Thread: any p90xers here?

That might be true, the second bit of your post. But I don't think it's going to be anything amazing that you couldn't achieve with any other normal exercises.

I'm about to make a super geeky analogy: Training a neural network.

If you're teaching a neural net to do text recognition (one letter at a time, for simplicity), you need to feed it lots of samples with correct answers (e.g. "<sample writing> - this is an R").

As you feed it more and more samples, it will learn to recognize other stuff.

But there is a difference in what order you feed it the samples. If you have 100 samples of each letter, you want to mix it up. If you feed it 100 Rs, then 100 Bs, etc. it will grow differently than if you feed random letters in random order (2600 samples, in no particular order). It will be MUCH more effective and versatile if you trained it by mixing it up early on.

I think this is a really great analogy to learning, as well as physical fitness. It's not a "scientific proof" by any stretch, but it's a kind of reasoning why "mixing it up" is better than "doing the same thing over and over for a LONG time and then changing it and then doing the same thing over and over, etc."

But I think the way they talk about "muscle confusion theory" is misleading. Yes, it's good to mix it up, and I wouldn't be shocked if the way they change the workouts and ordering around DID make a huge difference. I DID find that after the first few weeks, when I switched over to the second set (which is just a replacement of the weight training modules, to cover different muscle groups together, with slightly different exercises), the workouts became a lot harder for a while (as was the case when switching back to the original model).

But it just comes down to switching it up a lot, rather than optimizing your body for one specific workout.

But I think it's more about balanced growth, than it is about "your muscles will get bigger and stronger faster if you confuse them" - THAT statement, I think is bs.

From Mark Rippetoe: "It works very well since it is most people’s first exposure to an exercise protocol that’s supposed to be hard, and the impression of most inexperienced people who have tried it is very positive. P90X works well for the same reason, as does HIT, Turbo-Jam, the first week of football practice, and all participation in the first phases of any reasonably challenging sport. A strenuous physical effort – no matter what it is – acts as a stimulus for adaptation, up until the point that the adaptation occurs and the program fails to further progressively load."

Originally Posted by Replicon

I'm about to make a super geeky analogy: Training a neural network.

If you're teaching a neural net to do text recognition (one letter at a time, for simplicity), you need to feed it lots of samples with correct answers (e.g. "<sample writing> - this is an R").

As you feed it more and more samples, it will learn to recognize other stuff.

But there is a difference in what order you feed it the samples. If you have 100 samples of each letter, you want to mix it up. If you feed it 100 Rs, then 100 Bs, etc. it will grow differently than if you feed random letters in random order (2600 samples, in no particular order). It will be MUCH more effective and versatile if you trained it by mixing it up early on.

I think this is a really great analogy to learning, as well as physical fitness. It's not a "scientific proof" by any stretch, but it's a kind of reasoning why "mixing it up" is better than "doing the same thing over and over for a LONG time and then changing it and then doing the same thing over and over, etc."

But I think the way they talk about "muscle confusion theory" is misleading. Yes, it's good to mix it up, and I wouldn't be shocked if the way they change the workouts and ordering around DID make a huge difference. I DID find that after the first few weeks, when I switched over to the second set (which is just a replacement of the weight training modules, to cover different muscle groups together, with slightly different exercises), the workouts became a lot harder for a while (as was the case when switching back to the original model).

But it just comes down to switching it up a lot, rather than optimizing your body for one specific workout.

But I think it's more about balanced growth, than it is about "your muscles will get bigger and stronger faster if you confuse them" - THAT statement, I think is bs.

I could agree with this.

Not that the neural net is analogous to muscles in every respect.