Playing Tetris increases cortical thickness and brain efficiency
David
http://www.biomedcentral.com/1756-0500/2/174/abstract
Yet more evidence for brain plasticity, cortical thickness also
correlates with IQ. Tetris also increased brain efficiency in the
frontal/parietal lobes. I wonder if such neurobiological changes can
also be induced by other cognitively demanding games other than
Tetris.
The study doesnt mention whether they used 2D or 3D Tetris (although
likely it seems it was the 2D as thats the standard game) but it would
be interesting if 3D Tetris could enhance 3D spatial temporal
reasoning (which correlates with Mathematical aptitude)
Josh
cortical thickness is murky, to say the least. According to Shaw, IQ
can be related to less cortical thickness - so does this mean that
playing tetris decreases IQ???
On Sep 19, 2:32 pm, David <david.j...@yahoo.com> wrote:
> http://www.medicalnewstoday.com/articles/162617.phphttp://www.biomedcentral.com/1756-0500/2/174/abstract
David
suggesting positive correlation between cortical gray matter and IQ
http://www.ncbi.nlm.nih.gov/pubmed/17118969
On Sep 19, 10:44 pm, Josh <joshlevi...@gmail.com> wrote:
> The study doesn't mention IQ at all. The relationship between IQ and
> cortical thickness is murky, to say the least. According to Shaw, IQ
> can be related to less cortical thickness - so does this mean that
> playing tetris decreases IQ???
>
> On Sep 19, 2:32 pm, David <david.j...@yahoo.com> wrote:
>
Jonathan Toomim
> http://www.medicalnewstoday.com/articles/162617.php
> http://www.biomedcentral.com/1756-0500/2/174/abstract
>
> Yet more evidence for brain plasticity, cortical thickness also
> correlates with IQ.
Cortical thickness in what parts of the brain correlates with IQ?
Hint: None of them are places in which this study found thickening.
So no, the supplementary motor cortex (the girls got better at
pressing buttons!) and left anterior superior temporal gyrus are not
known (at least by me) to be associated with IQ.
> Tetris also increased brain efficiency in the
> frontal/parietal lobes. I wonder if such neurobiological changes can
> also be induced by other cognitively demanding games other than
> Tetris.
More specifically, practicing Tetris increased brain efficiency at
playing Tetris in the frontal/parietal lobes. Shocking, isn't it?
When you practice playing a game 1.5 hours/week for 3 months, your
brain gets more efficient at playing that game. Wow!
Though really, even calling it "Tetris-playing efficiency" seems like
a poor description. Activation increased in the occipital and
parietal areas (visual and spatial areas), and decreased in frontal
areas. To me, this sounds like when they were just starting, the
girls' frontal lobes were heavily activated because they were just
learning the task, and as a result, they weren't able to fully
leverage their visuo-spatial processing capabilities; after they
learned the task, they no longer needed the frontal activation, and
they were better able to make sense of the visual and spatial
information of the game.
> The study doesnt mention whether they used 2D or 3D Tetris (although
> likely it seems it was the 2D as thats the standard game) but it would
> be interesting if 3D Tetris could enhance 3D spatial temporal
> reasoning (which correlates with Mathematical aptitude)
Vocabulary in one's native language correlates with mathematical
aptitude too. Is studying vocab going to improve mathematical
aptitude? Is playing Tetris going to help you learn to take integrals
and manipulate fractions, or is it just going to make you really good
at packing irregularly shaped objects into containers while wasting a
minimal amount of space? Or maybe even neither? I see no reason to
reject the null hypothesis (that playing Tetris makes you better at
anything other than Tetris) in this case.
By the way, the study does mention that it was 2D tetris (or at least,
it mentions enough information for you to be able to determine that--
for example, the players use "four buttons to move and/or rotate the
piece," which rules out 3D tetris (which would require at least 6, I
think)).
Jonathan
Gwern Branwen
> By the way, the study does mention that it was 2D tetris (or at least, it
> mentions enough information for you to be able to determine that--for
> example, the players use "four buttons to move and/or rotate the piece,"
> which rules out 3D tetris (which would require at least 6, I think)).
>
> Jonathan
A bit off-topic, but you could probably get away with fewer than 6
keys by making keys harder to use. For example, http://3dtris.de/
gives you the entire D-pad, 4 keys, for moving the piece around. But
you could cut this in half by providing, say, only Up and Left, and
making them overlap. (So if I wanted the piece one block to the right,
I would hold Left until I had wrapped all the way around, and
similarly for up.) One could cut this in half again by making the Up
key, say, traverse each possible position and wrap to the next row or
column.
Similarly with rotating the piece rather than moving it. 3dtris
provides 6 keys - left & right for the x, y, and z axis. (3*2) But as
before, one of the rotations is redundant because of symmetry, and so
on.
Hence, one could play a 3d Tetris with 2 logical keys. Or fewer - it's
all binary information encoded somehow in the sequence of keystrokes
over time.
So the number of keys has no necessary connection. It is still very
probable that 4 keys indicates classical 2D Tetris, though.
--
gwern
Oye
regions can become bigger due to plasticity.
http://www.youtube.com/watch?v=f0G8Lz5vFds
Like he says, Einstein did apparently NOT have some 10-15% larger
brain regions compared to the median when he was born. He achieved
larger brain regions because of his work. That would imply all brain
regions which involve: logic, mathematics, problem-solving, abstract
thinking, physics, scientific thinking, critical thinking etc etc in
other words much of the brain tissue in the frontal lobe, specifically
the prefrontal cortex which is now accepted to be very important in
intellectual activity and the Parietal Cortex.
