ScienceDaily - Clarity in Short-Term Memory Shows No Link With IQ

[likeprestige]

Title: Clarity in Short-Term Memory Shows No Link With IQ

Date: 29/11/2010

Source: http://www.sciencedaily.com/releases/2010/11/101129121123.htm

Excerpts -

""The number of things people can remember is robustly correlated with
fluid intelligence -- the larger number remembered, the higher the
IQ."

"Resolution in memory is not predictive of IQ at all."


likeprestige

[likeprestige]

I've uploaded the journal in the files section.

See:
http://brain-training.googlegroups.com/web/2010%2C+Fukuda+%3D+Quantity+not+quality+-+the+relationship+between+fluid+intelligence+and+working+memory+capacity.pdf?gda=nRfR0acAAABL-AAtWJyv-45o99_wkvZJc6H8KMRrLGGmQBx9sEiuuBgFJI4A-Oijr61isSnt4f6_MHAF4RJ3FzYNjhChmZg4ROx9rehCDhQym-QDNGcdOYPP9WV6hLlwDMb9toWyxZaaW44nnuEat6tx6mSshmnTCv3Gg2__lN19RDxxWbDkPnGlHau8Fxisn-AdAGJN6tlhmrMR3uGvvPr01Poh-10xeV4duv6pDMGhhhZdjQlNAw&gsc=4nHs7QsAAADc687o-GeOzJ-Z2ydM9NcZ

likeprestige

[Sarah]

Hi,

Don't post many times.

What does this mean for fluid intelligence training and Brain
Workshop?

Is this a possible argument for why single-n-back may be better than
dual-n-back or than the other little goodies such as pentuple and
multi-stim mode?

Be nice to know.

Thanks everyone,

Sarah

On Nov 30, 7:49 pm, likeprestige <plastic...@live.com.au> wrote:
> I've uploaded the journal in the files section.
>

> See:http://brain-training.googlegroups.com/web/2010%2C+Fukuda+%3D+Quantit...

[Gwern Branwen]

On Tue, Nov 30, 2010 at 4:22 AM, Sarah <opentob...@hotmail.com> wrote:
> Hi,
>
> Don't post many times.

A 4-hour gap isn't unreasonable if that's how long it took him to
track down the paper. A good mail interface (like Gmail's
conversations) reduces the annoyance even further.

> What does this mean for fluid intelligence training and Brain
> Workshop?
>
> Is this a possible argument for why single-n-back may be better than
> dual-n-back or than the other little goodies such as pentuple and
> multi-stim mode?

Seems plausible enough to me. It supports the general idea that the
fancier modes aren't better than DNB or SNB for increasing IQ but may
deliver other benefits; from the Science Daily article:

> "The discovery that clarity doesn't factor into a person's IQ score doesn't suggest that memory resolution is unimportant, the researchers noted. The importance of clarity or resolution of things remembered is indeed vital, for example, to a radiologist studying images of a patient's internal organs with potential disease conditions."

(I've skimmed the paper, but no particularly novel observations come
to mind. Maybe Pontus or Jonathan have more interesting things to
say.)

--
gwern
http://www.gwern.net

[Jonathan Toomim]

Almost everything in Brain Workshop is primarily dependent on working
memory capacity rather than resolution. The only things that I can
think of which aren't are the lure trials/interference mode settings
and (maybe, at least for me) the cartoon faces stimuli and a couple of
the shapes stimuli. Multi-stim mode may also depend on memory
resolution to some extent.

In order for the mode to depend on resolution, the probe needs to be
similar to the target, but not (always) identical. If the target were
blue, the probe might be aquamarine. If the target were a hexagon,
the probe might be a heptagon or octagon. If the target were a circle
of 100 pixel diameter, the probe might be a circle of 80 pixel
diameter. Et cetera. With pentuple n-back, you have 5 orthogonal
dimensions, but on each dimension the target and probe are either
identical or dissimilar, meaning that it's not resolution-based.

By the way, where does this idea come from that single n-back is
better than dual n-back? I've seen it bandied around here a few
times, but I haven't seen any data that support it.

Jonathan

[Gwern Branwen]

On Tue, Nov 30, 2010 at 2:49 PM, Jonathan Toomim <jto...@jtoomim.org> wrote:

> In order for the mode to depend on resolution, the probe needs to be similar
> to the target, but not (always) identical.  If the target were blue, the
> probe might be aquamarine.  If the target were a hexagon, the probe might be
> a heptagon or octagon.  If the target were a circle of 100 pixel diameter,
> the probe might be a circle of 80 pixel diameter.  Et cetera.  With pentuple
> n-back, you have 5 orthogonal dimensions, but on each dimension the target
> and probe are either identical or dissimilar, meaning that it's not
> resolution-based.

So you think the existing sets of stimuli are all too dissimilar and
discrete for this to be relevant?

> By the way, where does this idea come from that single n-back is better than
> dual n-back?  I've seen it bandied around here a few times, but I haven't
> seen any data that support it.

http://www.gwern.net/N-back%20FAQ.html#jaeggi-2010

--
gwern
http://www.gwern.net

[Aman Idle]

So does that mean that the better you perform on digit recal the higher iq or i got that one wrong?

--
gwern
http://www.gwern.net

--

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.

[Pontus Granström]

You got it right, digit span is highly correlated with IQ. Then comes the question, does digit span performance depend on functions casual for intelligence? As I said before, digit span has been immune to the flynn effect for the last 30 years where as RAPM scores have increased by ~20 points.

[Aman Idle]

Gosh i shall resume my training with digit recal then! Someone told me something about it not doing anything sometime ago and i took a break from it.

[Pontus Granström]

I think you just should continue your n-backing since it transfer to digit span tasks, perhaps even more in single n-back (position n-back).

