12 days T-N-B

[y offs et]

After 12 days of T-N-B, my thoughts.

I am having difficulty in remembering sequences of colour. Once, I
thought to recommend having the option to substitute shapes in its
place. Now I realize colour difficulty is a personal weakness, while
shapes is a personal strength of mine, and I was trying to make it
easier. I will develop a colour skill.

I have realized what some may consider a flaw, and others an extra
challenge, in the brainworkshop program. When the chain of instances
double back upon itself, the user must develop an additional skill,
that is, to "stack" one memory visual upon another, in the same
location of the grid.
A grid of say, 25 squares, with the random pick constrained to within
squares not already chosen in the chain length, would fix this.
Again, this is a solution to make the exercise easier, and is open for
debate.

As a qualifier, consider that in T-N-B we are using one auditory skill
and two visual skills.

[Ginkgo]

Do you happen to have any difficulty distinguishing between colors in
general? I don't, just a thought regarding preference of shape over
color.

TNB was immediately difficult and stimulating. As I began to learn the
task, I noticed the tendency to want to verbalize the color, in effect
converting it from a visual, to an internal-auditory. I try to go as
far as possible w/out verbalizing colors as an attempt to strengthen
that visual process.

[y offs et]

Yeah, I tried verbalizing and dropped it. Enough noise in my head
without adding to it.
Anyway, T-N-B is helping me develop "a mind's eye", in that I am
getting better at "seeing" an overlay of what has gone before.
Perhaps if the colours stayed visible for a fraction of a second
longer...

[y offs et]

I think I've come up with an improvement - for me that is.

OLD_STYLE_SQUARES = True
COLOR_8 = (191, 120, 68, 255)

time will tell.

[y offs et]

Day 16 - improvement strategy

Note - Coming to T-N-B from D-N-B, I noticed a key difficulty which
impeded the transfer of previously learned skills.
My solution -
KEY_POSITION = 102
KEY_AUDIO = 108
KEY_COLOR = 101

About my colour weakness, and your whatever weakness, an earlier read
study concluded that the attempt to multi-task results in the transfer
of attention resources from strengths to weaknesses, with the result
of erosion of scores in the strength category. I agree, although
there is a possible strategy here.

IMO the ?-N-B exercise is strongly left brain logic oriented, until
you get behind or lost. Then there is the panic reset, and possible
key stab based on "what?" - a feeling, a guess, a gamble etc.. This
moment, IMO is more right brain motivated. It easy for logical
people, as we most all are overwhelmingly sided, to forget there is a
huge world of non-logical people who base many of their daily
decisions on "hunches". They are called lucky, sleep walking, using
astrology, etc.. I contend they are using sub-conscious resources
that they are more attuned to than people of our strengths, as our
very strength demands we suppress illogical decisions. This is a
field that are yet to be fully understood.

However, I have found that a valid ?-N-B improvement strategy is to
deliberately concentrate on the weakness category, and trusting that
guesses made because of reduced attention in usually high competence
categories, have a high percentage of being correct.

Now, some will say I have just stated an obvious, basic strategy. And
you are correct.
However, this post is not aimed at you.

[Gwern Branwen]

On Sat, Oct 3, 2009 at 4:01 PM, y offs et <tre...@shaw.ca> wrote:
>
> Day 16 - improvement strategy
>
> Note - Coming to T-N-B from D-N-B, I noticed a key difficulty which
> impeded the transfer of previously learned skills.
> My solution -
> KEY_POSITION = 102
> KEY_AUDIO = 108
> KEY_COLOR = 101
>
> About my colour weakness, and your whatever weakness, an earlier read
> study concluded that the attempt to multi-task results in the transfer
> of attention resources from strengths to weaknesses, with the result
> of erosion of scores in the strength category.  I agree, although
> there is a possible strategy here.
>
> IMO  the ?-N-B exercise is strongly left brain logic oriented, until
> you get behind or lost.  Then there is the panic reset, and possible
> key stab based on "what?" - a feeling, a guess, a gamble etc..  This
> moment, IMO is more right brain motivated.  It easy for logical
> people, as we most all are overwhelmingly sided, to forget there is a
> huge world of non-logical people who base many of their daily
> decisions on "hunches".  They are called lucky, sleep walking, using
> astrology, etc..  I contend they are using sub-conscious resources
> that they are more attuned to than people of our strengths, as our
> very strength demands we suppress illogical decisions.  This is a
> field that are yet to be fully understood.

Did you see the uploaded study about latent inhibition & faith in intuition?

--
gwern

[y offs et]

No, but the titles seem to say it all.

By the way, your Flash sig comes thru my Email as an attachment,
taking your reply with it. Thus, every one of your posts presents me
with a blank Email page with attachments. This is the first time I
actually decided to open one.

