What game(s) would people most like?

[maxim...@gmail.com]

I've recently made a game that improves object individuation ability, which I believe is an important component of non verbal working memory that is an untapped opportunity to improve. Game and scientific support here:http://twitchmath.com/objectIndividuation.html

Thanks to those who provided feedback so far :)

I'm also planning on making some additional games relating to both non verbal working memory and relational ability, and I wanted to check in if there was a game that a lot of people wanted to see made. 

If no one makes any other suggestions I'll first make a game that stresses non verbal working memory in a format similar to n back but probably closest to lumosity's follow that frog. I'll also probably add some features like variable timing. I believe the follow that frog type format where the possible position are irregular and distributed differently every time you play MAY be better than n back because n back may begin to rely a bit on crystallized intelligence as you develop familiarity because the grid is the exact same every time. 

That said I think follow that frog is fairly limited because it takes about a minute for the game to start every time you play because it completely halts and provides the steps numbered for you every time the path increases by one, also you can pause whenever you want, which I don't think provides the best training.

[whoisbambam]

are you Russian or Ukranian or?

i (k)new a Maxim Smirnov once....he was Russian.........so i am assuming 'Maxim' has that ancestry

[maxim...@gmail.com]

neither actually, I'm american :) 

On Thursday, April 9, 2015 at 8:15:11 PM UTC-7, whoisbambam wrote:

are you Russian or Ukranian or?

i new a Maxim Smirnov once....he was Russian.........so i am assuming 'Maxim' has that ancestry

[Marc Howard]

Hi Maxim-

This is AMAZING--I really enjoyed playing and increasing levels.  It would be great to have more information within the tool that either links or clearly explains the practical implications of this game.  Anyway, I think you're really on to something--count me in to test any updates.

Cheers,

Marc

[maxim...@gmail.com]

Hi Marc,

Glad you enjoyed it! Great point I'll come up with a clear concise explanation of the idea and put it on the site.

[maxim...@gmail.com]

Okay I put an explanation up on the website, let me know if it makes sense!

PS. I believe you may feel a significant difference in your cognition once you increase the number you can do from say 3 to 4 or 4 to 5 and then an additional difference as you are able to do one more from there. It depends though from person to person I would guess, if you are already strong in this area it may not matter. Please let me know if you feel a difference!

[Ron Williams]

I'd like to suggest some sort of reduction of the 'jump' between levels. I found that level 4 was quite easy, but level 5 virtually impossible, due to the number of distractors and complex interactions. 

I know you said that the non-correct circles are set at n+1, so that you can't just follow those and effectively be at a lower level, but that assumes people are trying to 'game' the system. 

I'd like to be able to set the distractor circles to be any number I like, so that I get practice in following 5 correct disks - there will still be some indecisive interactions, where circles bump and you aren't sure which is which and that might be enough to practice on.

What about there being two levels per 'level'? One for the correct disks and one for the incorrect disks. I.e. you don't go up a level of correct disks until you can do it with a sub-level of n+1 incorrect disks.

So each 'level' would start  n/n-2 and progress to n/n+1 and then to the next level:

i.e.   4/2 -> 4/5 -> 5/3 -> 5/6 ... etc.

Really, truly, I would only follow the 'correct' disks, because any rational person realises that just getting a higher level is pointless in itself.

--
You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brain-trainin...@googlegroups.com.
To post to this group, send email to brain-t...@googlegroups.com.
Visit this group at http://groups.google.com/group/brain-training.
For more options, visit https://groups.google.com/d/optout.

[maxim...@gmail.com]

Hey I very much agree with you. It's a big mental jump between 4 and 5 and any other two numbers. To address this exact issue I'm working on a way to have one or two in between stages in standard mode.

In the mean time, I made it so that you can reduce the number of incorrect balls to n-2 in custom mode exactly as you discussed.

PS. The fact that its a big mental jump between 4 and 5 illustrates that its the limit of what your mind is able to do that you are pushing up against. I started here too and am now able to do 7 sometimes and 6 almost always, and I found it helps with many other things that rely on non verbal working memory.

[Brandon Woodson]

I don't train WM directly, and haven't in years, but I have to admit I was curious and tried it. Fun game.

Protip: don't fixate; keep your attention quickly moving. I got up to 8 without any fails and virtually no effort, then started thinking too much and couldn't even manage 6 on a couple subsequent trials. Started freeing my attention as it had been in its naive state during the first 8 runs, and score easily shot back up again.


- Brandon

[maxim...@gmail.com]

Hey thanks! I'm curious was it you who completed level 7 of the game successfully something like 8-10 times today?

[Marc Howard]

@Brandon--loved your pro tips--hopefully I'm not too selfish in asking if you had anymore or if you could explain how to keep your mind consistently "naive".