[jttoto]

I wouldn't completely discard resolution of memory's relation to
intelligence. If we are to assume that resolution of memory is
analogous to resolution of proactive interference, then resolution of
memory does correlate rather strongly to reading span, which
correlates to reading ability.

Source:
http://www.interactivemetronome.com/IMpublic/Research/Temporal%20Processing/Brain%20Injury/Research_Executive%20functions_reading_Whitney2001.pdf

This makes sense, since when reading many impertinent words may enter
enter into one's working memory, especially in grandiloquent prose
commonly seen in the GMAT or LSAT. An individual's ability to read
efficiently may depend on one's ability to screen out irrelevant
stimuli, and is extremely important in most standardized test. This
study, as with most, used culture-fair tests, which tend to be visual,
as a proxy of Gf.

[Aman Idle]

hm still got the digit recall on lock. 9 numbers remembered with ease, but backward digit recall only reach about 7.

[Pontus Granström]

The diffrence is pretty normal though. A 9 digit span and a 7 backward span are however above average.

[Benjamin]

I think the reading portion of the LSAT becomes quickly decoupeled
with IQ as IQ rises. More of a floor than a ceiling. I'm pretty sure
it's not much of a differentiator as scores go up.

On Dec 2, 4:38 am, jttoto <jtdem...@uncc.edu> wrote:
> I wouldn't completely discard resolution of memory's relation to
> intelligence.  If we are to assume that resolution of memory is
> analogous to resolution of proactive interference, then resolution of
> memory does correlate rather strongly to reading span, which
> correlates to reading ability.
>

> Source:http://www.interactivemetronome.com/IMpublic/Research/Temporal%20Proc...

[αrgvmziΩ σV]

Has anyone looked into the statistics of digit span? I don't currently
have the time, but if anyone can provide information or a reference to
this, I would be most appreciative. Further, I would like to know what
kind of practice effects are seen with it.
I've long been able to do reverse and forward DS (both auditorily and
visually) at a minimum of 11 items with 100% accuracy. FYI: I have not
trained with any NB variants for the past several months and still
maintain this span (even with a small alcoholic buzz).

argumzio

On Dec 2, 10:31 am, Pontus Granström <lepon...@gmail.com> wrote:
> The diffrence is pretty normal though. A 9 digit span and a 7 backward span
> are however above average.
>

> On Thu, Dec 2, 2010 at 5:29 PM, Aman Idle <aman.i...@googlemail.com> wrote:
> > hm still got the digit recall on lock. 9 numbers remembered with ease, but
> > backward digit recall only reach about 7.
>

> > On Thu, Dec 2, 2010 at 1:38 PM, jttoto <jtdem...@uncc.edu> wrote:
>
> >> I wouldn't completely discard resolution of memory's relation to
> >> intelligence.  If we are to assume that resolution of memory is
> >> analogous to resolution of proactive interference, then resolution of
> >> memory does correlate rather strongly to reading span, which
> >> correlates to reading ability.
>
> >> Source:
>

> >>http://www.interactivemetronome.com/IMpublic/Research/Temporal%20Proc...

>
> >> This makes sense, since when reading many impertinent words may enter
> >> enter into one's working memory, especially in grandiloquent prose
> >> commonly seen in the GMAT or LSAT.  An individual's ability to read
> >> efficiently may depend on one's ability to screen out irrelevant
> >> stimuli, and is extremely important in most standardized test.  This
> >> study, as with most, used culture-fair tests, which tend to be visual,
> >> as a proxy of Gf.
>
> >> On Nov 30, 12:51 am, likeprestige <plastic...@live.com.au> wrote:
> >> > Title: Clarity in Short-Term Memory Shows No Link With IQ
>
> >> > Date: 29/11/2010
>
> >> > Source:http://www.sciencedaily.com/releases/2010/11/101129121123.htm
>
> >> > Excerpts -
>
> >> > ""The number of things people can remember is robustly correlated with
> >> > fluid intelligence -- the larger number remembered, the higher the
> >> > IQ."
>
> >> > "Resolution in memory is not predictive of IQ at all."
>
> >> > likeprestige
>
> >> --
> >> 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<brain-training%2Bunsu...@googlegroups.com>
> >> .

> >> For more options, visit this group at
> >>http://groups.google.com/group/brain-training?hl=en.
>
> >  --
> > 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<brain-training%2Bunsu...@googlegroups.com>
> > .

[Pontus Granström]

The average for students is around 6.

To unsubscribe from this group, send email to brain-trainin...@googlegroups.com.

[wzeller]

Is 6 the average for reverse digit span, forward digit span, or both?  What's the standard deviation?

[αrgvmziΩ σV]

Okay, I finally had a chance to dig around, and found this data table
(best seen in fixed text):

Demographic and behavioral data (mean+/-S.D.) for each age group
Age Group 11–20 21–30 31–50 51–70
N 140 160 150 100
Years of education 11.69+/-2.84 13.96+/-4.33 14.27+/-3.40
13.01+/-3.77
Estimated IQ 100.15+/-6.83 105.54+/-7.28 108.85+/-5.93
109.56+/-6.64
Forward digit span 6.41+/-1.32 6.71+/-1.39 6.31+/-1.29
6.05+/-1.22
Reverse digit span 5.16+/-1.51 5.34+/-1.82 4.39+/-0.85
4.01+/-1.69

Apparently, a reverse and forward span of 11 is a 3 sigma score for my
cohort. Looks like I have work to do. And mnemonics are for
cheaters! :)

Original PDF: http://tinyurl.com/24ygmnf

argumzio

[αrgvmziΩ σV]

Just a note of clarification: training for the digit span task for the
sake of the digit span task is really all I care about. In fact, I
have said myself it is a pointless task, but since it is included in
some psychometric tests, it would be foolhardy not to "train for the
test". ;) Besides, it's jolly good fun to know where one stands in the
statistics before one seriously begins training on it.