I'm sure your sig is clever (still haven't found a way to view it),
but I'm not about to change my email settings, and wonder about the
others.

On Oct 3, 9:43 pm, Gwern Branwen <gwe...@gmail.com> wrote:

>  signature.asc
> < 1KViewDownload

[Gwern Branwen]

On Sun, Oct 4, 2009 at 12:17 PM, y offs et <tre...@shaw.ca> wrote:
>
> No, but the titles seem to say it all.
>
> By the way, your Flash sig comes thru my Email as an attachment,
> taking your reply with it.  Thus, every one of your posts presents me
> with a blank Email page with attachments.  This is the first time I
> actually decided to open one.
>
> I'm sure your sig is clever (still haven't found a way to view it),
> but I'm not about to change my email settings, and wonder about the
> others.

It is not Flash anything. It is a cryptographic signature generated by
FireGPG/GnuPG.

If your email service cannot handle an RFC standard for ASCII text
files which was first developed almost a decade before shaw.ca was
even registered, and which would be recognized as such (and not a
Flash file) by even the most primitive 'file' utility... then I would
recommend you not entrust them with a single email, as they are
obviously incompetent at both email & security.

--
gwern

[y offs et]

After frustrating over the colours some more, I made some symbol
graphics. I find I am again advancing.

They, plus my config.ini file, are available here.

http://www.uploadpedia.com/33RZGLXNDCV1/Brain-Workshop-zip.html

[childofbaud]

I'm terribly off-topic, but the Google Groups web interface seems to
be incompetent at e-mail and security as well. What does the sig
actually say? I've always wondered, but the opportunity never arose to
ask before now. :-)

On Oct 4, 1:02 pm, Gwern Branwen <gwe...@gmail.com> wrote:

[Gwern Branwen]

On Tue, Oct 6, 2009 at 11:00 PM, childofbaud <sait...@gmail.com> wrote:
>
> I'm terribly off-topic, but the Google Groups web interface seems to
> be incompetent at e-mail and security as well. What does the sig
> actually say? I've always wondered, but the opportunity never arose to
> ask before now. :-)

It's just some base-64 gibberish; however, it's unique, carefully chosen gibberish such that if you run my email, my public signature, and the gibberish through something like GPG, you can know whether you received my email exactly as I sent it.

--
gwern

[y offs et]

Day 25

Attention Paul!

I am having a bad day, and after fretting over a potential drop in
performance, I decided to try familiar shapes, as apposed to the
previous download. Then I realized alphabetic letters are familiar,
so I made up a selection of. My scores immediately jumped 15% with no
discerned increase of effort on my part - in fact it seemed easy. I
would highly recommend you consider swapping such for the colours for
T-N-B into your next release.

[y offs et]

I think I may have found a possible bug.
Today I had an 8 run at T4B (pats on own back) before getting the
"decreased" message.
At that time I decided to take a break and closed the program.
When I re-opened, I thought I was going crazy, until I realized I was
still at T4B.
Huh?!

[Jonathan Toomim]

I fixed that bug a while ago. You're using 4.4, aren't you?

The bug was basically this: restarting BW would set you at the level
you were at for the last session you completed, and would reset the
below-fallback-threshold session counter to 0. This also meant that
if you got >80% (or whatever your advance threshold is) on a session,
then restarted BW, you wouldn't advance. Newer versions of BW work as
they should, and will advance you or make you fall back at the same
time regardless of when and whether you restart BW.

Thanks for the bug report, though.

Jonathan

[Michael Campbell]

Jonathan Toomim wrote:
> I fixed that bug a while ago. You're using 4.4, aren't you?

I'm not the original poster, but was curious; is there a version past 4.4 for
Linux? The website didn't appear to have one, but I'll admit I didn't look very
thoroughly.

[y offs et]

I am too using 4.4. Doesn't seem to be a link to any newer version
(windows)?

[Reece]

You can get 4.5 for Windows here: http://brainworkshop.sourceforge.net/misc/
.

[y offs et]

Day 33(ish) Time for a recap.

I haven't felt the need to morph my graphics lately, so here is my
final solution.
http://www.uploadpedia.com/33ADLSAW7AJZ/BW-zip.html
Thanks to John for 4.5. Nice graphics, but I'll just continue with
mine.

Raise threshhold 91% - Why? : The wonders of random. In the
beginning, I found my lazy old brain would think it perceived a
pattern and would gear down the concentration, only to be rudely
awakened again, because random has no pattern. The flip side of the
random coin is occasionally it sends a real easy trial, and wow- over
80%, and oh no- completely lost(at least with TNB). I tried 90% but
still didn't feel I had developed a good base.

