Technical issues with Brain Workshop

[Matan]

I just switched to Brain Workshop from other N-back program and I'm
having some problems.
I plan to play 2 types of N-back's. In order to have easy access to
each of them and not mix the stats between them, I created 2 different
shortcuts to the exe file, adding different "--configfile X.ini --
statsfile X.txt" to the target location under properties.

For some reason, the N level getting reset every day.
During the day, the sessions progress is getting tracked in the
"today's last 20" list and the N level is stayed in the correct level
even if I close the program.
But the day after, the "today's last 20" is cleared(I suppose it
should be like that), but the N level getting reset to D2B as well.
The stats file still hold the record of the previous day so i'm not
sure why does it happen.
(I don't know if it is related to the problem but I also can not
create new user - it creates it but when I close the program i'm
getting an error message, and when I reload the program this user
disappear. Iv'e used the "--configfile X.ini --statsfile X.txt"
method I described above as a wor around and it did create new config
and stats files).
What can I do?

Another question I have is about the trial's per session.
From the config file I can see it is calculated like this "NUM_TRIALS
+ NUM_TRIALS_FACTOR * n ^ NUM_TRIALS_EXPONENT".
With the default values, it means that by the time I reach to D5B, the
training time per day for 20 sessions jumps from ~25 minutes to ~50
minutes, with each session lasting ~2.5 minutes(assuming default 3
second per trial). Should it be like that? Because it wasn't like this
in my previous program.

Thanks.

[foobeer]

Hi Matan,

Matan wrote/schrieb:
> ...

> For some reason, the N level getting reset every day.
> During the day, the sessions progress is getting tracked in the
> "today's last 20" list and the N level is stayed in the correct level
> even if I close the program.
> But the day after, the "today's last 20" is cleared(I suppose it
> should be like that), but the N level getting reset to D2B as well.
> The stats file still hold the record of the previous day so i'm not
> sure why does it happen.

I think that is intended, you could alter the starting-level in the
config.ini:

BACK_DEFAULT = 2

(this is Level 2 for DNB, other xNB below that line, see comment above in config.ini) Suiting this to your needs should not make much work.
This however, would make your stats inconsistent to some extent because
your D2B performance would not be incorperated. On the other hand, this
is hardly important for most people. :)

> (I don't know if it is related to the problem but I also can not
> create new user - it creates it but when I close the program i'm
> getting an error message, and when I reload the program this user
> disappear. Iv'e used the "--configfile X.ini --statsfile X.txt"
> method I described above as a wor around and it did create new config
> and stats files).
> What can I do?

What is the error-message?

> ...

Bye, foobeer

[Matan]

OK I found the "BACK_DEFAULT = 2" but I still don't understand how do
I avoid my N level getting reset to level 2 every day?
If I played for few days and I can comfortably play D4B, why would I
want to start with D2B every day? I don't want to get back to the
default level every day - I want to start the day with the level I
left off on the last day.

The error message refer me to "brainworkshop.exe.txt" which has the
following message:
http://www.text-upload.com/read.php?id=283111&c=8545656

Lastly, if someone can comment about my " trial's per session"
question it will be nice.

Thanks.

[Włodzisław]

change back default to a different number???

anyway dude stop crying over nonsense it takes few seconds to set it up.

no big deal.

[Paul H]

Starting at level 2 is intended as a mental warm up and to keep the exercise varied. If the n-back level is changing it's less likely that you will fall into using set strategies, which could help transfer the benefits of n-back training to other tasks. It should only take a few minutes to work from level 2 up to your max level, assuming you really have mastered the lower levels.

That said, in the next version of BW there will be a new setting which allows you to switch off the behaviour of resetting to the base n-back level every day. The level stays the same as your previous game. This feature is already implemented in the SVN version of BW. 

Paul

--
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.