One that I really would like to try is the Letter-Number Sequencing
task in which one has to rephrase a jumbled set of numbers in letters
in numeric and then alphabetic order. This is included in WAIS-IV as a
supplement, from what I understand.

As for the theoretical aspects of WM and STM, I leave that to the
psychologists, who probably know better than I. However, one point
does stand out: reverse DS taxes WM more directly; whereas forwards DS
taxes STM alone. It's in the literature somewhere.

argumzio


On Dec 3, 4:31 pm, likeprestige <plastic...@live.com.au> wrote:
> Remember, dual-n-back training does transfer to increased digit-span
> scores. I don't really see it as a necessity to just train on the
> digit-span task.
>
> Jaeggi, 2008:
>
> "Additional analyses showed that there was training-related transfer
> on the digit-span task, but not on the reading span task, in which
> both groups improved equally in the posttest session. However, the
> training time-dependent gain in Gf remained intact after controlling
> for the gain in working memory capacity as measured by a performance
> increase in both the digit-span task and the reading-span task.
> Furthermore, the training-time-dependent gain in Gf also remained
> intact after controlling for the averaged n-back level reached in the
> last training session. In sum, these data indicate that the transfer
> effect on Gf scores goes beyond an increase in working memory capacity
> alone. We discuss this point in more detail below."
>
> I thought there was a study on digit span training, where it concluded
> that training did not transfer to other tasks, no?
>
> Also, I think I uploaded a study a while ago that concluded that the
> digit span task was not an adequate measure of an _adults_ working
> memory capacity but was a good measure for a child's. The adults digit
> span was more correlated to their short-term memory instead. From the
> above study, I wonder if you can now separate the notion between short
> term memory and WMC contributing to the calculation of a persons IQ.

[αrgvmziΩ σV]

> One that I really would like to try is the Letter-Number Sequencing

> task in which one has to rephrase a jumbled set of numbers _and_ letters

> in numeric and then alphabetic order.

Mea minima culpa!

argumzio

[wzeller]

The task is called Trail Making B. It's available on Lumosity.

> To unsubscribe from this group, send email to brain-trainin...@googlegroups.com.

[Dom]

Hi argumzio,

Those are some impressive digit span scores nonetheless. Just a
question... How well do you go at the "spatial working memory
updating" task down at cognitivefun.net?

Thx.

[Benjamin]

Interestingly, I performed the best on the letter-number sequencing
compared to both forward and backward digit span. I'm not sure if
this is anomalous, but for whatever reason I found it much easier than
both forward and backward digit span. If you end up taking the WAIS-
IV, do let what you though about these three WM tests.

Anyone know how Sequencing and forward and backward digit span compare
statistically?

[moe]

About a month ago I trained on the digit span in hopes of far transfer
(A few studies and anecdotes claimed far transfer with digit span
training so I said what the hell) but the only thing I left with was a
higher digit span :(

N-back is clearly a superior WM/Gf training task but digit span
training may be also be useful according to the anecdotes/studies and
it may tax a particular component of WM more strongly than n-back
training does so training on it may be a good idea.

[moe]

Hey Ben, how many digits/characters does those 3 WM tests go up to on
the wais-iv?

[hallu]

I've read the study and it doesn't say anything about the participants
using phonological loops or chunking for remembering. I wonder if they
would get the same results on a fast consonant span task.

> > > >> brain-trainin...@googlegroups.com<brain-training%2Bunsubscribe@go oglegroups.com>

> > > >> .
> > > >> For more options, visit this group at
> > > >>http://groups.google.com/group/brain-training?hl=en.
>
> > > >  --
> > > > 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<brain-training%2Bunsubscribe@go oglegroups.com>

[Pontus Granström]

This is very interesting because according to this table I had a 4 sigma digit span.

To unsubscribe from this group, send email to brain-trainin...@googlegroups.com.

[Dom]

Hey moe, not long ago I cited your very high "spatial working memory
updating" score down at cognitivefun.net.

Did you receive any benefits from training on this task? If not, do
you think it's still work training on it?

Thx moe.


P.S - Pontus, you seem to have given up on the task of reaching 4
boxes, no?

[Pontus Granström]

Forgot about it! Been training some other stuff.

--
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.

[Benjamin]

if i remember right, it "only" went up to 9 for forward, 8 for
backward (but it might have been 9), and 9 (but maybe 8) for number-
letter.

They start off at a mere two or three numbers or letters, give two (i
think) chances at each particular level, then upgrade only if you get
one right. As I recall the fellow who tested me said I got all the
way through all of them but missed a couple on the forward and
backward on the last couple levels but didn't miss even one on the
letter-number sequencing.

I think the difficultly in the WAIS is the cumulative effect of
constantly having to recite numbers is somewhat draining. Although
I've hit well over 9 for forward and backward digit span at cog fun,
the WAIS was quite a bit harder because of the continuous nature of
it.

[moe]

For the most part these games don't "work" for me. My guess is that
if you've trained with a particular game/exercise religiously for 2-4
weeks and are not seeing gains then additional training is probably a
waste of time. In my case the only thing that seems to be "working"
is walking/jogging 4-5 days per week 30 minutes per day and doing math
problems incessantly. My Ef seems to have gone up a bit with this
routine as I'm now able to post passing scores on PASAT-8 (prior to
this routine I could do no such thing).