4 second trials - Why? :
interesting(?) comparison
- a computer is binary, in that, there are two possible bit states (ON
or OFF) - base two.
- DNB is base three - no match, match position, match sound, match
both - four possible states.
- TNB is base four - no match, match position, match sound, match
colour(symbol), match position and sound, match position and colour,
match sound and colour, match all three - eight possible states.
Yes, it seems slow at the beginning in T2B, but in T3B or better it
gives enough time to run through the list, sort of.

20 sessions - yes and no : As I have a high raise level value, I've
been hitting <ESC> frequently when I know I've failed. The rational
is I'm just getting more practice, and in total I may be doing 25-30
trials in total but only documenting 20/day.

progress - from a beginning of struggling with T2B and being lost at
T3B, to now where I use T2B as a warm-up and hit T4B once almost every
day, with 3-8 session flight times. I have yet to hit the T4B heights
twice in one day, but it will be soon.

I'm a 60 year old male trying to stave off the eventual fate of all,
and I am quite happy with my progress.

[y offs et]

It's interesting how doing n-back proves that time is relative and
based upon our perception of it's passing.

When I'm doing well, the next instance comes with metronome exactness
as expected from a machine.
When I'm resetting after a tricky double-back, the next instance
always comes way too quickly, as if a second had been removed. The
same perception happens on an upped level, and it is so persistent.
It's like some time had vanished.

For the longest time I thought the program had a bug, being the mere
human.

On Oct 18, 7:25 pm, y offs et <tree...@shaw.ca> wrote:
> Day 33(ish)  Time for a recap.
>
> I haven't felt the need to morph my graphics lately, so here is my

> final solution.http://www.uploadpedia.com/33ADLSAW7AJZ/BW-zip.html

[Reece]

I've noticed something similar when playing with 1 second trials and
then switching to 2 second trials -- it feels like you have "forever"
to decide whether there was a match or not. After having played with 1
second trials, I now find the default 3 second trials unbearably slow
and actually perform better with 2 second trials than 3 second trials.

When playing PASAT over at cognitivefun.net, I similarly noticed that
upping the level seemed "so much faster" despite actually being
slower seeing as I was playing the Speed Run version which adapts to
the player's speed.

[Gore Lando]

A huge grid would be interesting, but I don't think stacking is a weakness of the program.

Stacking is a cognitive skill, one that's natural to us, and doing it (while keeping everything else straight in your head, etc.) is mental activity, and challenging in this context.

I think avoiding stacking (so likewise with the so-called intuitive approaches) probably doesn't do much good.  The point is to stretch yourself as far as you can -- if stacking allows you to rise to higher difficulty levels, then you're on a higher difficulty level and still challenged.

--~--~---------~--~----~------------~-------~--~----~
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
-~----------~----~----~----~------~----~------~--~---

[Gore Lando]

Oops, I didn't understand your message quite clearly.  You suggest that it would be easier with a large grid!

Could be, for some.  But I think the same argument I made applies ... "stacking" in your mental map probably takes effort and exercises some cognitive function too ... that you find it difficult is indicative.  And difficulty is good, eh?

Personally, I find some doubling back of the positions easier, and sometimes harder.  A sequence like this makes me feel ... like I'm dyslexic, I think:

_Q|__|__|
__|__|__|
__|__|__|

__|__|__|
_A|__|__|
__|__|__|


_Q|__|__|
__|__|__|
__|__|__|

__|__|__|
_A|__|__|
__|__|__|


If that makes sense to anyone.

On Tue, Sep 29, 2009 at 11:20 AM, y offs et <tre...@shaw.ca> wrote:

[y offs et]

60 days - sort of

I would have hoped to be able to report a breakthru to T5B by now, but
it doesn't seem likely in the immediate future. However, following
the progress of new users of N-Back, I'd like to offer some insights
and encouragement.

When I began TNB, I was struggling with T2B and lost in T3B. In that
time I was frantic, racing to keep up. My attention span was so
limited that 60 sec. was too long, and my mind would thankfully know
when the end was nearing with relief.

Now, with 91% advance level, I ace both within 2-3 trials, warming up
to T4B. I've extended to 4.5 sec/trial = 108 sec. @ T4B, and often
the end of the trial is unexpected, as my mind has eased into a groove
of controlled processing. I've found at earlier levels that this is
the precursor to a new breakthru, so I am confident of T5B soon.

On the subject of real world immediate benefits of doing N-Back, I'd
like to recall the saying "luck favors the prepared mind". T2B, or
recalling events twice removed, was a skill I did not possess 2 months
ago, while now it is almost automatically available to me. Thus, in
my other explorations into new software or whatever I study, the
information is now processed with a new mental tool in addition to the
tools already at hand. I have realized that I am learning more
quickly.

The point is, struggling at a level tends to commit the skill of
levels below to automatic, and thus a new mental tool to add to your
faculties.

[y offs et]

Just a quick note. I lied! By some fluke I managed a 91% T4B on the
exact 60th day. Of course, I immediately struck out at T5B, but
hooray anyway.