[Jonathan Toomim]

1. Ignore the "AttributeError: 'Win32EventLoop' object has no attribute
'_next_idle_time'" message. It's a minor bug in pyglet with no negative
consequences that I know of except that error message.
2. There's a bug in BW 4.8.1 and earlier that causes a whitespace
character to appear at the start of the new user text entry field. If
you don't delete that character, user account creation will fail. This
bug has been fixed in the development branch, and will not be present in
4.8.2 or 4.9 or 5.0 whatever we decide to call the next release.
3. Resetting the N level every day was an intentional design decision.
That's how the program was intended to work. Specifically, it resets at
4:00am each day. The development branch has an option in the config.ini
to not reset the n level each day. Part of the motivation for this
decision is that it is still likely to be useful to spend some time at
the lower N levels so you can work on getting up to 100% accuracy at
lower working memory loads. My suggestion is that you set BACK_DEFAULT
to whatever the lowest n-level is at which you do not consistently get
100% correct, and just work through to your max N level each day as
quickly as you can.
4. The trials-per-session thing was another design decision. Mine, in
fact. One of the ideas behind it is that as you advance to higher
levels, you can train your extended concentration ability in parallel to
training your working memory capacity. I wouldn't suggest aiming for 20
sessions at the higher N levels, though. It's almost certain that what
matters for training efficacy is the amount of time spent training, not
the number of "sessions" of arbitrary length you perform. If you want
the trials-per-session behavior to mirror your previous program, set
NUM_TRIALS_EXPONENT to 1. (The development branch also lists the amount
of time you've spent each day. Something else to look forward to, once I
get my act together and fix the bugs that I recently introduced and
release the darn thing.)

For either of the two design decisions I mentioned (#3 and #4), if you
think that the default configuration is lame, please tell me so. If
enough people tell me that they'd like to see the defaults changed, I'll
probably change them.

Jonathan

[whoisbambam]

Włodzisław,

Noone is crying.

Ppl come from varying backgrounds of technical ability.

If he wants to cry, he can. This is a free world.

If you don't like freedom of speech, freedom to question, that is
fine--but do not impose your values onto others.

If you do, then you will hear my POV as you have here, whether YOU
like it or not.

And I personally will pursue freedom to ask questions to any degree--
make your move.

Respect the rights of others, please.

[Matan]

Thanks everyone for replying.

Jonathan, I now understand the design decision you mentioned in #3 and
#4 and it make sense. I will give it a try.
I think an important improvement for #3 will be the option to choose
to cycle back "X" levels each new day. So If I reached N7 and set in
config.ini to cycle back 2 levels, it will start with N5 the next day.
It will act as a warmup and train lower working memory loads.
I agree with your reasoning behind the choice made in #4. BUT for that
to be properly used, you should list the amount of time spent each day
like you have in the development branch. As you said, the actual
sessions does not matter but the time you spend on the task. Since the
N level varies a lot with your default config(and with that the time
on each session varies A LOT), the sessions per day becomes irrelevant
- the time spent is what matter's. Currently I will calculate things
manually every day. It will be much nicer to have it calculated by the
software.

Matan.

[whoisbambam]

Matan,

the time AND imo the LEVEL of difficulty for your brain.

I am confident that the harder you struggle with a given level (while
still be able to track with some accuracy) is extremely important, and
to do this, preferrably, for 30minutes uninterrupted, 5days a week,
for 6months, after which time, imo, there would be little cognitive
growth benefit for many of us--therefore, perhaps a maintenance of
once a week.

Having said this, i was too frustrated to do so. I always wanted like
70% accuracy, manual N-back. This was probably a mistake, but i just
couldnt handle it.

furthermore, when i first started, i couldnt even do 30minutes
straight. I tried various stimulants--i cant remember all of them. But
I know that I liked to take a cup of hot cocoa prior (about 2 heaping
tsp of hershey's non-alkalized baking cocoa). And, I liked to take a
swig of chocolate milk about 10minutes prior, and then after each
session, I would take a swig before playing the next. And, after
15minutes, i took a short nap (10-20min). And, I placed peppermint
under my nose--very concentrated NOW foods peppermint oil.

when i first started, at dnb2, struggling, it was extremely mentally
exhausting.

I only got up to dnb5, 2ticks, average 80% or so accuracy (a couple
games 60s cause i was distracted, a few hundred percents, but mostly
80s-90s)

i dont play much now.

and my ADD has returned.

i can only do dnb5 at 2.3ticks with the beforementioned accuracy now.

i did want to master dnb6 and then quit--maybe someday.

i do believe it helped my brain, but i do not believe it improved my
IQ

i remain very skeptical that dnb can improve IQ