Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

No gain from brain training

nature.com

Post reply on HN
Print view View original

11–20 of 28 posts

Re: No gain from brain training

#11

Post by a-priori »

mattmanser wrote: post #3

The volunteers were self-selected so probably already did this sort of thing online, the exposure extremely short and the testees from a broad age range who are not the intended audience for these type of games. All outlined in the article, but I restate it as it's a travesty that attention grabbing 'research' like this still gets published and worse still reported on.

How else do you expect to select volunteers? Usually, they're undergraduate students who are compensated with credit towards courses. I participated in a few experiments myself as an undergrad.

I presume that they administered a questionnaire where they asked, among other things, whether the volunteers have done this sort of thing before. From that you can divide people into an "experienced" and an "inexperienced" group, and compare their results.

Re: No gain from brain training

#12

Post by timwiseman »

mattmanser wrote: post #3

The volunteers were self-selected so probably already did this sort of thing online, the exposure extremely short and the testees from a broad age range who are not the intended audience for these type of games. All outlined in the article, but I restate it as it's a travesty that attention grabbing 'research' like this still gets published and worse still reported on.

Good points. This far outside my field, but the largest thing that stuck out at me was how limited the amount of training was.

If I did less than 4 hours worth of free weight training over a period of 6 weeks, I would not expect to see noticeable results at the end of it.

Re: No gain from brain training

#13

Post by pygy_ »

robg wrote: post #6
DennisP wrote: post #2

They don't mention specifically whether they tested dual n-back, which as far as I know is the only task that some studies have shown to increase fluid intelligence.

That's a very difficult task. Can you point to those studies?

Here's one: http://www.pnas.org/content/105/19/6829.full

The full text is online.

The study has a control group. The subject had to perform the task for about 25 minutes per day, every day.

The effect is dose-dependent. The longer the subjects trained, the better they fared on an unrelated fluid intelligence test.

Re: No gain from brain training

#14

Post by drats »

robg wrote: post #6
DennisP wrote: post #2

They don't mention specifically whether they tested dual n-back, which as far as I know is the only task that some studies have shown to increase fluid intelligence.

That's a very difficult task. Can you point to those studies?

Small sample, improvement mostly in the people who were scoring worst to begin with if I recall correctly. http://www.iapsych.com/articles/jaeggi2008.pdf Jaeggi et al also have a 2010 paper extending/replicating the results.

More interesting to me is that after reading a popular article about an increase in lucid dreaming since the 1980s I thought there would be a connection to gaming. Turns out there are a few academic papers exploring this. This is important to the type of culture technology engenders and the type of internal actors or agents we are at the core, passive or active. I think computer games rather than TV or flim, the active hyperlink rather than the "take our word for it" print hold promise in the sense of making us better at being individuals. Attention-seeking bleating about the lowering of attention spans in recent decades doesn't take into account that there is something to be said for brevity; and that there are far too many system-building waffling patriarchal walruses "performing" their alpha-male genius through turgid prose (especially in philosophy and related disciplines).

Re: No gain from brain training

#15

Post by sayemm »

As a counter to this, Dominic O'Brien, the world's memory champion, is a terrific example of how far mental training could take you: http://en.wikipedia.org/wiki/Dominic_O%27Brien

He started training his memory at the age of 30 and is also dyslexic. Good youtube interview too: http://www.youtube.com/watch?v=N60_d8vvl9I

Re: No gain from brain training

#16

Post by ceejayoz »

sayemm wrote: post #15

As a counter to this, Dominic O'Brien, the world's memory champion, is a terrific example of how far mental training could take you: http://en.wikipedia.org/wiki/Dominic_O%27Brien He started training his memory at the age of 30 and is also dyslexic. Good youtube interview too: http://www.youtube.com/watch?v=N60_d8vvl9I

Mental training, or innate ability?

Re: No gain from brain training

#17

Post by phpnode »

ceejayoz wrote: post #16
sayemm wrote: post #15

As a counter to this, Dominic O'Brien, the world's memory champion, is a terrific example of how far mental training could take you: http://en.wikipedia.org/wiki/Dominic_O%27Brien He started training his memory at the age of 30 and is also dyslexic. Good youtube interview too: http://www.youtube.com/watch?v=N60_d8vvl9I

Mental training, or innate ability?

Almost certainly a loci based system: http://en.wikipedia.org/wiki/Method_of_loci

Re: No gain from brain training

#18

Post by Groxx »

It does seem that brain training doesn't directly improve anything but what was trained for.

I buy that. Taken from a more computational standpoint, that'd be the holy grail: we'd have to be nigh perfectly abstracted to allow improvement in any one area to improve another area. We're generalists, but we're not that good of generalists.

Training does encourage you to learn ways to learn what you're being trained on, and those ways of learning things may very well translate to learning other things. Direct cause-effect? No, thus no guarantee, which is what science wants very much. But I'd argue some gain is indeed, uh, gained.

Re: No gain from brain training

#19

Post by YooLi »

robg wrote: post #9
danielford wrote: post #8

I'm desperately trying to think of any human activity that would substantially improve after ten minutes of practice, three times a week, for six weeks. My general inclination is to agree with their hypothesis, but testing it like this is just embarrassing.

Try doing sit-ups or push-ups that much and that often. I'm pretty sure you'll improve over six weeks. That said, more works needs to be done.

Agreed, but you would only improve at sit-ups or push-ups, and there is a point where you will plateau. Doing push-ups 3 times a week for 10 min. won't make you good at pull-ups, which it seems is what people are expecting with these brain 'training' games. If you want to improve your mind, you do have to work it, but the work needs to be varied and consistent: work on your memory, read, learn to play a musical instrument, learn new physical skills, do math problems, etc. You have to do it all and keep doing it (unless you want to specialize, in which case you do one thing over and over as has already been said in the comments).

Re: No gain from brain training

#20

Post by mattmanser »

robg wrote: post #7
mattmanser wrote: post #3

The volunteers were self-selected so probably already did this sort of thing online, the exposure extremely short and the testees from a broad age range who are not the intended audience for these type of games. All outlined in the article, but I restate it as it's a travesty that attention grabbing 'research' like this still gets published and worse still reported on.

All volunteers are self-selected. Nature is the right venue. It will spur much more research which is the only way to evaluate the science.

Admittedly I wasn't clear there about the volunteers, these are people who volunteered via a BBC programme called 'Bang goes the theory' and I suspect probably also recruited via the BBC's website.

They do not represent a broad demographic, they are extremely biased. They are not, and never will be, a good group to base any sort of scientific test on. It violates the scientific method and these 'results' should never have been published in any serious form. To be frank it is medieval quackery.

Owen should go read some Karl Popper and bow his head in shame for even trying to draw any conclusions from results obtained by recruiting test subjects from a tv programme.

I'm not saying that he's wrong, he may be totally right. But his test was inherently flawed, yet he presents his results to Nature? He should never have written to Nature and his 'test' is as scientifically significant as 'Ask HN: Which browser do you use?'.

It happens all the time, but it's really beginning to piss me off.

Post reply on HN