I have trained and played a LOT with Tangrams. They exercises both
logical, visuo-spatial, creative, problem-solving abilities. AND also
I have found that one area which seems to be involved in understanding
irony, sarcasm and humour also controls much of the visual ability.
AND THEN I have actually experienced that my irony/sarcasm/humour
ability seems to be significantly better even if I havent trained
those abilities very much. But I have trained the visual ability very
much and its the same brain region! I have completed around 3000-3500
tangram silhouettes/puzzles/pictures.
And more: When I was a new-beginner at tans I wasnt so good, I had to
use the key with the answers at perhaps the 500-1000 first puzzles.
After that I could solve most of them without. Practice do better.
Now I can mentally visualize most of even the most difficult puzzles
in my head without use the physical tan pieces and I never consult the
answer key.
Not to boast, I wouldnt be so good at Tangrams without all that
playing and training, thats for sure.
But this is a good example of Brain Plasticity. And Im convinced its
not just that I have learned the Tangram language and remember the
concept well, some parts of my brain have changed both in structure
and integrity.
On Sep 19, 11:32 pm, David <david.j...@yahoo.com> wrote:
> http://www.medicalnewstoday.com/articles/162617.phphttp://www.biomedcentral.com/1756-0500/2/174/abstract
wzeller
Aren't you just saying that learning changes the brain, and more practice causes more learning and more brain changes? Does anyone really disagree with that position?
>
> On Sep 19, 11:32 pm, David <david.j...@yahoo.com> wrote:
>> http://www.medicalnewstoday.com/articles/162617.phphttp://www.biomedcentral.com/1756-0500/2/174/abstract
>>
>> Yet more evidence for brain plasticity, cortical thickness also
>> correlates with IQ. Tetris also increased brain efficiency in the
>> frontal/parietal lobes. I wonder if such neurobiological changes can
>> also be induced by other cognitively demanding games other than
>> Tetris.
>>
>> The study doesnt mention whether they used 2D or 3D Tetris (although
>> likely it seems it was the 2D as thats the standard game) but it would
>> be interesting if 3D Tetris could enhance 3D spatial temporal
>> reasoning (which correlates with Mathematical aptitude)
>
> --
> 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.
>
Reece
certainly didn't have the technology back then that we have today).
Most people aren't interested in what they're not good at, so I think
we may have a hard time convincing lower IQ individuals that they
should be practicing math + physics more, regardless of the merits of
doing so.
@wzeller, I think Jonathan made a good point when he mentioned that it
is probably unsurprising that people who play lots of Tetris got
better at Tetris (and that has certainly been my experience). Why dual
n-back is interesting (more interesting than Tetris for example) is
because practicing dual n-back makes us not only better at dual n-back
**but also** at things which seemingly have little to do with dual n-
back. I honestly believe playing speed dual n-back with position+image
stimuli is making me a better Tetris player, however I don't expect
playing Tetris to make me better at dual n-back (it doesn't even seem
to translate into higher DNB scores when using tetrominos for image
stimuli -- probably because for any N level above maybe 1 or 2,
working memory demands are most certainly higher in DNB).
Oye
tetris could also induce neuroplastic changes and I
answered yes, I do indeed think so. Tangrams is a good example. Then I
perhaps answered a bit in general but when I believe all or most brain
regions can be trained I also would believe that all exercises which
train them could induce plasticity. And therefore I think that many
many other cognitive/noncognitive activities coud induce brain
changes. When it comes to cognition I think games like Chess, Go,
Tangrams, Monopol, 3D Visuospatial games like Rubix Cube, Computer
Strategy games, FPS (first person shooter), word games, Scrabble, a
lot of mindgames such as Tower of Hanoi et etc and activities like
reading, writing, arguing, learning anyway every activity you use your
head.
But no, youre probably right, most people dont disagree with that
position. But the author wondered what could induce those changes.
> >> Yet more evidence for brain plasticity, cortical thickness also
> >> correlates with IQ. Tetris also increased brain efficiency in the
> >> frontal/parietal lobes. I wonder if such neurobiological changes can
> >> also be induced by other cognitively demanding games other than
> >> Tetris.
>
> >> The study doesnt mention whether they used 2D or 3D Tetris (although
> >> likely it seems it was the 2D as thats the standard game) but it would
> >> be interesting if 3D Tetris could enhance 3D spatial temporal
> >> reasoning (which correlates with Mathematical aptitude)
>
> > --
> > 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.
>
> - Show quoted text -- Hide quoted text -
>
> - Show quoted text -
Oye
that some of Einsteins brain regions was larger: does it mean he was
born with them or did he achieve them?
But even if we dont have good evidence for it, it is really consistent
with the study of jugglers which over training period acheived larger
brain area for coordinating complex juggling movements.
You see the guy in teh youtube video talks about it or the taxi
drivers who gets a larger Hippocampus. But again they could just have
a larger one from before, and because theyre good at it they choose to
work as a taxidriver perhaps.
And another which shows more gray matter in some brain regions after
aerobic exercise.
Arkanj3l
driving marathon running musician.
Jonathan Toomim
Arkanj3l
On a slightly-more-on-topic note, jugglers are basically experiencing
the same thing, correct? Brain optimization, but only for that
specific task.
