Why some women are not asked to dance

[noobster]

Re: Balance: I don't think i would have fit that data with a quadratic curve - the main feature seems to be heteroskedasticity (and having peak ability at 40 with 60 year olds performing the same as 20 year olds is pretty encouraging ). (I tend to think that balance is one of the things we just tend to neglect when exercising)

There's a quadratic (Figure 1) and a cubic (Figure 2) fit. The cubic is the less encouraging with imputed maximum about age 30.

Are we looking at the same curves? Figure 1 shows the worst performing 20 y/o still making it to almost a minute, where a fair number of sixty year olds didn't even make it to ten seconds. The curve is not at all symmetrical. Cutting off the top of the curve at 2 minutes also obscures the differences among the highest performers.

 

[Brummbaer]

I tend to think that balance is one of the things we just tend to neglect when exercising

Whilst I think it's relevant to just being 'fit' at whatever age you are, we know balance deteriorates with age, yet it appears that some of the 'great' milongueros were only truly esteemed when older. Maybe that's just a feature of when they were 're-discovered' during some AT renaissance. Or maybe we're considering a totally different style/type of social AT. Or maybe that's just 'feelgood' as we age. It's an interesting study, but I wouldn't get too hung up on it. We are where we are in our allotted timespan. Just don't become sedentary.

 

[Bailango]

Whilst I think it's relevant to just being 'fit' at whatever age you are, we know balance deteriorates with age, yet it appears that some of the 'great' milongueros were only truly esteemed when older. Maybe that's just a feature of when they were 're-discovered' during some AT renaissance. Or maybe we're considering a totally different style/type of social AT. Or maybe that's just 'feelgood' as we age. It's an interesting study, but I wouldn't get too hung up on it. We are where we are in our allotted timespan. Just don't become sedentary.

The last part is the key. As we age genetics to plays a large part. We see those in 80s and 90s that are very fit including dancing or running marathons or playing tennis. We also see those who get frail. The state of health plays a major role. I ponder whether the loss of balance is age related or is abated by loss of strength in muscles and bones.

 

[Brummbaer]

The state of health plays a major role. I ponder whether the loss of balance is age related or is abated by loss of strength in muscles and bones

Our lives are complex. You can't put it down to one single thing. Personally I'd be tempted to consider as well the aging of the balance sensors of the inner ear along with some expected high-frequency hearing loss; possibly exacerbated by noise during lifetime. But that's my bias. Being 'healthy' and generally 'fit' helps. So does having an interest in life, even if that only interest is AT.

Just take it that the survey shows general indicators of aging. The converse isn't true, i.e. correct those and you won't age.

 

[Gssh]

Are we looking at the same curves? Figure 1 shows the worst performing 20 y/o still making it to almost a minute, where a fair number of sixty year olds didn't even make it to ten seconds. The curve is not at all symmetrical. Cutting off the top of the curve at 2 minutes also obscures the differences among the highest performers.

Parabolas are per definition symmetrical . I personally think they should have fit a loess curve, especially as their sample is weighted heavily towards younger people and the floor and ceiling effects will make any normal theory argument anti-conservative. Normative modeling is hard, and there here is a reason that the people who have the most experience trying to apply global growth curves to individuals (pediatricians) have basically given up on parametrizing them and rely on empirical curves.
I think the take home message of these graphs is obscured by the ml fit lines: the variability between people of the same age is huge, and while there are limits, especially beyond 80 or so, we probably should think less about global trends but about sources of variability. I think it would have been nice if the authors had done some exploratory analyses linking residualized scores with socio-demographic and life-style factors. They don't even give a gender split, and considering how different men and women develop and age this bothers me a lot.

 

[TTanguero]

There's actually a trick to make that exercise easier - just imagine yourself holding onto something with one of your hands even with your eyes closed.

Wow! Just tried that, and it's so much easier!

 

[LadyLeader]

There's actually a trick to make that exercise easier - just imagine yourself holding onto something with one of your hands even with your eyes closed.

I have been wondering how totally blind people manage their balance. They must have overcapacity compared to me in the other balance producing systems compensating the lack of vision.

 

[LadyLeader]

To reach my enrosque goals I have been training my turns several days per week for the last years and that has surprisingly given me a much better balance and core too.

I have done regular pilates and other training for a long time without knowing that the turning function had that improving potential.

 

[noobster]

Whilst I think it's relevant to just being 'fit' at whatever age you are, we know balance deteriorates with age, yet it appears that some of the 'great' milongueros were only truly esteemed when older. Maybe that's just a feature of when they were 're-discovered' during some AT renaissance. Or maybe we're considering a totally different style/type of social AT. Or maybe that's just 'feelgood' as we age. It's an interesting study, but I wouldn't get too hung up on it. We are where we are in our allotted timespan. Just don't become sedentary.

