Hacker News new | ask | show | jobs
by stiff 4685 days ago
I just don't understand how you can state bold claims like this as facts without substantiating them in any way by references or any sort of logical deduction:

If a person learned to become exceptionally 'talented' in one area, then he/she's more likely to replicate that same level success in a completely new area

The only limiting factor on skill acquisition is choice/taste.

Doesn't genetics also have some role here? The references you just included do nothing to provide any sort of evidence for those statements. To what extents cognitive skills are transferable is still much debated and science is far from having an unanimous answer. Good overview of the issue is here:

http://www4.ncsu.edu/~jlnietfe/EDP504_Notes_files/Are%20Cogn...

1 comments

Genetics do play a surprisingly significant role, and my second reference (book) even mentions it in the title. IQ is strongly believed to be linked to genetics, but again, as I listed in one of the sources, even IQ doesn't necessarily factor into skill acquisition, only ease of skill acquisition; two distinct but important points. You can learn to learn better -- that is the main point. It doesn't matter where you start off (i.e. IQ via genetics), you can still get a skill snowball-effect going relative to where you started. It seems you missed the second part of my last comment where I mentioned it was more segmented and context based (like your citation states), but that the general trajectory is still the same. Learn a skill in one context, any subsequent skills in the same context become easier. The thing is that there are things that generalize to multiple contexts, and that some contexts are larger than they seem[1].

Now, it is certainly much harder for some people to acquire certain skills, and it may be near impossible to excel at them up to a certain level (e.g. physical athleticism). But skill level is again different from the skill itself, and it is the skills themselves that matter in skill acquisition. For example, if running ability is found to be genetically limited, then getting any better at it may be harder for you, but picking up a sport like football would probably still be much easier for you if you trained your running first regardless. IQ being genetically determined can impact other things like motivation; i.e. if you have a low IQ things will be harder for you, making you less likely to pursue skill acquisition in the future, and conversely, a high IQ may predispose you to acquiring more skills since they come so easily to you, but the point is that neither absolutely determine how many skills you can/will acquire.

I made a 'bold' statement because it's just that practical of a perspective to take[2]. And science backs it up not only in terms of all the positive effects exercising brain plasticity brings[3], but also in the fact that IQ isn't an end-all be-all metric. It is significant, and highly correlated to other important things like life expectancy[4], but IQ itself is not the limit. A built-in limit to skill acquisition would have to come somewhere else down the line if there is one. I study this stuff on my free time because it interests me, I'm not an expert/scientist, so trying to discover a well-defined concrete limit to skill acquisition is beyond my domain, but stuff like epigenetics and plasticity is making it harder to believe there is one.

[1] I believe the last cite in my previous comment shows an example of this. Also, I myself have a pretty average IQ of ~114 or something, but this view has allowed me to learn a surprising number of skills before I even knew any of the science around it, so of course I'm biased and wanted to share.

[2] Well, that and the fact that most people simply don't bother to look at painfully constructed source lists anyway, making the endeavor of compiling them less worth it.

[3] One example: http://news.stanford.edu/news/2003/september24/dementia.html

[4] http://www.mrc.ac.uk/About/AnnualReview09-10/SevenAges/Elder...

The problem here is that we humans really want to believe certain things, and our emotional involvement blinds us to what we really managed to establish scientifically, e.g. people vehemently oppose any suggestions of determinism, attempts to deny free will, whatever it might mean, etc., regardless of any logical argument.

I think you are falling for this in certain places as well, for example when talking about genetic skill level limitations you jump right to physical athleticism, which is something people are somehow able to accept more easily, while there is strong evidence that intellectual abilities are open to limitations of the same kind.

IQ is an imperfect measure that is at best correlated with the "quality" of ones genetic endowment, but it doesn't mean that the genetic limitations are any less real. This is the very old and heated debate of "nature vs. nurture", and I would appreciate your comment much more if you also included views of the "opposing" side and moderated your claims to what the research really says, while I think you are making some big extrapolations. That is not to deny the possibility of "learning to learn" or to discourage learning, but you used very strong phrases.

I actually don't think our views are opposing, but I think the point I'm trying to make is just tricky because falls in a small area that doesn't oppose determinism. I actually view the mind as a total algorithm, because there are a shocking number of personality traits that appear to be deterministic[1]. Thus, I don't actually believe in free will, but I find it to be a useful model by which to live by (kinda like how classical mechanics helped us get to the moon despite relativity ultimately being more accurate). There are many things that can still work within a deterministic system though, they may just require the right set of inputs[2] to get the desired outputs. So I just proposed a perspective (one that leverages our apparently limitless ability for plasticity and memory) as an input, and maybe it'll trigger some people to deterministically consider it for helping themselves, leading to useful outputs.

The system is much too complex to assume that just because we don't have what amounts to total 'free' will that we're hopeless to improve anything at our level of operation. If someone's determined (pun intended) to be a defeatist, then alright, but some others are just waiting for the right inputs to take them down a more useful branch of execution. Because just like a program, even though everything is neatly outlined and determined, that doesn't mean you know what every output ever will be. That's why I don't think 'hardcore' determinism to the point of discouraging choices is a useful view to take, much like how hardcore philosophical skepticism is a dead-end line of logic; neither really provide anything you can build off of, so while they may ultimately be true, they're poor models for productivity. I mean, it's possible that you're right and I'm just having a hard case of cognitive dissonance, but it seems to me like extrapolating deterministic genetic algorithms to argue against useful high-level perspectives is still making a lot of assumptions about the implications of such a system. Meanwhile, I'm just reporting observations that been found with regards to skill acquisition.

[1] The phenomenon depicted on this episode of Radiolab with regards to Transient Global Amnesia is particularly damning (http://www.radiolab.org/2011/oct/04/ ), because it shows that when given all the same inputs, you're likely to perform the exact same actions over-and-over again. The separated twin studies on IQ also show a remarkable number of personality similarities amongst twins (besides IQ), which indicates a possible genetic component to random things like sense of humor. There's no hard evidence that any of these things are genetically determined of course, but meh. Let's also not forget that epigenetics and GMOs exist, though it may be a while before that becomes useful for GATTACA-like situations, lol.

[2] Yes, genetics and other deterministic factors count as inputs. If you want to reach a branch of logic that requires AND-ing with a genetic component you don't have, then tough luck, but a simple OR with some other less deterministic input is equally possible. Of course, this is just another hypothetical model to cope with our lack of understanding.