Oye
responsible for: decision making, I think problemsolving, working
memory, empathy etc
If you train your working memory very very much, then there are
studies which show synaptic changes, neuroplasticity in the regions
which is responsible for the working memory especially the Prefrontal
Cortex. The Orb Cortex is the anterior part of the PFC. So then the
Orb Cortex should be neurologically changed.
Then the question: If the Orb Cortex is changed due to the plastic
theory via for ex Brain Workshop why wouldnt that also improve the
other abilities which the Orb Cortex also is responsible for and not
only the working memory capacity?
On Sep 21, 12:20 pm, Arkanj3l <kenneth.bruskiew...@gmail.com> wrote:
> ... by taxi-driving, running marathons and playing music. Duh!
>
moe
highly correlated with iq (right posterior intraparietal sulcus). In
my experience juggling is the only activity that has yielded a
subjective improvement in my spatial ability. Whats odd is that my
discernment speed/efficiency doesn't seem to have improved but my
spatial efficacy seems to have improved (eg I'm not any faster at
rotating objects but it's as if my brain doesn't have to work as hard
as it did previously). It looks as if juggling increases simultaneous
memory capacity which I guess can be measured by subitizing ability,
spatial rotation, etc. After juggling, ordinary things/objects appear
more salient, it's actually kind of annoying at times when you can't
stop counting street lights or when the brain constantly takes in the
number of bolts/etc on objects. Anyone else gain from juggling?
Josh
that IQ is very difficult to train or practice, partly because IQ
involves ability to perform in novel situations so by definition you
cannot have practiced them before. If you have practiced them before
(eg, already seen the test item and memorized the answer), the test is
no longer assessing IQ.
Studies then show that practice leads to brain changes. Why does this
mean that IQ is automatically affected, when direct studies of
practice on IQ have found disappointing results?
Or are we assuming that the brain only varies in one dimension only.
Hence, the dimension that practice affects is the same dimension that
causes differences in IQ?
Pontus Granström
subtests, so chances are pretty high that an performance increase in
wmc is more or less linked to an increase of G/IQ. Working memory
tasks can never be solved with long term memory (knowledge) just as I
cannot remeber to do 100 pushups I need to have muscular quality to
complete them.
Josh
eg, tetris, juggling, are not "working memory" tasks. But even if we
restrict the discussion to working memory tasks like n-back, how is it
that just because they use the same brain areas then improving
performance in one will have a "pretty high" chance of improving
performance in the other. Again, are we assuming that there is only
one dimension along which a brain region can vary?
On Sep 21, 10:11 am, Pontus Granström <lepon...@gmail.com> wrote:
> Working memory tasks overlap more with areas linked to G than other
> subtests, so chances are pretty high that an performance increase in
> wmc is more or less linked to an increase of G/IQ. Working memory
> tasks can never be solved with long term memory (knowledge) just as I
> cannot remeber to do 100 pushups I need to have muscular quality to
> complete them.
>
Pontus Granström
region can vary only that if we know that working memory tasks are
g-loaded and we improve your WMC that we would have to improve G as
well. To flip the question is to say that we can improve your WMC
without it actually changing anything in the brain or that all
improvement in the brain that effects wmc does not effect g.
I got an really interesting article on this business in the files
section, "distributed neural systems for general intelligence".
Josh
is involved.
Consider the following hypothesis... g is determined by 5 parameters.
WMC is determined by 5 parameters. One of these parameters is in
common across both g and WMC. Hence, you would find a moderate
correlation between g and WMC. This model would be supported by
various empirical findings that show that g and WMC are related, but
can also be differentiated. If you then train WMC, this might
influence one of the other parameters that determine WMC. Hence, WMC
would go up, but g would be unaffected.
On the other hand, your statement only holds if there is truly only
one parameter involved. You say that "if we know that working memory
tasks are g-loaded". g-loading only indicates that there is a
correlation, not the strength of the correlation unless you assume a
one parameter model. Then you say "and we improve your WMC that we
would HAVE to improve G as well." This only applies in a one parameter
or one dimension case...
On Sep 21, 11:00 am, Pontus Granström <lepon...@gmail.com> wrote:
> No we are not assuming that there's one dimension in which a brain
> region can vary only that if we know that working memory tasks are
> g-loaded and we improve your WMC that we would have to improve G as
> well. To flip the question is to say that we can improve your WMC
> without it actually changing anything in the brain or that all
> improvement in the brain that effects wmc does not effect g.
>
> I got an really interesting article on this business in the files
> section, "distributed neural systems for general intelligence".
>
Pontus Granström
almost 0.5 that's 4-5 times higher than other sub tests. If we get
white matter gains in areas linked to g and improved neural circuits
that working memory recruits as well as g, the g-increase is more or
less a fact. It's a raw biological concrete change in the brain that
will enable us to solve more complex problems. This is what I think
happen among other things!
Lou
noticeable improvement in my hand eye coordination. Any activities
that required me to identify an object and grasp it I could do much
better: cleaning, doing the dishes, playing catch, playing speed.
Even
when I went grocery shopping, when I lifted the items off the shelf,
it seemed relaxing, enjoyable, and easier. It also seemed to be
easier
to predict the destination of rapidly moving things, like cars or
tennis balls, as if my brain would do it on its own without a need
for
effort.
I've noticed bingo also seems to have transferable effects. In bingo
you have to pay attention to the the numbers being called (auditory),
rapidly scan through your cards (visual,) click the number once found
(tactile,) and immediately begin on the next number being called.
There is a lot of task switching, remembering, and searching. After
about a month of bingo, searching for titles on a shelf, fallen
contacts, phrases in text---anything related to visual searching
seemed much easier. I'm thinking switching between unrelated
activities is probably easier too, though it is hard to gage it after
having played for so long.
