Progression Advice?

[cucumbe...@gmail.com]

Hi everyone,

I've been doing dual n-back training for exactly 1 week but unfortunately had some slow progress today. The first two days I was stuck between to D3B and D4B and felt I wasn't making any progress at all, so on day three I decided to use the following progression strategy:

1. choose n at 5 seconds/trial
2. practice until I get 100% twice in a row
3. decrease the time by 0.1 - 0.2 seconds/trial
4. repeat steps 2 and 3 until I get 100% completion twice in a row at 3 seconds/trial
5. progress to n+1 at 5 seconds/trial and start at step 2

I used this strategy at D3B and progressed fairly smoothly shaving off 1 second / trial / day. My progress at D4B however was was slower progressing only about 0.5 seconds / trial / day for the first two days, and today I only progressed 0.3 seconds/trial. So I'm currently on D4B at 3.7 seconds/trial in the 80%-95% range and starting to question whether I should try a different progression strategy (or at least some strategy for overcoming plateaus).

So...

1. What do you believe is the most efficient general progression strategy and why?
2. What do you believe is the most efficient strategy for overcoming plateaus and why?
   

I also noticed when I do make a mistake, it is because of certain types of stimuli sequences. For example, in the sequence (S left, L right, S left, L right, L left, L left, S right) S and L are somewhat phonetically similar and there's a lot of overlapping positions. Updating these sequences slows me down dramatically and usually results in me making a 1 or 2 mistakes. How do you force yourself to maintain your speed when faced with these kinds of sequences?

[nemamea 1]

There is no such thing as linear progress in brain training and there is no successful or non-successful brain training strategy. Whichever brain training app/mode/strategy you choose, you will come to some point where you will not be able to recognize any changes on your cognition or score (results). If you have a plan to get 10 or 20 IQ points in few weeks you can forget about it. You will either accept this empirical fact, or you will quit brain training very soon and will never come back again.

In any case, you should read previous similar questions asked on this forum which have been ellaborated infinite number of times already.

[cucumbe...@gmail.com]

You have neither answered nor addressed the question I asked. I never asserted progression is supposed to be linear or unbounded; I was doubtful as to whether my progression strategy was reasonable and optimal and asked for people's views and justifications on what they think constitutes a more reasonable and optimal progression strategy and cases for deviation; a legitimate and pragmatic question in brain training as well as general learning and as far as I've been able to see in this forum has neither been asked nor answered adequately (it's not the same as asking "how do I progress past 3,4,5,etc. back?"). I also never made any claims about IQ so it's asinine to address such an absurd contingency rather than the question. My interest in Dual n-back is to improve WM regardless of whether it yields a direct increase in IQ (I'll even go as far as saying I believe studying math and philosophy are more likely to yield increases IQ than any brain training game)

I would also point your claim "there is no successful or non-successful brain training strategy" is self-contradictory and flies in the face of evidence. Empirical evidence has shown that certain tasks do not provide any reasonable expectation of "far transfer" and are therefore non-successful brain strategies (e.g. playing Chess does not to appear to have significant far transfer effects). Conversely, increases in skills such as mathematical and verbal ability have reasonably strong correlates with IQ (even most of the rules that show up in a typical "culture-fair" test like Raven Matrices test are easily recognizable set theoretic, geometric, and logic procedures), suggesting that using those domains for brain training are likely more promising for far transfer effects than Chess and even Dual n-back (a view somewhat supported by the observation that each year of compulsory schooling correlates to a 1 to 5 point increase in IQ per year) and therefore are examples of a possibly successful brain training strategy.

Additionally it is trivially clear both logically and empirically that certain progression strategies are better than others. For example, picking an n-back that's easy and doing it for a year before increasing the difficulty is clearly a horribly inefficient progression strategy (at least in relation to mine). Furthermore it is possible that perhaps requiring a faster time per stimulus (say 2 seconds/trial) at n back before progression to n+1 back may be more optimal than my own strategy (which allows progression at 3 seconds/trial), though it is not clear to me if that should be the case as one could argue they can improve working memory speed before advancing, but another may counter saying they can develop speed at n+1 back anyways and avoid unnecessarily easy sessions (and it is particularly that type of discussion I was seeking to elicit and understand when making this post).