[αrgvmziΩ σV]

If you're referring to the Corsi block tasks, I haven't really tried
that task, as it doesn't seem all that interesting. Is there a
particular reason you'd like to know?

Maybe I'll try it some time after grinding my skull into DS.

argumzio

[αrgvmziΩ σV]

That isn't quite the format I had in mind, since WAIS-IV, I presume,
you have to repeat a random string of letters and numbers given to you
aurally, which would definitely be more difficult and interesting.
Thanks for letting me know!

argumzio

[αrgvmziΩ σV]

Yeah, seems only that those who have somewhat debilitating cognitive
impairments/abnormalities will see any kind of far transfer on non-NB
WM-related tasks.

Phenomenal WM improvements or no, it's nice to memorize a string of
digits instantly. I suspect that having a high DS will directly
influence one's PASAT performance, so there may be near transfer. And
I need not mention you also have a ridiculous PASAT score, moe. ;)

argumzio

[αrgvmziΩ σV]

I don't use my phonological loop on DS, because it isn't my strong
point. My natural strategy is to map out the buttons I have to press
visually and kinesthetically. I haven't tried doing this in day-to-day
instances, but it might be a good idea to try. Procedural capacity is
unlimited, so the extent to which it can be improved is according to
the degree of use and effort one puts into it.

argumzio

[moe]

"Phenomenal WM improvements or no, it's nice to memorize a string of
digits instantly."

Definitely, I agree.

"I suspect that having a high DS will directly
influence one's PASAT performance, so there may be near transfer."

Increased PASAT performance was a benefit that I was hoping for from
digit span training but the increase in digit span didn't seem to
improve my PASAT performance. It wasn't until I started running/
jogging and hitting the math pretty hard (I guessing these two
activities improve Ef significantly) that my PASAT performance started
to improve.

[hallu]

I do better on the visual-n-back, but actually use my phonological
loop for the first few digits mixed with a visual memory of the
number's shape for the later ones. I also pick up similarities between
the visually remembered numbers.

None of the studies say what a real digit span is. If they have taken
random people for the study and just told them to remember as many
digits as possible, the participants probably used chunking, because
it is a natural behavior to do so for large numbers. But what is the
real digit span, the largest chunk alone or the whole number? Without
clarification on this issue they probably used whole numbers as
representations of the average digit span. Also if the task is called
"visual digit span" but the participant uses his phonological loop,
the task automatically becomes an auditory digit span, only the source
stays visual. This issue isn't addressed as well.

Mapping the buttons. I was thinking about doing this for the digit
span while playing corosi forward and will have to try it (i've
reached once corosi 13, but the game doesn't record stats beyond 12).
I think position n back, corosi forward and mapping the buttons for DS
are similar tasks. It's about remembering the visual movement. For the
digit span it's the movement on the numpad, for corosi it's the
movement on a random surface, position n back is also a movement on a
numberless numpad.

Item span. One of study says, that the larger number remembered, the
higher the IQ, but it is never mentioned which brain process is
responsible for remembering that number. For me there are a bit too
many holes in this field. On the other side there was a study here
showing that there was no difference between mathematicans and
musicians in IQ, so in the end it might not matter what sense people
use to remember the items.

[hallu]

Ok, I've played the auditory digit span and reached the same level as
in corosi forward, 12. "Auditory digit span", that's the name of the
task, but to be honest what my mind was doing didn't have much to do
with the words 'auditory' and 'digit', it was more a visuospatial
span. I believe it is impossible to test visual and auditory spans
separately. If someone has more visual capabilities, his mind will
start unintentionally coding auditory stimuli into visual
representations to the point, where the auditory stimuli will be
transformed into a visual span and become a visual item span test.

Maybe the decision to choose letters in dnb was wrong, because for
visual people it's hard to stop seeing the shapes of the letters in
the place of the n-back position (i was using my phonological loop
instead). Random distinct sounds could have been better for auditory
stimuli, because with them sub-vocalization would be eliminated and it
would be impossible to imagine a sound in a given position.

[αrgvmziΩ σV]

I've done some additional calculations from cogfun, whose stats I
suspect are slightly more representative of the general population
than the aforementioned study's.

Here they are (assume adult-age norms):

Task Mean +/- SD

[v] Reverse DS 4.52 +/-1.721
[a] ReverseDS(en) 4.65 +/-1.774
[v] DS 5.02 +/-1.969
[a] DS(en) 5.56 +/-1.925

What is interesting to note here is that the most difficult task is
visual Reverse DS, while is the easiest is auditory forward DS (en).
This hints that the conversion from visual stimulus to auditory
percept, or phonological loop (assuming this is the modality used, in
general, and considering the auditory mode is easiest for most people,
this is not a mistaken assumption), taxes the cognitive systems
involved in the task slightly more heavily.

argumzio

> >http://groups.google.com/group/brain-training?hl=en.- Hide quoted text -
>
> - Show quoted text -

[hallu]

On Dec 6, 1:30 am, likeprestige <plastic...@live.com.au> wrote:
> When it comes to day-to-day instances would it be just the visualization of
> the numbers?

Bank cards. It occurred to me, that the first thing coming to my mind
while thinking about my PINs is the visual movement/motion I have to
perform on the numpad.

[Reece]

I've discovered that I can increase my performance by ~+2 digits on
both visual forward and reverse digit span tasks (when tested at
cognitivefun.net) if I mentally rehearse the first 6-7 digits
(silently to myself) as they are presented and remember subsequent
digits by mentally representing them visually. When doing such, I
obtain a visual forward digit spans of 10-11 and visual reverse digit
span of 9-10 (I don't chunk the numbers or otherwise make use of other
strategies like mnemonics). I've been taking creatine for a few months
now (following Jonathan's recommendation on this forum awhile back)
and it has provided, by far, the most noticeable increase to my visual
WM and spatial WM of any supplement I've tried -- definitely the
most bang for the buck I've ever gotten from a supplement, that's for
sure. Scores on Lumosity's Memory Matrix have doubled since creatine
supplementation began and I had seen no gain on Memory Matrix
whatsoever for an entire year prior to creatine supplementation. Pre-
creatine, my visual abilities were poor enough that I never would have
even considered trying such a digit span strategy.

