|
|
show details 1:42 AM (15 hours ago)
|
--
You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To post to this group, send email to brain-t...@googlegroups.com.
To unsubscribe from this group, send email to brain-trainin...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/brain-training?hl=en.
Sumo wrestling is a weight-intensive task, and one's performance in
sumo wrestling is often weight-limited. Sumo wrestling doesn't make
you weigh more. Eating like your mom does.
Reaching with your mouth to try to bite fruit hanging above your head
won't make your neck longer. Putting a metal coil between your chin
and your shoulders will. (http://en.wikipedia.org/wiki/Neck_ring)
We don't know why the n-back tasks improve performance on matrix
reasoning tasks. We hypothesize that it's because it's training
working memory, but the data support other hypotheses (e.g. it helps
via spatial reasoning ability) just as well. However, the data
collected so far imply that it does improve matrix reasoning task
performance.
If we eventually discover that, as we suspect, the n-back tasks
improve matrix reasoning tasks because they improve working memory,
then we still won't know why they improve working memory. It very well
could be that they improve working memory because they require one to
access ("read from") and modify ("write to") working memory at the
same time. It could be because of the rhythmic temporal structure of
the task. It could be because of the progressive and adaptive
difficulty of the task. It could be because of something completely
different. We simply don't yet know.
What we do know is that, according to three separate controlled
clinical trials, it improves performance on matrix reasoning tasks.
In the past, almost everyone believed, as you seem to, that one can
become more intelligent by studying more, and by getting educated.
Eventually, understanding of scientific and statistical methodology
improved to the point where we could test that hypothesis, so in the
20th century, they tested it, expecting to learn exactly how much it
helps and what types of education help intelligence the most. The data
that they collected showed that, essentially, nothing (except maybe
music?) helps. As a result, some people abandoned that hypothesis.
TL;DR: Just because math uses working memory doesn't mean that it
improves working memory. Education makes you more knowledgeable and
more capable. It does not make you measurably more intelligent. N-back
training seems to.
Jonathan
I probably shouldn't have used the word "same". Because I'm not a
consumer I don't know a whole list, however, I was just referring to
ADD or ADHD (sorry if I'm mistaken) alleviating medications such as
ritalin, dexedrine & other.
On Jan 16, 5:36 pm, Działo, Christopher <chrisdzi...@gmail.com> wrote:
> Thomasthetankengine,
> *
> *
> Regarding your comment "I do not know of anything else that will
> provide me with
> the same benefits (apart from drugs)." which drugs are you referring to?
>
> On Sun, Jan 16, 2011 at 1:33 AM, Thomasthetankengine <
>
They are backed up by the data; as I clearly adduced the current
evidence as an example of a speed factor here. The RPM result is a
confound of the Gf-Gs aspect. Quite elementary, in my view.
You really should try not sounding like a child when you talk nonsense
about my "superiority flag". What utter bollocks. :)
If I were here to look good for everyone, I would have kept my mouth
shut long, long ago, so that clearly is not what is at issue here.
This is what I call argumentum ad hominem on your part. (I would
invite others to look into my chat history if they wish to see others
engaging in similar behavior, with the purposeful intent to sully my
internet persona. O woe is me. So much for those "losses" you speak
of.)
Do you suppose DNB has aided you in your muscling the needed thought
to fashion the dialectical tools of a simple linguistic exchange, or
does the whole affair merely seem to stand clearly in your mind? Do
new thoughts... occur to you? (Questions to keep in mind with your
ongoing training; don't forget that if you're still young, that will
likely cloud any improvements you might notice.)
Well, I'm beginning to digress, but not very far; just around the
periphery. Contrary to what you seem to imply what I'm saying, the
issue isn't whether people should or should not use DNB; what does the
use of this tool actually bring about? It is doubtful Gf has much to
do with it so far, although still possible - but to a small extent.
Charitably,
argumzio
On Jan 16, 12:33 am, Thomasthetankengine <thomasttt...@hotmail.com>
To be clear, the Gs aspect I speak of applies to a global cognitive
functioning ability, not the perceptual aspect, which is only one
small subset of the global that I'm talking about. I.e., does DNB
improve the g-speed factor more so than the g-fluid factor; in other
words, can one engage in a sequence of mental operations more quickly
(after training), or think of new and original ideas to bring order to
seeming chaos? That is the sticking point, I think.
By the way, thanks. It's nice to be appreciated. :)
argumzio
To unsubscribe from this group, send email to brain-trainin...@googlegroups.com.
Explanation for differences for g do not therewith imply that the
explanations, when improved, will lead to bona fide improvements in g.
WMC increase will _not_ give you the ability to think of original
solutions to various problems, only to follow more things and handle
them more efficiently. Pretty darn obvious; kind of like how someone
participates on a forum as two members, thinking no one will have the
brains to lift the veil and see it as clear as day. (Don't fret, I
even know some people here, beyond their handles... and they are too,
too obvious.)
Tests which require this in a straight-forward manner will invariably
link g to these constructs, as if these were the province of g. But
any psychometrician will tell you that g cannot and should not be
reduced to WMC. Speed, as you will eventually discover, has
practically no relationship with g; this is why it is its own,
separate factor – why there is a g-speed apex outside of the g-
intelligence apex.
In my view, it's best to stick to what is currently known rather than
get too carried away with preliminary results (Jaeggi et al.) and soar
to the heights of theoretical to-and-fro's. As much as others hope Gf
really is improved by this, one has to have at least a pinch of self-
respect to acknowledge that the current evidence couldn't possibly be
the basis to arrive at such a conclusion.
argumzio
On Jan 16, 5:53 am, Thomasthetankengine <thomasttt...@hotmail.com>
wrote:
> Although I am yet to obviously delve into all of the little goodies
> that are mentioned in ur suggested readings Argumzuo, from first
> glance it seems that Gs and Working memory are the primary constructs
> that help explain individual differences in G.
>
> That being said...
>
> The goal of n-back is to increase one's working memory capacity, and
> others such as your self have suggested that it is also or otherwise
> (just) related to increasing Gs. If this is the case and because both
> constructs 'act' or 'contribute' globally to cognition, then it _may_
> be fair to say that there are actual gains in G? Based on the premise
> (n-back = Gs and WMC).
>
> I understand though that I have a lot of reading ahead of me.
> Hopefully hear a response from you, thanks.
>
> Thomas
> ...
>
> read more »
--
You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To post to this group, send email to brain-t...@googlegroups.com.
To unsubscribe from this group, send email to brain-trainin...@googlegroups.com.
--