Josh
is not that high. It is only people who have an agenda to push the
working memory explanation that claim that this is impressive. Second,
it is by no means 4-5 times higher than other subtests. What about
word analogies? These have much higher loadings on fluid intelligence,
and also don't seem to be dependent on working memory demands for
their difficulty as all items, from the easiest to the hardest,
consist of only 3 words, or really only 2 representations at a time.
Doesn't this indicate that fluid intelligence and working memory can
be distinguished?
You then bring up white matter gains. Again, how it this necessarily
related to g or fluid intelligence? How does transmitting information
faster lead to more intelligence? Wouldn't wrong answers as well as
right answers be transmitted faster?
What are "improved neural circuits"? Without specifying what the
improvements are, how can you make the argument that the improvements
are relevant to both working memory and g?
On Sep 21, 1:39 pm, Pontus Granström <lepon...@gmail.com> wrote:
> Well true, but working memory tasks overlaps quite much with g areas
> almost 0.5 that's 4-5 times higher than other sub tests. If we get
> white matter gains in areas linked to g and improved neural circuits
> that working memory recruits as well as g, the g-increase is more or
> less a fact. It's a raw biological concrete change in the brain that
> will enable us to solve more complex problems. This is what I think
> happen among other things!
>
Pontus Granström
dependent as well and as a such shows a high overlap with the "real g
part" of the brain.
This analysis revealed only a modest overlap for Picture
Arrangement (0.10), Block Design (0.09), and Picture Completion
(0.08), indicating that visuospatial skills are vulnerable to damage
in much larger areas of the right hemisphere than those involved in
g (Fig. S3). However, working memory and verbal skills overlapped
more substantially with the left hemispheric correlate of g, most
notably for the Arithmetic (0.42) and Similarities (0.39) sub tests.
Arithmetic has 4 times higher overlap than other tests, which means
that those skills are statistically linked to g rather than
biologically linked. Arithmetic shows the highest g-load in the whole
WAIS according to factor analysis.
The SLT revealed that all verbal sub tests (loadings: 0.57–0.66) as well as the
Arithmetic (0.67) and Block Design (0.57) sub tests exhibit high
loadings on g, consistent with previous accounts (10).
I know that arithmetic skills are very indicative of g, in the swedish
airforce battery test numerical dependent tests shows the highest
g-load although the test requires much more than this.
Brain scans shows that n-back is more or less identical to
"g-activity". Besides WM is a label and one should be careful with
lumping different WM-tests together!
White matter is interesting for intelligence since grey and white
matter in some sense constitutes intelligence at a biological level.
Myelin is what scientist try to target for IQ improvements. Since
n-back seems to cause myelin increases it's not a bold leap to believe
that this causes increases in what we label as Gf.
Improved neural circuits means that we get strengthen pathways for
this involved Gf problems. Just as the brain forgets certain things
that are not frequently used activating such circuits (gf circuits)
would improve gf.
Josh
> dependent as well and as a such shows a high overlap with the "real g
> part" of the brain.
>
dependent as well, as opposed to other subtests like picture
arrangement and block design? All of these subtests involve working
memory, but only one correlates highly with g. This suggests that g is
not just working memory, but something more.
> This analysis revealed only a modest overlap for Picture
> Arrangement (0.10), Block Design (0.09), and Picture Completion
> (0.08), indicating that visuospatial skills are vulnerable to damage
> in much larger areas of the right hemisphere than those involved in
> g (Fig. S3). However, working memory and verbal skills overlapped
> more substantially with the left hemispheric correlate of g, most
> notably for the Arithmetic (0.42) and Similarities (0.39) sub tests.
>
> Arithmetic has 4 times higher overlap than other tests, which means
> that those skills are statistically linked to g rather than
> biologically linked. Arithmetic shows the highest g-load in the whole
> WAIS according to factor analysis.
>
> The SLT revealed that all verbal sub tests (loadings: 0.57–0.66) as well as the
> Arithmetic (0.67) and Block Design (0.57) sub tests exhibit high
> loadings on g, consistent with previous accounts (10).
>
> I know that arithmetic skills are very indicative of g, in the swedish
> airforce battery test numerical dependent tests shows the highest
> g-load although the test requires much more than this.
>
> Brain scans shows that n-back is more or less identical to
> "g-activity". Besides WM is a label and one should be careful with
> lumping different WM-tests together!
>
brain areas process both tasks, it does not mean that the same
dimensions or parameters of those brain areas affect the tasks in the
same way.
> White matter is interesting for intelligence since grey and white
> matter in some sense constitutes intelligence at a biological level.
> Myelin is what scientist try to target for IQ improvements. Since
> n-back seems to cause myelin increases it's not a bold leap to believe
> that this causes increases in what we label as Gf.
>
hence, improving the brain will improve intelligence. However, it
again is based on the one dimension or one parameter model. Why is it
not possible that myelin may be improved, but myelin does not
determine differences in IQ? Not only is there very weak evidence for
any empirical relationship between myelination and IQ, but there isn't
even a plausible or coherent story that explains how better myelin
would lead to greater IQ.
> Improved neural circuits means that we get strengthen pathways for
> this involved Gf problems. Just as the brain forgets certain things
> that are not frequently used activating such circuits (gf circuits)
> would improve gf.