Regarding Letter-Number sequencing... I took the Trail Making Test
Part A and Part B on Lumosity and scored much better on Part B (where
one is required to link A-1-B-2-C-3-D-4, etc) than part A (where one
is required to link 1-2-3-4-5-6-7-8, etc). I haven't been able to find
much information on normal times for normal functioning people in
their 20s, however I gather that my 26 second score on Part B is very
quick (or at least appears so in comparison to my 32 second score on
Part A). I know this is not the same kind of Letter-Number sequencing
as Benjamin was discussing acing one the WAIS-IV, however I thought it
was an interesting result, granted it is (from what I've read) rather
uncommon for someone to perform better on Part B unless impaired in
some way which is necessary for good Part A performance.

[Aman Idle]

Are you vegetarian if i may ask? I also take creatine but i must admit i do eat meat!

--
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.

[Reece]

I eat very little meat (most of my meat consumption comes from a
weekly pepperoni pizza lol) -- I was one of those creatine-deficient
vegetarians who just didn't know how much better my mental world could
be with a little more meat / creatine :)

> > brain-trainin...@googlegroups.com<brain-training%2Bunsubscribe@go oglegroups.com>
> > .

[Aman Idle]

Lucky you! I eat at least a cheeseburger a day: / i might not benefit much then! I can't be bothered to go modafinil route although i will try at some point it b4 i graduate in may, but for now i'm looking for alternatives!

To unsubscribe from this group, send email to brain-trainin...@googlegroups.com.

[Pontus Granström]

Shame on you both, I had porridge and coconut oil for breakfast. =)

[Aman Idle]

lol, perhapds you're naturally gifted. Some of us have to look for that thing to get us going. I have tried various nootropics, even racetams (pira and aniracetam) still nada!

[Pontus Granström]

You know since I am used to my "brain boosters" it's hard to notice the effects since I am used to them. If I stopped taking mainly omega-3 and astaxanthin who knows how I would be.

[Aman Idle]

astaxanthin is an anti oxidant right?  Have you heard about curcumin on mind and muslce forum? It's another anti oxidant that has promising benefits for cignition, many deceases and depression. Rervatrol is another one. By the way do you mind listing what you take apart from those two?

[Pontus Granström]

Omega-3 (10+ years)
astaxanthin. (3+ years)
Coconut oil (breakfast) (since Karl posted about it)
Spirulina (before/after training) (a few months)
Creatine (after training) (a few months)

I read an interesting article yesterday about how spices can be used for therapeutic purposes, that was very nice since I usually add a lot of basil,garlic and oregano to my homemade pasta sauce. I never ever get a cold either, mostly due
to high intake of astaxanthin and d-vitamin (from spirulina).

[Aman Idle]

Oki in terms of getting smarter, omega would be the obvoius one on your list.

[Aman Idle]

The rest i guess is more overall health. What's your omega 3 ratio and do you taek liquid? I have heard that more DHA is important for a smarter brain.

[Pontus Granström]

I use eskimo brainsharp a very clean and natural fish oil but it also contains q-10 and some other goodies.

[Aman Idle]

hm is that molecularly distilled and free of all forms of heavy metals? I buy most stuff on iherb, they dont stock eskimo there though!

[Pontus Granström]

Yes it's very clean and got a lot of positive clinical studies.

[Aman Idle]

cool i probably give it a go, perhapds becasue of the research behind it. Do yuo reccomend pills or liquid? Can you fin the research on pubmed or where is it published?

[Pontus Granström]

Here's there homepage they got the studies there as well.

[Pontus Granström]

http://www.cardinova.com/

[Aman Idle]

hm no publications found under eskimo brainsharp http://www.cardinova.com/research.asp?productId=16

[Pontus Granström]

No but on the fish oil they use.

[Aman Idle]

But i want the brainsharp one. That sounds more appealing, but no research behind that one.

[Pontus Granström]

No but they use the same oil for all the products they just add some other stuff depending on it's specialisation.

[Aman Idle]

awesomw, thanks. Don't enough about the rest on ya list, but someone told me unless u really really need an anti oxidant then u better without so ill think about it. Curcumin is still tempting though!

[αrgvmziΩ σV]

Hello, likeprestige. My apologies for not answering your question
earlier. Time management has become an absolute necessity for me these
days. I will, at any rate, proceed to answer your question:

In accordance with my previous message, I do not exactly mean mere
"visualization" of the numbers. That is, the visual symbols that are
taken as numerals by most people are not what I visualize. Instead,
I've taken the number pad on your keyboard as a matrix onto which I
may map continuous paths among the numbers throughout this 3x3 space.
For instance, given a possible string of "4 8 3 6 8 1 5 1 2 3 5 4 4
4", I would map it out like so:

(4→8→3) (6→8→1)
o x o o x o
↗ ↗
x o↘o ⇆ x ↙o o br⇆

o o x x o o

(5→1→2→3→5) (4→4→4)
o o o

o x o ⇆ x3
↙↖
x→ x→ x

(Best seen as fixed text.)

