtual tutor can give you exercises at your current level, and can challenge you to practice new things.

A virtual tutor can also help you reflect on what you have learned. Logs and journals are great tools for learning because they make you integrate and summarize. A virtual tutor can help you keep practice logs, it can help you keep track of your progress.

So, what would this look like? Here is one (rough) idea: virtual tutor sketch .

It's important to note that there are many things a virtual musical tutor can't do. It's hard for a computer to give you good tips about musical expression, or to correct your posture, to help you relax. Perhaps most importantly, an virtual can't give you a real connection human connection. That can only come from a real in-the-flesh teacher.

Nevertheless I think that it's still worth developing a virtual tutor. There are many good things a virtual tutor can do, and perhaps it can inspire someone to eventually find a real tutor.

I'm working on a visual novel engine to help create a virtual musical tutor: https://github.com/musikata/musikata.kabuki, and a curriculum: https://github.com/musikata/musikata.curriculum .

If it sounds interesting to you, let me know here!

[music-ed-notes-blog]

How Cooking is Like Designing a Music Curriculum

Designing a curriculum is like designing a 12-course meal.

Imagine that you are chef, planning an elaborate dinner. You know the flavors and ingredients you want to present, the wines you want share. But how do you present them in a sequence that makes sense?

If you serve your spicy garlic dip too early, it will overwhelm delicate flavors later. If you start with bread, your guests fill up too soon. You need balance. You need sequencing. You need to know what your guests can handle, how their taste buds work.

Similarly, we also need balancing and sequencing when we plan a music curriculum. We need to understand how people learn music.

Fortunately, we don't have to start from scratch. Some research has already been done on how people learn music. Music Learning Theory (MLT), which I have written about earlier, provides us with information about how children learn music. It gives us hints about WHAT needs to be learned (the content), and WHEN to learn it (the sequence).It's like a recipe book that gives us tips on what ingredients to use, and when to serve them.

What MLT doesn't tell us is HOW to present content. MLT gives us a framework, but it's our job as curriculum designers to figure out how much to present, and how to present it. (For more on this idea, see Eric Bluestine's Book "The Ways Children Learn Music").

This process of figuring out what to present is not trivial. We need to balance constraints. We need to think like engineers. Or chefs.

We need to think about how we present the dishes to our patrons. What portion sizes do we use? What kinds of garnishes? How often should we throw in something spicy?

We need to think about our tools and technology. How many spoons will we need? How many frying pans will the kitchen need? Will there be enough time for things to bake in the oven?

We need to think about costs and prep time. Will saffron dishes be too expensive? Will it take too long to cut all the artichokes?

This is why curriculum design is both challenging, and interesting. It's a puzzle that's an art and a science. There's the puzzle of making everything fit together. And then there's the puzzle of presenting it in a way that's fun and beautiful.

So let's start cooking.

I'm thinking about online curriculum design here: https://github.com/musikata/musikata.curriculum. 

Do you know about any interesting curriculum design project? Let's jam on reddit. Or contact me here: http://music-ed-notes.tumblr.com/ask .

[music-ed-notes-blog]

MacGyver, Time, and Islands

Recently I attended the "Urban Issues Forum" at the Massachusetts All-State Music Educators Conference.

These are a few of my thoughts. They are likely somewhat naive, since I am not a music teacher myself. My apologies in advance.

I was struck by two things.

The first was that music teachers are the MacGyvers of teaching. They find ways to make things work with limited resources. Give 'em chewing gum and paper clips, and you might just end up with a jetpack.

Music teachers in general have low budgets and high workloads. Yet they make their music programs work by cobbling together partnerships and volunteers. I heard about how teachers partner with local universities and the Boston Symphony Orchestra (BSO) to bring in student volunteers and professional musicians as coaches. Music teachers find ways to get people involved.

The second thing that struck me was how infrequently music teachers get together. Three teachers I was sitting with told me they only meet at the All-State conference.

How could it be that we have scores of clever and hard-working people, all trying to solve similar problems, but who rarely get together to talk shop?!