I do think that age can bring some intangible benefits to the dance aesthetic as well. I have not seen too many young people in social tango for comparison, but in salsa I do notice that while the younger people have more superficially impressive acrobatics, older dancers often have the ability to load small movements with a subtle but very beautiful expressiveness and flavor.

I don't know if that's an effect of age and perspective per se, or more so a result of many accumulated years of experience and muscle memory.

 

[noobster]

Parabolas are per definition symmetrical . I personally think they should have fit a loess curve, especially as their sample is weighted heavily towards younger people and the floor and ceiling effects will make any normal theory argument anti-conservative.

Oh I see. Sure I agree. But I would note that a better fit to the data, which seem to show a steep upward slope on the left-hand side and a more gradual reduction on the right-hand side, would probably place the apex even earlier, maybe in the late teens or early 20s versus the 30s or higher.

I think the take home message of these graphs is obscured by the ml fit lines: the variability between people of the same age is huge, and while there are limits, especially beyond 80 or so, we probably should think less about global trends but about sources of variability. I think it would have been nice if the authors had done some exploratory analyses linking residualized scores with socio-demographic and life-style factors. They don't even give a gender split, and considering how different men and women develop and age this bothers me a lot.

The sociodemographic bits may not have been of interest to this particular group, but I note that they did demonstrate that reduced two-point discrimination on the sole of the foot correlated with impairment in the balance test, suggesting that decreased somatic sensation plays a role.

 

[noobster]

The last part is the key. As we age genetics to plays a large part. We see those in 80s and 90s that are very fit including dancing or running marathons or playing tennis. We also see those who get frail. The state of health plays a major role. I ponder whether the loss of balance is age related or is abated by loss of strength in muscles and bones.

It's probably multifactorial. I think you can offset the age-related degradation of almost any skill to a great degree by regular training and practice.

For what it's worth, I do not notice any degradation in dancing balance from when I was in my 20s. Closing my eyes does not seem to affect my balance when dancing either. But I've been social dancing for a long time now, and I don't regularly practice standing on one foot with my eyes closed. Although perhaps I should start.

 

[Brummbaer]

I have been wondering how totally blind people manage their balance. They must have overcapacity compared to me in the other balance producing systems compensating the lack of vision.

Balance comes from the vestibular system of the ears.

 

[Brummbaer]

I do notice that while the younger people have more superficially impressive acrobatics, older dancers often have the ability to load small movements with a subtle but very beautiful expressiveness and flavor.

= combination of wisdom with experience.

 

[Gssh]

Oh I see. Sure I agree. But I would note that a better fit to the data, which seem to show a steep upward slope on the left-hand side and a more gradual reduction on the right-hand side, would probably place the apex even earlier, maybe in the late teens or early 20s versus the 30s or higher.

Without having the data it is hard to say much (and i am not really excited enough about their data to try to recreate it from their plot), but i would guess that a lot of the specifics of the shape of the cubic are due to the tight clusters at the left and right edges - data in the middle has less leverage, and having the majority of variability there doesn't help. It looks to me like what the cubic is fitting is that there is no further loss (because they are already at the lower bound) beyond 75 or so. This is a big source of the lack of fit of the quadratic, as the quadratic is very steep there, and prediction error for the 90s is huge (it is a bit hidden in the graph as they smewhat cheekily stop the fit line at a predicted value of 0, but if you extend the line to the last datapoint that is somewhere around 93 or you can see that the predicted value is maybe -87 for these people or something in that ballpark). So i am somewhat sceptical overall of using the stationary points of the fitted curves.

The sociodemographic bits may not have been of interest to this particular group, but I note that they did demonstrate that reduced two-point discrimination on the sole of the foot correlated with impairment in the balance test, suggesting that decreased somatic sensation plays a role.

I actually liked the two-point discrimination data much better - i probably would have lead with that, especially as the balancing tests seems to have awful measurement properties for this population: people on the upper bound, people on the lower bound, and based on the plots i would guess it is not linear to the underlying construct. It is a typical pub health measure: fast, minimal equipment, easy to administer, and good enough to talk about populations. I would not use it where variability and scaling could become important like when talking about mechanisms or individuals.

 

[LadyLeader]

Balance comes from the vestibular system of the ears.

Yes, vestibular system is collecting information together with eyes and muscles & joints. All this is put together and then muscles get the instructions for balance actions