>
(including tetris), and presumably this would be related to brain
changes. However, the point of IQ is that it is not readily improved
by practice. So maybe "strengthening pathways", whatever that is
supposed to be, does not actually improve the brain in the dimension
that leads to differences in IQ. I'm still not getting how improvement
in the brain necessarily means that IQ is also improved, when tests of
IQ itself do not show such improvement. Even though you acknowledge
that the one parameter model is not necessarily true, you seem to keep
falling back on it to justify your arguments.
Another issue I find very obtuse is the implied knowledge of neural
circuitry that is never explicitly stated. You don't give any detailed
explanation of how neural circuits work, and yet you claim that
strengthening them is all that is needed for intelligence to occur.
Researchers such as Garlick and Hawkins look in much more detail at
the brain mechanisms that underlie cognition, and their ideas don't
correlate with such a simple account.
Pontus Granström
for general intelligence, here you will find answers to some of your
questions regarding sub tests, their g-load and their g-overlap. It
also discusses white matter and intelligence.
Quotes
The latter account argues that g should involve inter regional communication
among many brain regions and therefore critically rely on the white
matter connections between them, whereas the former account
argues for a distinct region or network of regions implementing g.
It thus remains debated whether g should be thought of as a single
ability upon which other cognitive processes might draw, or
whether it itself is constituted by the multiple cognitive processes
from which it is psychometrically derived.
If G relies on white matter and we engage in activity that activates
g-areas and increase our white matter in that area, g has increased
per definition.
Josh
isn't sufficient to support your claims. In fact, it really isn't
relevant at all. The reason I say this is because it is based on a
clinical population, ie, people with brain damage. Hence, this means
the "parameters" we have been talking about are not the same as for
the normal population. Hence, while variations in the parameters may
cause differences in the clinical population, it does not follow that
the parameters also cause differences in the normal population.
Imagine this simple analogy. You have 2 groups of people. You give 1
group pens that work, and the other group pens that do not work. You
give both groups an IQ test. The group with pens that work does well,
while the group with pens that don't work do poorly. Simply because
their pens don't work and they cannot fill out the answers on the test
form. Based on this, you conclude that when IQ tests are given to
people in the general population, it is the quality of their pens that
determine how well they do on the IQ test!
This is flawed as the pens do not normally cause variation in people
doing IQ tests.
Exactly the same is true when you look at people with brain damage,
and then try and generalize the results to people without brain
damage. Just because missing a part of the brain causes a deficit on
an IQ test, it does not mean that this part of the brain is what
causes differences on the IQ test if this brain region is intact and
everyone has the same parameter. Other parameters may then be
responsible.
On Sep 22, 12:39 pm, Pontus Granström <lepon...@gmail.com> wrote:
> I suggest that you take a look in the file distributed neural systems
> for general intelligence, here you will find answers to some of your
> questions regarding sub tests, their g-load and their g-overlap. It
> also discusses white matter and intelligence.
>
> Quotes
>
> The latter account argues that g should involve inter regional communication
> among many brain regions and therefore critically rely on the white
> matter connections between them, whereas the former account
> argues for a distinct region or network of regions implementing g.
> It thus remains debated whether g should be thought of as a single
> ability upon which other cognitive processes might draw, or
> whether it itself is constituted by the multiple cognitive processes
> from which it is psychometrically derived.
>
> If G relies on white matter and we engage in activity that activates
> g-areas and increase our white matter in that area, g has increased
> per definition.
>
Pontus Granström
has been a keystone for neurological research for a long time going
back to the guy who got a steel bar through his head. Naturally you
have to be careful with all kind of research so that you can control
for the desired effect, I agree. In this study they were suffering
from focal brain damage which means that they have a certain area in
the brain that is damaged. Much of modern neuroscience is based on
research done on people with focal brain damage! This is quite
different from having a completely different brain due to genetics or
other factors.
Using people with brain damages is not the same thing as using pens
that do not work when taking IQ-tests, unless you want to decide which
pens do not work and so on. It all becomes a matter of which aspect
you study or what you try to deduce logically.
The study has been peer reviewed as well and I believe it was published in PNAS!
Pontus Granström
various reasons! They managed to identify that a "solely g area" is
tapped by working memory tasks. Which might support the idea of jaeggi
who claimed that WMC and G might share same capacity constraints, in
fact they do share 50%. This part seems to depend heavily on
interconnections/white matter which n-back seem to increase. Yes I
believe the case is strong in favor of n-backing.
Josh
reading a single study and treating it as gospel. Brain damage is one
method of looking at brain function, but does not necessarily show
what causes differences in performance for people who do not suffer
brain damage. It is well known that findings from clinical populations
do not necessarily generalize to the normal population. The analogy I
give with pens is quite appropriate in that the variance in the pen
population is greater than the variance in the general population when
taking IQ tests.
As for the paper being peer reviewed, it seems more that you are
making inferences that are not stated in the paper. Where does the
paper state that n-back will increase IQ? Where does it even state
that increasing white matter will increase intelligence?
Shaw et al (2006) is based on a normal population and shows that
increased intelligence is based on less gray matter. This was
published in Nature, which is a much more prestigious journal. Doesn't
this indicate that less gray matter is likely to be more important?
On Sep 22, 11:28 pm, Pontus Granström <lepon...@gmail.com> wrote:
> I do not understand this at all, studying people with brain damages
> has been a keystone for neurological research for a long time going
> back to the guy who got a steel bar through his head. Naturally you
> have to be careful with all kind of research so that you can control
> for the desired effect, I agree. In this study they were suffering
> from focal brain damage which means that they have a certain area in
> the brain that is damaged. Much of modern neuroscience is based on
> research done on people with focal brain damage! This is quite
> different from having a completely different brain due to genetics or
> other factors.