If it isn't clear, then to describe the basic idea: through geometric
shapes and lines, I connect the points on the number field. This is to
serve as the same fundamental operation that I would be using as if
they "really" were numbers. (To be clear, there is nothing "numerical"
about any particular thought. The degree of lateral thought is so
deeply embedded in our thinking that it doesn't really matter if
something "really" is what one purports it to be. All that matters is
that the basic operation remains the same. In this case, following
these relationships serves the same purpose as repeating a string of
numbers aloud or in one's head phonologically.) I manipulate the
numbers in this way in an intuitive and automated manner, such that
the numbers become appropriately chunked. In some cases, I lose focus
near the end and memorize the single digits, such that I'll have two
chunks, and a string of x separate digits.

Incidentally, I've only a few minutes ago managed to score nearly 100%
at n=14 on [v] Reverse DS (cogfun). The only reason it isn't perfect
is because I accidentally switched two numbers during my first trial
at that level. I'm now working on n=15. I'm discovering that moving to
higher n-levels isn't all that hard to do.

argumzio


On Dec 5, 6:30 pm, likeprestige <plastic...@live.com.au> wrote:
> argumzio, sounds interesting. Not sure I fully understand though. When

> it comes to day-to-day instances would it be just the visualization of

> the numbers? However, this method wouldn't really include what you
> suggest... making it a more procedural activity (mapping out the
> numbers with buttons).
>
> Thanks,
>
> likeprestige

>
> On Dec 6, 8:27 am, αrgvmziΩ σV <argum...@gmail.com> wrote:
>
> > I don't use my phonological loop on DS, because it isn't my strong
> > point. My natural strategy is to map out the buttons I have to press
> > visually and kinesthetically. I haven't tried doing this in day-to-day
> > instances, but it might be a good idea to try. Procedural capacity is
> > unlimited, so the extent to which it can be improved is according to
> > the degree of use and effort one puts into it.
>
> > argumzio
>

> > > > > > >> --
> > > > > > >> 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<brain-training%2Bunsubscribe@go oglegroups.com>
> > > > > > >> .
> > > > > > >> For more options, visit this group at
> > > > > > >>http://groups.google.com/group/brain-training?hl=en.
>
> > > > > > >  --
> > > > > > > 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<brain-training%2Bunsubscribe@go oglegroups.com>
> > > > > > > .
> > > > > > > For more options, visit this group at

> > > > > > >http://groups.google.com/group/brain-training?hl=en.-Hide quoted text -

[Benjamin]

That reminds of a (not terribly) interesting anecdote.

Back when I was working retail, I frequently had to remember 7 digits
numbers associated products to enter them into the computer for check
for the "sales" we had all the time. One day I was up on a ladder
look at 2 paris of in-line skate and someone wanted to know if either
was on sale. I was bereft of both pen and pencil, but not a good
working memory. I looked at both numbers and rehearsed one
audiotorially and visualized the other...and success...remembered them
both. I ended up using this a few more time...to keep myself from
going crazy doing a medial sales task.

:)

[polar]

Hi Reece, great that creatine worked for you - may I ask, have you
maintain the benefits long-term (or are you still using creatine)?

[Reece]

Hi polar,

Since this September, I have been taking a creatine dose of ~ 1.0g
t.i.d. (from what I've read, creatine is better absorbed when a small
amount is taken more frequently)

It's been working quite well since the beginning, the only noticeable
"side effects" being increased thirst and a tendency to overheat more
easily when exercising if not properly hydrated. The overheating is
quite noticeable if I really exert myself (eg. running at 10mph on my
treadmill), however it doesn't cause me any problems whatsoever when
not exercising.

[Aman Idle]

what  does t.i.d. mean?

--
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.

[Działo, Christopher]

Aman,

t is an abbreviation for "ter in die" which in Latin means three times a day.

[Aman Idle]

Pontus im trying to order brainsharp over to the uk http://www.cardinova.com/products.asp?productId=16 but i can't see where to place the order. How did you do it?

[Pontus Granström]

I got it from a store.

[Aman Idle]

oh oki, i guess i gotta head over to iherb for my fish oil. Could not really see where to order brainsharp online.

[Aman Idle]

ah found it on amazon. Do you take pills or liquid?

[Pontus Granström]

Pills.

>>> >>>>>>>>>>> > > brain-trainin...@googlegroups.com<brain-training%2Bunsu...@googlegroups.com>

>>> <brain-training%2Bunsubscribe@go
>>> >>>>>>>>>>> > > oglegroups.com>
>>> >>>>>>>>>>> > > .
>>> >>>>>>>>>>> > > For more options, visit this group at
>>> >>>>>>>>>>> > >http://groups.google.com/group/brain-training?hl=en.
>>> >>>>>>>>>>>
>>> >>>>>>>>>>> --
>>> >>>>>>>>>>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>>>>>>>>>> For more options, visit this group at
>>> >>>>>>>>>>> http://groups.google.com/group/brain-training?hl=en.
>>> >>>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>> --
>>> >>>>>>>>>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>>>>>>>>> For more options, visit this group at
>>> >>>>>>>>>> http://groups.google.com/group/brain-training?hl=en.
>>> >>>>>>>>>
>>> >>>>>>>>> --
>>> >>>>>>>>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>>>>>>>> For more options, visit this group at
>>> >>>>>>>>> http://groups.google.com/group/brain-training?hl=en.
>>> >>>>>>>>
>>> >>>>>>>> --
>>> >>>>>>>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>>>>>>> For more options, visit this group at
>>> >>>>>>>> http://groups.google.com/group/brain-training?hl=en.
>>> >>>>>>>
>>> >>>>>>> --
>>> >>>>>>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>>>>>> For more options, visit this group at
>>> >>>>>>> http://groups.google.com/group/brain-training?hl=en.
>>> >>>>>>
>>> >>>>>> --
>>> >>>>>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>>>>> For more options, visit this group at
>>> >>>>>> http://groups.google.com/group/brain-training?hl=en.
>>> >>>>>
>>> >>>>> --
>>> >>>>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>>>> For more options, visit this group at
>>> >>>>> http://groups.google.com/group/brain-training?hl=en.
>>> >>>>
>>> >>>
>>> >>> --
>>> >>> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >>> For more options, visit this group at
>>> >>> http://groups.google.com/group/brain-training?hl=en.
>>> >>
>>> >> --
>>> >> 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> >> For more options, visit this group at
>>> >> http://groups.google.com/group/brain-training?hl=en.
>>> >
>>> > --
>>> > 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<brain-training%2Bunsu...@googlegroups.com>