Speaking as a software engineer, this is puzzling to me. Software engineers get together all the time to jam. For example, http://www.meetup.com/bostonpython/ .

So why is that there are so few music ed jams?

The lack of meetups certainly isn't for lack of desire. Many of the teachers I spoke with wanted to get together more.

Instead there are structural problems that prevent music teachers from meeting.

The biggest constraint is time.

Music teachers, almost universally, are maxed out. They're hustling to put together concerts, travel between schools, go to staff meetings, and to plan their classes. And many, many, other things. Somewhere in there are things like 'family’, 'friends', and 'sleep'.

In addition music teachers have many scheduling constraints. A music teacher simply can't rearrange her schedule in order to make it to a lunch meeting downtown. She needs to do things like find a sub for the period, and talk with her supervisor to get permission.

It may also be that music teachers, and people interested in music education in general, simply don't know what other folks are up. To my knowledge, there is no central "Boston Music Education Calendar". Message boards are starting to appear (like the UMEC facebook page!), but adoption is still low. Communication in general among the various islands of Boston music education seems to be low.

Time and communication are just some of the things that make it hard for music teachers to meet.

Hard, but not impossible.

Can we find ways to help music teachers get together more? I think we can. After all, these are the MacGyvers of the teaching world we’re talking about.

My personal interest in this is that I want to see more music education innovation happening in Boston. Geting people together helps make innovation happen.

I want to explore some solutions.

[music-ed-notes-blog]

Thumbs vs. Abs: An Analogy for Teaching Audiation Online

Which would you rather have, strong thumbs or strong abs?

If you want strong thumbs, then do this:

If you want strong abs, then do this:

The point is that virtual games won't build real muscles. If you want real abs, you must train with your real body.

Similarly, you can't learn audiation through virtual games alone. Audiation involves both the mind and the body. To learn to audiate you must physically sing and move, you can't just click and type.

It took me some time to realize this. It was only after I read about several music education pedagogies (Campbell & Scott-Kastner provide a nice overview), that I began to appreciate the importance of physical performance for learning audiation.

So if physical performance is essential, how do we develop online media that encourage physical performance?

I can think of two basic strategies.

Record Physical Performances

One strategy is to create media that can record and evaluate performances.

For example, we could build a web page that uses a phone's microphone to record a student singing. Like this page .

But there are several challenges that make recording physical performances difficult. One reason is that recording devices are not yet widely accessible. Many would-be students still do not have smart devices.

Another challenge is that it is technically difficult to develop effective interfaces for recording and evaluating live performances. Basic interfaces like tuners and pitch detectors have been developed, but they are often unreliable and do not work well when there is lots of background noise. If the technology is not seamless it can distract students prevent them from interacting musically.

In the future recording technologies and software will likely be more widespread and more reliable. But for the present it is difficult to create effective and accessible media that use these technologies.

Unrecorded Physical Performance

Another strategy is to use media to encourage physical performance, but to not record it.

In this strategy, online media exist purely as presentational media. They provide examples and prompt performances. They do not record or evaluate performances. Students self-evaluate.

This strategy has several advantages. One is that it is much easier to implement. It does not require creating interfaces to sense and evaluate performances.

More importantly, the media does not distract a student as much from musical interaction. There is no mediated interface to get in the way of performing.

The main disadvantage of this strategy is that students do not get external feedback. I do not know yet whether a lack of feedback will prevent students from advancing.

Hybrid Approaches

There may be ways to combine the two strategies. For example, most interactions may be unrecorded. But occasionally, a student can choose to record her performance and submit it to a human mentor for evaluation. However, using human mentors may be prohibitively expensive if there are many students.

I recently found out that Frank York and Lyn Henderson of James Cook University have done work in this area, and I want to read more about it.

Want to jam more about this idea? Use the Reddit thread: http://www.reddit.com/r/MusicEd/comments/2zgjwq/thumbs_vs_abs_an_analogy_for_teaching_audiation/ .

[music-ed-notes-blog]

Audiation: The Meaning Behind the Sounds

Can we learn audiation online? I want to find out.