>
> Using people with brain damages is not the same thing as using pens
> that do not work when taking IQ-tests, unless you want to decide which
> pens do not work and so on. It all becomes a matter of which aspect
> you study or what you try to deduce logically.
>
> The study has been peer reviewed as well and I believe it was published in PNAS!
>
Pontus Granström
more importance does not make the less important worthless. Same goes
with studies that are published in prestigious and less prestigious
papers. I speculated that since working memory tasks are so heavily
g-loaded and shares an area with g that depends on whiter matter and
n-back is proven to increase white matter that
it would be a neurological foundation for a real increase in IQ. Just
as real as having a gene that says "create more white matter in that
area". This was my speculation! If you take a look in the files
section in which I've uploaded a lot of files you'll find support of
overlapping between g and n-back and data on why working memory
capacity is so closely related to G.
Jonathan Toomim
Intellectual ability and cortical development in adolescents
http://www.nature.com/nature/journal/v440/n7084/abs/nature04513.html
As I recall, that study found that more intelligent kids tended to be
late (but more complete) bloomers in terms of brain development, such
that their cortical thickness was thinner when they were young but
thicker when they were older. Taking that and saying that
intelligence and cortical thickness are negatively correlated is a
gross overgeneralization.
(I just thought that this would be the best way to express my support
for Josh's warnings about overgeneralization.)
Jonathan
Reece
reached on Tetrisfriends.com or available as a separate game in many
"clones") might be beneficial. This would seem to me as being quite
similar to single N-back at a very high N... Players would need to
both remember where each piece was placed (since pieces are invisible
once placed), in addition to updating their memory after every line
clear in order to account for where pieces now are, something which
would only get harder to do as the player makes mistakes.
Josh
intelligent actually had a thinner cortex by late adolescence. The
reason for this was because their brain went through a much more
rigorous pruning process. This makes sense if you read Intelligence
and the Brain by Garlick. He points out that fluid intelligence
depends on identifying abstractions, which are commonalities across
situations. By pruning the connections, information that is relevant
to only specific concrete instances can be lost. This means that a
brain with less connections (assuming the right connections are lost
and retained) can more easily see the commonalities across situations
- which is what IQ tests get at.
On Sep 24, 9:54 pm, Jonathan Toomim <jtoo...@jtoomim.org> wrote:
> I take it by Shaw et al (2006) you mean this?
>
>
> read more »
Jonathan Toomim
Using a longitudinal design, we find a marked developmental shift from
a predominantly negative correlation between intelligence and cortical
thickness in early childhood to a positive correlation in late
childhood and beyond.
"And beyond" includes adolescence and early adulthood. See also
Figure 4.
Yes, the pruning that occurred during late childhood and adolescence
was more rigorous in the higher intelligence groups, but this was not
enough to fully compensate for the greater cortical growth rate and
longer duration of growth in those groups (Figure 3).
Perhaps I misspoke earlier: I meant that the more-intelligent
children had cortices which were thinner when they were young and
thicker when they were older ^compared to the lower-intelligence
groups. I suppose I assumed, given the context, that it would be
clear that I would be making a comparison between rather than within
individuals; that assumption may have been a poor one.
Jonathan
Josh
this is the key message of Shaw's study. Specifically, all
intelligence groups had thicker cortices in early childhood, and this
cortex then thinned out in later childhood. This is accompanied by an
increase in intelligence. This is shown in Figure 3 where all groups
had thinning or decreasing cortical thickness over late adolescence.
Hence, I cannot see how you can claim that thicker cortex leads to
greater intelligence when Shaw's study is clearly showing that
thinning the cortex leads to greater intelligence in late childhood.
Jonathan Toomim
legs then lengthened out in later childhood. This is accompanied by
an increase in intelligence. This is shown in [common sense] where
all groups had leg lengthening or increasing height over late
adolescence. Hence, it is clearly shown that increasing leg length
leads to greater intelligence in late adulthood.
(You could make a very similar argument to "prove" that men think with
their genitalia.)
A) Correlation is not causation.
B) Looking at what changes during development is often uninteresting
unless you're looking at how it happens differently in different people.
Yes, the cortex thins as mammals approach adulthood. Yes, cognitive
ability tends to increase as adults approach adulthood. No, having a
thinner cortex does not increase cognitive ability. Yes, adolescents
and adults with greater cognitive ability tend to have thicker
cortices in the frontal and parietal lobes. No, thickening the cortex
of the frontal and parietal lobes won't necessarily cause an increase
in cognitive ability.
Jonathan
Josh
correlation goes just as much against when you are looking at how
things happen differently in different people. For instance, children
of higher socioeconomic status have higher IQ. Therefore, higher
socioeconomic status causes higher IQ. Is this statement suddenly true
because we are looking at differences across people rather than
development within a person??
The reality is that typically all of the psychological studies are
based on correlation, so causation is difficult to infer. My point was
firstly - you cannot argue causation (thicker cortex causes increased
intelligence) when the correlation is in the reverse direction (cortex
thins at a time that intelligence is increasing). In order to argue
causation, you do need the correlation to be in the right direction,
whether it is across individuals or across age. Correlation is
necessary but not sufficient to establish causation. Second, due to
the problems with correlation, especially in psychology, getting at
causation typically requires identifying the underlying mechanisms. As
I said, it would seem most plausible that thinning of the cortex would
lead to greater intelligence. Do you have a specific argument against
this?