>>> .
>>> > For more options, visit this group at
>>> > http://groups.google.com/group/brain-training?hl=en.
>>> >
>>>
>>> --
>>> 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<brain-training%2Bunsu...@googlegroups.com>

[Aman Idle]

thanks.

[Pencil]

Last time I checked DNB improves (possibly) working memory.

[ailambris]

You guys realize that the fish oil you are taking is most likely ethyl-
esterized and highly oxidative? Why not consider a micro algae source
of DHA instead?

[Aman Idle]

wait what does that mean? I just placed the order : /

you mean it's not safe? Pontus said it has a lot of research behind it!

[Mike]

Ailambris, what algae brand do you buy? 

That fish oil is highly oxidative isn't a sign that it's a potent anti oxidant? can a product be very stable at room temperature be a good  antioxidant? I'd say yes, because vitamin C stays on the counter for weeks and still is a potent antioxidant... how does the molecule not bind to oxygen while on the counter, but does when in the body? I guess this is out of scope and I should google this, but if you have quick explanation I'm curious.

also I always read that fish oil was better than flax seed oil and algae EPA/DHA. but I don't know why. if you have other good reasons for favoring algae EPA/DHA, please tell! this is very important.

[Mike]

Maybe the decision to choose letters in dnb was wrong, because for
visual people it's hard to stop seeing the shapes of the letters in
the place of the n-back position (i was using my phonological loop
instead). Random distinct sounds could have been better for auditory
stimuli, because with them sub-vocalization would be eliminated and it
would be impossible to imagine a sound in a given position.

...Hi Hallu, 

This is very interesting. I thought (and posted) about this before. I noticed that I improved my phonological loop--unconscious/automatic remembering of sounds--mostly by playing Quad n back for a few weeks. Having to concentrate on the shapes and colors and positions forced me to compute the letters completely unconsciously--something I wasn't doing before, when playing dnb only: rather I was translating sounds into letters that I'd write on the board. I would remember sounds better that way.

After a few weeks and months of quad n back training (during which I played very little dnb), I tried dnb again--I noticed that I could then play up to dual 6-7 back without the need to visually 'write down' letters on the board. remembering audio letters was completely automatic. passed those levels I had to rely on writing on the board again to get good scores. During the few months where I trained only at quad, I only played 3 or 4 times at dnb, yet the scores were in line/curve with my previous improvements at dnb--in other words I continued improving at lot at dnb without training at dnb. 

I think when you combine the phono loop and visual strategies + chunking and others, you get the best scores at dnb. Probably this is also true on other tasks. Probably that a good integration of both visual and phono wm boosts intelligence the most. 

Maybe I would have have developed my phono loop anyways, by just training at dnb. But maybe really playing quad pushed my phono loop to new levels.

It's hard to tell if this improved 'ability to remember sounds as such' gave me substantially different real life benefits than what my previous training at dnb alone would have given me. it's hard to tell, because I definitely improved my visual memory as well. and maybe both are related--both rely on some common mental processes/networks. I don't know if what I noticed is a definite indication that I improved a mental process completely different than what I was using before--but it looks like it, at least to an extend. 

It seems--as you noticed, too--that you can train at dnb yet mostly train your visual memory. your performance on a memory task like digit span can rely on your phono loop, or on more visual strategies. another indication that they are 2 separate things (chemical properties/processes/networks, who knows) is that the phono loop seems more limited, and seems harder to expand than visual wm, at least in my case. It doesn't exclude that they share processes, but it indicates that at least some of their parts are independent.

I have a spontaneous feeling that visual wm is more important for general intelligence and reasoning than audio wm, but the phonological loop (audio wm) could be useful and good to have too (and I agree with you we should try to increase it more actively--with qnb, or with sounds that don't have conventional visual equivalents/symbols. btw cognitivefun.net has german letters this can be an alternative to using unknown sounds). maybe both visual and audio wm can act in synergy on some tasks. eg I think when you combine the phono loop and visual strategies + chunking and others, you get the best scores at dnb. it is only when I start using both visual writing down letters in combination with the instinctive phono loop that I reach my best scores.

Maybe that the phonological loop, while it seems at first sight to be less important for general intelligence, is still important for certain tasks like listening to a lecture and conversations for example. it doesn't seem so important to me for internal reasoning, but it could be useful for audio input, eg for learning in school, social life, paying attention to what others say--being more audio input friendly and extroverted, instead of introverted. I did notice improvements along those lines in my life, but maybe they are not due only to my improved phono loop.

I recently saw an interview from Feynman in which he said he once compared the way he was doing a mental task (counting words in a paragraph or something) with a fellow mathematician. Feynman was using his phono loop, and couldn't talk or sing while doing the task, while the other guy was seeing numbers and he could sing while doing it. 