I recently read about Edwin Gordon’s Music Learning Theory (MLT), and wrote about it here. (Thanks to Libby & Stefani at Berklee for hipping me to this!). Audiation is on my mind.

Audiation: The Meaning Behind the Sounds 

I think of audiation as the process of understanding music.

Let me offer an analogy.

Imagine that you are hearing a poem in Japanese. If you don’t understand Japanese you just hear a series of sounds. There may be soft ‘shh’ sounds, or sharp ‘kuh’ sounds. The sounds themselves are interesting. But they don’t have any meaning for you. They’re hard to remember later, it’s hard to repeat them.

Some years go by and you decide to learn Japanese. You come across the poem again. But this time it’s different! You hear words, the poem puts images and thoughts into your head. You can remember the poem later and think about it. You can repeat it back to a friend.

Furu ike ya / kawazu tobikomu / mizu no oto (An old pond / a frog jumps in / the sound of water). Poem by Bashō, Portrait of Matsuo Bashō by Yokoi Kinkoku, c. 1820

Learning to audiate is like learning the language of music. When you can audiate you can hear the parts of music, you can understand the structure of music. You can learn to think in music and even compose music in your head.

This is valuable. Musicians use audiation to hear each other better and to work out their musical thoughts. Music fans can use audiation to hear songs on new levels, and to remember songs better.

Thinking as an engineer, audiation can help us design and analyze music. “Why does this piece make me feel a certain way? How do I write down this melody that I have in my head?”

Audiation can also be a great comfort. I audiate when I go to the dentist. When the dentist moves in with her drill, I play Beethoven symphonies in my head. The music in my mind blocks out the thoughts of the drill. It makes the experience better.

The term ‘audiation’ comes from Edwin Gordon’s Music Learning Theory (MLT). But its central ideas are also found in other active learning music pedagogies such as Suzuki, Dalcroze, and Kodaly.

I think audiation is important and I want to help people learn how to do it. But this can be a challenge.

Teaching Audiation 

People can learn to audiate in many ways. However, certain methods have been found to work well. Some of the better known methods are Gordon Music Learning Theory, Suzuki, Jacques-Dalcroze, Orff, and Kodály. These methods are sometimes called ‘active learning’ methods because they involve singing and moving.

These methods have been tested in the physical world, and they really do work.

So if they work, what’s the problem?

The problem is that these methods are not accessible to many people. They depend on having regular access to a live teacher. And many would-be learners are hesitant to try them because of fear, time, or expense. Or, would-be learners don’t even know about the existence of these methods.

It would be great if we could make these methods more accessible to people. Online media might be able to help us do it.

But this is a big might. Because there are challenges that we have to address. Here are just a few: 

Can we create online tools that people can interact with through singing and dancing? 

Can we create expressive virtual instruments that allow people to perform? 

Can we online media help learners understand the subtle nuances of music? 

Can we design media in ways that encourage beginners to get started?

Can we create media that complement what live teachers already do?

These are just a few of the challenges. I want to explore how we solve them.

What do you think? Know about any good resources to check out? I posted this on Reddit too for discussion: http://www.reddit.com/r/MusicEd/comments/2xx7vn/audiation_the_meaning_behind_the_sounds/

[music-ed-notes-blog]

Gordon MLT, Part VIII

Today I continued to read about Gordon MLT in the MLT guide: http://giml.org/docs/AboutMLT.pdf

This continues the reading I described in http://music-ed-notes.tumblr.com/post/112134436493/gordon-mlt-part-vii.

Instrumental Learning

Instrumental learning is like normal audiation learning with an extra step of learning to 'sing' through an instrument. Student still learn to audiate through singing first. But they also practice playing simple patterns through their instruments. So that they gradually learn to control their instruments like their voices.

The MLT guide gives three general principles for instrumental learning:

Learn songs by rote first. Students develop a feel for their instrument when they learn by rote, without thinking about notation.

Learn to play patterns, not individual notes.

Think in terms of solfege names, not note names.

Improvisation

The MLT guide offers some advice on how to learn improvisation, and states that sequencing is important.