Josh
of the thread is that "increased cortical thickness would lead to
increased intelligence". You may or may not agree with this yourself,
but it is what is being debated in the thread. You then give the
analogy about how leg length increases at the time that intelligence
increases, therefore leg length causes greater intelligence. You
cannot of course conclude that leg length causes greater intelligence.
However, it cannot be dismissed as a causal mechanism, since there is
a correlation here (of course, the positive correlation does occur due
to age causing both an increase in leg length and greater
intelligence).
But this misrepresents the situation being discussed in this thread.
What if leg length increased over childhood, and intelligence did not
increase. Would you then argue that leg length caused the lack of
increase in intelligence? No you would not, because there is not a
correlation here. Leg length is not a plausible cause of intelligence
due to the lack of correlation over childhood. This is exactly the
same as cortical thickness. Cortical thickness can both increase and
decrease over childhood at the time that intelligence increases.
Hence, why would you conclude that increased cortical thickness leads
to increased intelligence?
Josh
argument is. Consider your statement "All intelligence groups had
later childhood. This is accompanied by an increase in intelligence.
This is shown in [common sense] where all groups had leg lengthening
or increasing height over late adolescence. Hence, it is clearly
shown that increasing leg length leads to greater intelligence in late
Ha ha. We are all supposed to see how silly this statement is.
However, let me take exactly the same statement, and just change a few
of the words. "All running groups had shorter legs in early childhood,
late adolescence. Hence, it is clearly shown that increasing leg
Ha ha. What a stupid statement to make... But hang on. It also happens
to be true. Age causes an increase in leg length, and increases in leg
length lead to increased running speed. So far from a developmental
argument being implausible, it is just that the specific example you
chose was implausible because we know, apriori, that leg length does
not cause intelligence.
But according to you, looking at different people such as groups with
differing IQ is plausible. Okay, let's take the following example. The
IQ of children is positively correlated with the IQ of their parents.
Therefore, child IQ causes parent IQ. According to you, this is
plausible since it is based on differences between the groups.
However, again we can see that it is not. Not because it is based on
either developmental or individual difference data, but because it is
based on an implausible causal mechanism (namely the temporal order of
events).
So the example you make is implausible not because of it being about
development, but because you have cherry picked an example where we
know the causal mechanism is implausible.
Jonathan Toomim
Arkanj3l
Arkanj3l
anyway.
Pontus Granström
me supersmart?
Josh
connections that represent abstractions are retained.
My objection to the whole "thicker cortex equals greater intelligence"
is that it is based on an antiquated notion of how the brain works,
ie, all of the neurons do their part, and having more neurons and
connections means that even better processing is able to be performed.
We now know that this is not how the brain functions. The brain is a
pattern recognizer. We learn patterns through experience, and then
apply these patterns in the future to predict the likely outcome of
events. However, if all information was transmitted in the brain, then
it would not be able to recognize the commonality across different
experiences as every concrete experience is different.
By reducing the connections, information that is irrelevant to general
patterns can be eliminated. This means that the brain can more easily
identify a common pattern in different situations. Hence, as was
stated above, it is actually found that cortical thickness declines in
childhood as intelligence is increasing, suggesting that less neural
connections can lead to a better ability to abstract information and
generalize it across situations (which is specifically what
intelligence tests measure).
On Sep 27, 2:37 am, Pontus Granström <lepon...@gmail.com> wrote:
> Josh do you mean that that nonexisting cortical thickness would make
> me supersmart?
>
> On Mon, Sep 27, 2010 at 11:31 AM, Arkanj3l
>
Pontus Granström
in g areas couldn't improve intelligence. Forgive me if my question is
stupid I am quite tired...
Josh
necessary for increasing intelligence. If someone has connections that
they shouldn't have and this is preventing them from identifying an
abstraction, then increasing white matter so that this information is
transmitted around the brain faster would make things worse rather
than better. But this is a lot of speculation...
The point that I have been trying to continually make is not so much
that white matter could not improve intelligence, but that the
original poster seems to be inferring that increasing white matter
would necessarily lead to an increase in intelligence. My point has
been that the brain varies in many more dimensions than white matter
alone, and other dimensions such as the cortex thickness itself would
seem to be more important for intelligence (given what intelligence
test items assess).
Hence, to establish that playing tetris improves IQ, it is not
sufficient to say that playing tetris improves white matter. You
really need direct evidence that playing tetris does actually change
IQ itself. Only once you have this evidence, could you go on and
speculate that the cause of the change in IQ due to playing tetris is
due the change in the white matter.
On Sep 27, 2:07 pm, Pontus Granström <lepon...@gmail.com> wrote:
> Sounds interesting, but does this really contradict that white matter
> in g areas couldn't improve intelligence. Forgive me if my question is
> stupid I am quite tired...
>
>
> read more »
Pontus Granström
intelligence. I would say that being able to apply "logical rules"
like A > B C <B therefore A > C is intelligence. Something that is
very hard to manage without working memory capacity, you need to hold
the information in your mind while executing a comparsion. Someone
might understand that A > B and can answer the question is A > B, but
if I ask him if C < B is C larger than A he might not be able because
he has problems with his WMC. Even though he might understand the
abstraction that something is less than another thing.