Personally I know that when I play the piano, I can talk at the same time if I play a walking bass or a repetitive pattern on the right hand--but I can't talk if I'm playing melodies or something that has an irregular rhythm. irregular rhythms, more complex pieces, and improvising seem to be using my phono loop--or at least some 'mental space' that my phono loop is also using. a sign that expanding your phono loop might actually be very useful for performance on some tasks. 

also: listening to music hinders performance on many tasks. Singing while working is something a lot of people do when the work is repetitive, but when the task is more cognitively demanding, it hinders performance. When I sing while working, even if it's just cleaning my apt, I seem to be less efficient and often find myself doing more repetitive or unthoughtful actions. an indication that the phono loop is linked to focus and concentration?

So these things seem to indicate that the phono loop, or the mental processes it's using, is actually very useful for many tasks, even internal reasoning, and even tasks apparently unrelated to audio input. if you can't sing a song while doing a task, it's probably because you need your phono loop for it, so increasing the phono loop might be very important for many tasks, even some that apparently aren't directly related to listening to sounds. certainly visual and audio working memories can act in synergy, certainly even for reasoning.

I would be very curious to know here in the forum 1/who, while playing dnb, gets better scores at audio input (and vice versa) and 2/who uses the phono loop spontaneously at dnb rather than the 'writing down letters on the board' strategy. a long time ago I remember a friend of mine found it strange when I repeated some directions I had received verbally (left, right, left, left, etc.) into visual information, with gestures, to remember it better. I think he was more relying on the verbal memory and was directly putting it into actions rather than translating it into visual working memory as an intermediate step. this I think is interesting and can account for some personality differences, for example the aptitude of listening and applying procedures, versus what we could describe as being an autodidact or a bottom up learner, often people who have more trouble at relying on directives or procedures. One does not exclude the other of course, but there could be a tendency for some people to rely more on one or the other. I think it's an interesting topic and it might give us some insight on the very different outcomes we can see in people with the same iq--how the brain spontaneously works to store information.

2010/12/6 hallu <hall...@gmail.com>

Ok, I've played the auditory digit span and reached the same level as
in corosi forward, 12. "Auditory digit span", that's the name of the
task, but to be honest what my mind was doing didn't have much to do
with the words 'auditory' and 'digit', it was more a visuospatial
span. I believe it is impossible to test visual and auditory spans
separately. If someone has more visual capabilities, his mind will
start unintentionally coding auditory stimuli into visual
representations to the point, where the auditory stimuli will be
transformed into a visual span and become a visual item span test.

Maybe the decision to choose letters in dnb was wrong, because for
visual people it's hard to stop seeing the shapes of the letters in
the place of the n-back position (i was using my phonological loop
instead). Random distinct sounds could have been better for auditory
stimuli, because with them sub-vocalization would be eliminated and it
would be impossible to imagine a sound in a given position.

On Dec 5, 10:27 pm, αrgvmziΩ σV <argum...@gmail.com> wrote:
> I don't use my phonological loop on DS, because it isn't my strong
> point. My natural strategy is to map out the buttons I have to press
> visually and kinesthetically. I haven't tried doing this in day-to-day
> instances, but it might be a good idea to try. Procedural capacity is
> unlimited, so the extent to which it can be improved is according to
> the degree of use and effort one puts into it.
>
> argumzio

[Mike]

As a summary, I think that:

1/the phono loop and visual wm seem to be--at least in part--independent processes; they can be stronger or weaker in different individuals, and, it seems, they can be improved relatively independently from one another.

2/quad n back and other n backs that prevent the translation of sound inputs into visual symbols might help improve better some brain processes that are more specific to the phono loop.

3/these 2 processes (phono and visual wm) could be used in synergy together on certain tasks (on dnb for a thing), for a superior performance.

4/spontaneously relying on one process more than the other during everyday tasks could maybe lead to what we perceive as personality differences. but we would need a survey for this.

==

here is a straightforward summary of the Baddeley and Hitch model of working memory which totally confirms most of what I perceived.

http://www.simplypsychology.pwp.blueyonder.co.uk/working%20memory.html

it's all there, really. I'm wondering if this is still accepted as the model for working memory today, but it seems so. it could be improved, for sure.

one of the things I'm most curious about is how a deficit in one or the other relates to things we notice in people in everyday life behavior.

also the central executive, which sounds to me like a CPU, seems very mysterious. that's where reasoning would take place, like confronting contradictory statements for example. Cev recently posted a study and some games that trained reasoning.

2011/2/4 Mike <mike...@gmail.com>

[hallu]

On Friday, February 4, 2011 10:42:28 PM UTC+1, Mike wrote:

I recently saw an interview from Feynman in which he said he once compared the way he was doing a mental task (counting words in a paragraph or something) with a fellow mathematician. Feynman was using his phono loop, and couldn't talk or sing while doing the task, while the other guy was seeing numbers and he could sing while doing it. 

Hi, I saw this too. He also mentioned he could read while counting in his head (while using his phono loop).  

Personally I know that when I play the piano, I can talk at the same time if I play a walking bass or a repetitive pattern on the right hand--but I can't talk if I'm playing melodies or something that has an irregular rhythm. irregular rhythms, more complex pieces, and improvising seem to be using my phono loop--or at least some 'mental space' that my phono loop is also using. a sign that expanding your phono loop might actually be very useful for performance on some tasks. 

There is a brain region (sorry, I couldn't find the english terms) which is responsible for both phono loop and rhythm. Since one WM can't be used for two things at once (according to the Baddeley's model), speed readers use this fact to eliminate subvocalization by taping a short rhythm with their right hand while reading (usually ..--). Using phono loop for something different than reading (like counting inside your head) is used too. I do not believe in speed reading claims, but reading in such manner is possible. However, I couldn't achieve that eye-no_sound_meaning transfer. People call it "that hidden second voice", but I think it's the mind memory of the sound. And if someone practiced subvocalization for 20 years, it is difficult to get rid of it.