It recommends this sequence for learning improvisation: 1. Sing first. 2. Learning tunes by ear. 3. Learning bass lines. A good improviser should be able to hear at least three tracks: the bass line, the original melody, and the new improvised melody 4. Start simple. 5. Focus on hearing. 6. Improvise with chunks of patterns.

General Thoughts

I like the MLT guide's approach to instrumental and improvisational learning. Both focus on making the instrument an extension of the player. So that playing the instrument becomes more like singing, rather than operating a machine.

It's interesting to think about how to teach instrumental playing through online technologies. Traditional instruments may be out of the scope of online learning.

However, there maybe potential to teach audation and instrumental learning through new online instruments. I've recently been reading about New Interfaces for Musical Expression (NIME). One idea couldbe to have everyone who studies through a particular online pedagogy learn a new online instrument. For example, a special phone app instrument. This would have the advantage of creating a shared context for everyone learning with the pedagogy, and provide distance from traditional instruments.

Sort of like going on a retreat. When you go to a new place, you think new ideas and you're not constrained by the rules of your old world.

[music-ed-notes-blog]

Gordon MLT, Part VII

I continued to read about Gordon MLT in the MLT guide: http://giml.org/docs/AboutMLT.pdf

This continues the reading I described in http://music-ed-notes.tumblr.com/post/112046667708/gordon-mlt-part-vi.

Classroom Activities

MLT does not prescribe full classroom lesson plans. It instead gives teachers knowledge and tools to complement what they already do.

MLT LSAs should only be a small part of what teachers do in class. The other parts of class should include exposing students to new musical content through listening, moving, dancing, chanting, and singing. The MLT guide refers to these activities as 'Classroom Activities'.

The guide gives some examples of how to coordinate MLT's general principles with classroom activities. For example, this is one possible sequence for learning a song by rote :

Listen to the teacher singing the song unaccompanied

Listen and move to the macrobeats

Listen and move to the microbeats

Listen and move to both macrobeats and microbeats together

Audiate the resting tone while listening

Audiate the entire song

Sing the song without accompaniment

Sing the song with accompaniment

The guide notes that teachers should sing for students, not with them. When teachers sing with students, the students can simply imitate without really audiating. Singing without guidance challenges them to audiate.

The guide also recommends teaching music before words, and also teaching students to hear bass lines.

Coordinating Learning Sequence Activities (LSAs) and Classroom Activities

The guide discusses how to coordinate LSAs and classroom activities.

One recommendation is to coordinate general content rather than specific patterns. Teachers shouldn't worry about tightly coupling specific patterns to specific songs. Instead they should coordinate general types of patterns and skills. So instead of thinking 'Let's do pattern #827, da-da-dum, because we're doing a song that has da-da-dum in it', we should think 'We're doing a song in duple meter, let's do a few patterns that use duple meter.'

It also recommends that new skills appear only in LSAs, and that new content only appears in classroom activities. I like this because it can use LSAs to help build metacognitive awareness. "We're learning right now, and we know it.".

My General Thoughts

The sections I read today helped me understand more about how MLT's LSAs work in a classroom setting.

I think of the relationship between LSAs and classroom activities through the analogy of behind-the-scenes movie clip. LSAs ar like a short 'behind-the-scenes' section that you might see at the end of a movie. The classroom activities are like the movies themselves. The short clips help you understand what you have already seen in a new way, and they make you notice new things when you see other movies later. But these clips are not the movies themselves, and they are not interesting without the movies.

I'm also wondering how the MLT guide's coordination advice relates to online music education. One of the key differences between online and offline music ed is that there is no unmediated classroom experience in online teaching. That is, in online music ed the classroom activities either don't exist, or they happen through a medium which makes the experience less immediate. Does that change how we teach audiation skills online?

I think it must have some effect. In a classroom setting students can dance, move, chant, and sing. They get immediate, nuanced feedback. They can see and hear a live teacher and see all the subtle things she does. I don't think that media in the present or the near future can reproduce those experiences.

But are there other experiences that mexia can provide? For example, we can teach songs as online musical quest games. Playing these games would offer a different experience from a live classroom. Some aspects of these games might not be as good as classroom learning. But other aspects could be good.