Or the problem becomes that what we define as intelligence is not the
same thing as what Gf-tests measure. Sometimes abstraction
removes what is in practice necessary. Solving a optimization problem
might abstractly been described as for example choosing the smallest
weight all the time since this will minimize the total weight, but in
practice we might be able to compare two objects and select the
smallest, something that is working memory capacity demanding. A
implementation of a abstract problem usually pin downs to for loops
and comparsions.
This means as Jonathan points out that the gain of measured
intelligence which relies on many functions might have different
effect depending on where your "weakness" is. Still all people seem to
benefit from IQ-training. If my trouble is a lack of focus when
tackling difficult problems a little n-back might be very beneficial,
if my problem is wmc which causes me to "drop information that needs
to be held online" n-back might be beneficial again. If need to be
faster due to the time limit of IQ-tests n-back might be beneficial
too, if you reach really high levels with a high demand for speed of
the mental updating ef. I would ask someone what underlies a G-change.
What is more g?
I do not believe that why n-back is beneficial is due to a raw
capacity of holding information in the head but rather the subtile
processes that assists this ability and increased.
I am also a bit suspicious to that you mean that solving gf-problems
is associated with the thickness of the total cortex. Since the brain
consists of many "isolated" parts I believe there's reason to believe
that this aplies for G as well. In the article I uploaded they have
identified a "g-area" that seems to be "g itself" or at least a
calibration area. Sure pruning the brain might be beneficial but
having more connections between the brain cells might be too. Having
more connections in a g area might obviously increase our ability to
integrate information. It's well known that women easier can verbalize
emotions due to their thicker corpus callosum.
However I am not saying this is the case, but rather a speculation.
Josh
define as intelligence." A very well-known scientific survey of
intelligence researchers revealed that 99.3% of them believed that
"abstract thinking or reasoning" was central to intelligence. No other
characteristic rated as high. The first definition on the wikipedia
page for intelligence is "abstract thought". Practically every
scientific definition of intelligence will mention abstraction in it.
Fluid intelligence test items become more difficult because they
involve more abstract concepts. Think of word analogies as a prime
example. IQ subtests that assess fluid intelligence such as
comprehension, similarities, arithmetic and picture arrangement all
require using abstractions to determine the correct response. Piaget
is a seminal developmental psychologist who studied the development of
intelligence and emphasized the development of abstract
representations over childhood. If you have never seen anywhere that
talks about abstraction in relation to intelligence then it indicates
you really haven't read any of the literature at all, and there is no
point in trying to discuss it with you until you have read at least
some relevant background material.
Pontus Granström
my view on intelligence. You got your view that a thinner cortex make
ones ability to "abstract" better. Might be some truth to it. As for
your definition of intelligence, there exists multiple definitions.
Some might say it's to grasp what to do, catch on (gottfredson) and so
on. Still this is a general and abstract description seen through her
emotional lens. While someone into neuroscience try to find the
scientific explanation to something. Compare with a doctor and
specialist in cell biology. A doctor says that person seems to be sick
or lacks something, while a scientist try to find a "neutral objective
explanation".
Still intelligence comes down to your score on tests developed from
intuition in the 50's. It's only the last years people have tried to
study what exactly intelligence tests measure what underlies the
performance. Where g is, or if it's only psychometric derived.
G only exists in the sense that finding a sequence for a object moving
or removing similar lines from figures seems to be correlated with
your vocabulary. Because that's the only thing we can measure. A lot
of research points to the fact that working memory capacity almost
explains everything in these kind of problems. I am not surprised at
all. It's very rarely a IQ-test actually contains anything that has to
do with logic to do at all. If you do not consider memorizing movement
logic.
Reece
some ways to make DNB more entertaining:
http://groups.google.com/group/brain-training/browse_thread/thread/e65bf778f9460e2f/b12557ba1ca239a0?lnk=gst&q=gwern+tetris#b12557ba1ca239a0
One interesting one gwern mentioned which caught my eye was an
invisible tetris in which piece N+1 is visible while piece N is not. I
tried playing completely invisible tetris today using Nullpomino
"Phantom Mania" and it was way beyond my abilities (not for this guy
apparently though: http://www.youtube.com/watch?v=jwC544Z37qo [starts
@ approx 5:08]).
Pontus Granström
documentary about the world record in donkey kong similar stuff but
this almost impresses me more. Al though this game is over in 5
minutes which isn't that long while dk goes on for hours.
Reece
some websites/forums "the best Tetris player in the world" -- he's
certainly one heck of a lot better than I am, though I don't anything
about competitive Tetris or if these statements are true. I often
wonder in cases like these (tetris, donkey kong, asteroids, etc) how
much of it was a result of practice and how much of it might result
from higher WMC, faster processing speed, or some other factors. We
would all certainly get better at Tetris + other games with practice,
however I have a very hard time believing most people would ever be
able to play at the level shown in that video.
On Sep 29, 10:05 am, Pontus Granström <lepon...@gmail.com> wrote:
> That's completely insane, I guess he has practiced a while? I saw a
> documentary about the world record in donkey kong similar stuff but
> this almost impresses me more. Al though this game is over in 5
> minutes which isn't that long while dk goes on for hours.
>
>
>
> On Wed, Sep 29, 2010 at 2:33 AM, Reece <rockyb...@hotmail.com> wrote:
> > I stumbled across an old thread in which a post by gwern mentioned
> > some ways to make DNB more entertaining:
Pontus Granström
you point out, not everyone would be able to reach this level even
with practice.
Pontus Granström
that are not frequently used and usually is a process that we want to
minimize by utilising brain training. So by this argument people how
engages i "G-activity" would have a brain pruned of things that are
not g-related.