Many performers of music cannot read it. Okay. There are other, often more intuitive, ways to learn music.

Scientists perform stat models. Most scientists cannot read them. This is less OK, but there are other ways to learn models.

Short thread in which I strain this comparison

If you don't read music, the Rzewski excerpt above (left) is meaningless. If you do, it is perfectly clear. You'd read it not by each individual note, but through higher structure like chords & arpeggio patterns & progression.

It's not the notes so much as their relationships.

If you don't read math stats, the social network model above (right) is most meaningless. But again, when you do read these models, you read the model in chunks, through its grammar and phrasing.

It's not the variables so much as their relationships.

In both cases, the music and the model, the formal notation is not enough to perform the damn thing. This is perhaps obvious for the music. But for the model too, implementation requires more. Models are usually performed on the log probability scale, for starters.

When I teach models to colleagues, I rely too much on notation. I want to do better.

Maybe I could teach more through performance. We cannot hear the model sure. But we can see how changes in one variable cause changes in others.

At the same time, I wonder sometimes whether I should just pause and provide a primer in reading math stats notation. It's not any harder than reading music. But I have found that many scientists have math anxiety, where I suppose they do not have sheet music anxiety.

Anyway, this odd thread inspired by a short conversation yesterday with @danielj_redhead, who is always tolerant of my associative fugues.

Also @cantabile has infected me with the superposition of musical and mathematical pedagogy.