It would be prudent to ensure you properly read and understand the question before answering.

[nemamea 1]

Thank you for a very nicely written answer. You imply some things which I did not say; you did not ask for explanation, but instead you just expanded my answer to match what you want, and than explained your arguments.

So, my answers are:

1. You did imply linear or quasi-linear progression in your questions:

I used this strategy at D3B and progressed fairly smoothly shaving off 1 second / trial / day. My progress at D4B however was was slower progressing only about 0.5 seconds / trial / day for the first two days, and today I only progressed 0.3 seconds/trial. So I'm currently on D4B at 3.7 seconds/trial in the 80%-95% range and starting to question whether I should try a different progression strategy (or at least some strategy for overcoming plateaus).

So...

    2. I wrote there is no successful or non-successful brain training strategy. I meant only in n-back; there is no empirical evidence that some n-back strategy is successful and other is not successful. One strategy can be better than the other one.
   

3. You are writting: Additionally it is trivially clear both logically and empirically that certain progression strategies are better than others. You are interpreting my statement maximally banally and trivially. This is clearly malicious.

4. In any case, you should read previous similar questions asked on this forum which have been ellaborated infinite number of times already.  You evidently did not read many similar topics on this google group. Similar questions as yours have been asked and answered many many times. People have spent much time to answer them, and there are some great answers and debats.

This would be all from me in this discussion. Have a nice day :-)

[cucumbe...@gmail.com]

Regarding point 1:

Fair enough, I understand how my wording could be reasonably construed that way.

As a mostly irrelevant aside I'd mention my general experience in learning or improving a skill has been more in series of brief but substantial bursts of progress rather than an a relatively steady linear progression.

Regarding point 2:

As far as I know, I agree there is a lack on empirical evidence on that matter, and that is why I asked about it here; I also disagree with your conclusion there is no successful or non-successful brain training strategy even if interpreted within the confines of n-back training in accordance with my statement

For example, picking an n-back that's easy and doing it for a year before increasing the difficulty is clearly a horribly inefficient progression strategy (at least in relation to mine)" which is a clear example of a poor progression strategy within the confines of n-back

The way you worded your statements appeared to me as if you believed the topic was trivial and does not warrant investigation.

On point 3:

No I am not being malicious; given your statements

there is no successful or non-successful brain training strategy. Whichever brain training app/mode/strategy you choose, you will come to some point where you will not be able to recognize any changes on your cognition or score (results). If you have a plan to get 10 or 20 IQ points in few weeks you can forget about it. You will either accept this empirical fact, or you will quit brain training very soon and will never come back again.

Which seems to me a fairly strong and absolute assertion in both wording and your choice of emphasis; my response was in accordance with that interpretation which to me does not seem an unreasonable view (in much the same regard how you pointed out my initial wording could reasonably construe an expectation of linear progression). I understand we all have nuances to our views and mistakes in translation occur. If I misinterpret you, then feel free correct me and clarify; it is counterproductive to presume it sufficient grounds for ill-intent.

I could play the same game with you claiming such things are, "you're making assumption of malice because you cannot debate without feeling personally attacked after all evidence shows most people can't" or "you say 'This would be all from me in this discussion.' as a common cop out to avoid having to defend points of contention because after all who hasn't heard the same line from an ex". Such statements only stand to create acrimony between participants and subtract from productive debate.

Regarding point 4:

I did in fact search for and read through a large portion of the material on this forum and I have not found any particular discussion to sufficiently answer my original questions regarding considerations of reasonable-ness and optimality of progression strategies. Feel free to leave links to any threads that you think do.

[nemamea 1]

You are smart and reasonable. I wish you a pleasant and beneficial brain training journey.