This is what I want to explore.

I think MLT has some good ideas, which have been proven and tested in classroom teaching. I want to see whether they can be adapted to online teaching.

[music-ed-notes-blog]

Gordon MLT, Part VI

Today I continued to read about Gordon MLT in the MLT guide: http://giml.org/docs/AboutMLT.pdf

This continues my reading from http://music-ed-notes.tumblr.com/post/111953698823/gordon-mlt-part-v.

Tonal Content

Tonal activities in MLT are aimed at developing a sense of tonality and building a vocabulary of tonal patterns. MLT organizes patterns by tone type (major, minor, etc.), and pattern function (tonic, subdominant, etc.) . Patterns go in sequence of familiarity. This means starting with major tonality with tonic/subdomininant patterns for Western learners.

In MLT the word 'tonality' encompasses the idea of modes such as dorian and dorian and phyrigian. It's like the overall set of spices you choose to cook with.

The phrase 'resting tone' refers to the solfege syllable associated with a particular tonality. For example, 're' is associated with dorian tonality.

The word 'keyality' refers to pitch center. For example, a piece in Bb Major has a keyality of Bb and tonality of major.

Early tonal patterns are typically arpeggiated in MLT because larger intervals force students to audiate. With smaller whole/half-step intervals it is easy for students to simply imimate and not audiate.

Tonal solfege in MLT uses a movable do with a la based minor. This has the advantages of (1) associating a solfege syllable with each tonality, and (2) associaing intervals with pairs of syllables. For example, the interval beween do and mi is always a major third in this system.

Rhythm Content

Rhythm activities in MLT focus on developing a sense of meter, and a vocabulary of rhythm patterns. Patterns are organized by meter (duple, triple) and function (macro, micro, divisional).

As with tonal activities, rhythm activities are sequenced by familiarity. For Western learners this means focusing on duple meters first.

Elements of Rhythm in MLT

MLT describes rhythm in terms of macrobeats, microbeats, and melodic rhythms. Students should audiate these elements together to establish the syntax of rhythm.

Macrobeats are what we perceive as the longest beat. I think of this as the beat we tap our feet to.

Microbeats are equal divisions of microbeats, usually in halves or thirds.

Melodic rhythm is the ongoing series of rhythm patterns in music. This part is not clear ot me yet.

There is some subjectivitiy in defining the macrobeat. Some people might hear the microbeats as macro, and vice versa. An interesting result from Gordon's research is that students with higher initial aptitude often use longer macrobeats. Interesting!

Meters

Meters describe how macrobeats are grouped and divided. Usual meter has macrobeats of equal length, usually in twos or threes. Unusual has macrobeats of unequal lengths, for example 5/8 time (1-2, 1-2-3).

Rhythm Function

MLT explicitly focuses on identifying rhythm elements through audiation rather than notation. _I especially like this focus. Beginners can get caught up in decoding rhyhthm notations and miss the sound and feel of a rhythm.

Rhythm functions progress in order of audiational difficulty, as tested through teaching experiments. This sequencing is in contrast to other methods that sequence by notation. The MLT guide notes that notation-first approaches can be counterproductive because longer notational rhythms like whole notes can be harder for beginners to audiate.

Rhythm patterns start from duple and triple meters, and then go into division patterns, elongation patterns, rest patterns, and tie patterns.

Rhythm solfege is based on how rhythm is audiated, not how it is notated. MLT uses Gordon solfege, which offers these featuers: (1) it is based on how rhythm is heard, (2) it is comprehensive and can uniquely describe rhythms.

My General Thoughts

I like many aspects of the MLT for teaching rhythm, especially the focus on hearing rhythms. I think it is easy for many beginners to get caught up in decoding rhythm notation, rather than hearing the rhythms themselves.

I also like hearing about how Gordon MLT has tested these theories, by doing actual experiments. I met a music ed. researcher here in Cambridge, who told me about friends of his who worked as research assistants for Gordon. He talked about how they would observe students and log their progress. This gives me more confidence in the method.