@maximumlabs: thanks much for the extra background text you added. What I was hoping is being able to site real-world examples or implications of the game. For instance will it help me learn to juggle better or drive through obstacles on a busy street or take out more targets in (for example) a game like Call of Duty?

I've been cautious with brain training platforms like Luminosity and DNB because of the lack of proof of game training directly benefiting a real-world shortcoming. Like most games the better you get at it the better you become in the game--but nothing else.

Hopefully this makes since as I'm not coming from any academic or psychology background but a regular 'ol guy wondering what tangible benefits I'm likely to get should I invest a few weeks of my time with this game.

You received this message because you are subscribed to a topic in the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/brain-training/6P6BWXBySXk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to brain-trainin...@googlegroups.com.

[maxim...@gmail.com]

Hey Marc,

The research I've read converges to a central theme, that object individuation capacity can be a bottleneck in the non verbal working memory (NVWM) system.

If object individuation capacity is a limiting factor for you as I believe it was for me and the cited researchers seem to believe it is for many people, improving this area should improve your non verbal working memory. In short, you'll get better at many things that depend on NVWM, which is an enormous number of things. 

Also, I think (depending on the individual) titis helpful in things like driving through obstacles on a busy street or call of duty than something like chess.  Those first two things depend on taking in a lot of information and rapidly putting into working memory. In chess, you take in a smaller amount of information but its more about maintaining that information internally for an extended period so that you can strategize deeply. Maintaining information is more related to a different ERP, the CDA, whereas object individuation capacity is related to the N2pc. 

Some background on CDA: http://www.ncbi.nlm.nih.gov/pubmed/20147415

It's tough to answer whether it will help you do those specific things you mentioned. It might, but only if it is the limiting factor for you in those areas.

How is that for an answer your question?

By the way, this next game I'm working on is pretty much entirely focused on the CDA, and it makes me noticeably better at chess, but thats a sample size of 1.

[maxim...@gmail.com]

Hey sorry, this is how this lone should read:

Also, I think (depending on the individual) it is more helpful in things like driving through obstacles ...

[Marc Howard]

Ho no worries—yes this was immensely helpful thanks for taking the time to explain a bit more.  Just a quick follow-up, and to really state what I am trying to get out of all this, is a way to measure progress in the real-world.  In the game its great to know that, for instance I can finally make it up to level 8 (actually I have not made it past 5 yet!)—but how can I correlate a higher game score with for instance better [insert real-world problem].  I suppose one can say overtime, “oh seems like i’m getting better at driving” or “oh seems like i’m getting more kills in Call of Duty” but one can see how that could be biased or placebo-effect like.

I would love to (and I’m not sure how yet) be able to setup a test that based on scoring higher in the TwitchMath game directly attribute a real-world benefit from in in a way that could be objectively measured.  Perhaps I’m asking too much though?

Marc

[Brandon Woodson]

@maximumlabs: Yes, 7 is easy. 8 is much harder untrained, but I think I could break it with a little practice. I did a battery as a benchmark for some training or another a few years ago, and consistently scored well on visual tasks where all of the (relevant) stimuli is presented all at once, much better than I do on sequential visual WM tasks anyway.

@Marc: Thanks. I just meant naive to short-term practice effect. There is some gain in WMC (or perhaps, protection from substantial WMC loss) for me - from my short experience earlier, and even from instances where I'm called upon to hold some information in short-term memory and perform much better at that than would be expected given my usual performance on, for instance, a digit span task - when I am not concentrating and don't allow my attention to settle too much. If my concentration becomes too fixed and inflexible, my available WMC is first down the drain. I guess you could liken it to hypnotism causing an entrancement if the range of stimuli experienced within some short time internal is insufficiently varied.

I don't have any immediate plans to train WM, but I do want to experiment one day with phonemic awareness training, maybe something along the lines of identifying individual phonemes or randomly selected sets of phonemes at rapid pace from high-speed speech (e.g., from speeded audio books). Maybe that will have some effect on auditory WM capacity in some individuals; I suspect I may be one such individual. But in truth, I don't have many good tips, because I don't have too much hands-on experience with WM training, just a bunch of untested ideas and hunches.


--Brandon

[maxim...@gmail.com]

Hey Marc,

Well here's where I noticed the difference most:

1) talking to extremely smart friends that I knew well, I found I could keep up with them and sometimes jump ahead when before I would be behind at times

2) playing piano improved in a very specific way that I can try to describe more in depth if you are a pianist (I'm a very amateurish one but I play frequently enough that I felt the difference was very noticeable)

3) increased starcraft apm by more than 20 %.

4) improved mental rotation ability (as measured by lumosity rotation matrix) significantly

5) improved overall nvwm (as measured by lumosity follow that frog) 

I think everyone is different though and you may not be limited by your object individuation ability in these areas or at all. That said, if you'd like to test it on a specific thing that would be awesome and I can try to help you find something suitable.

Do you have any things that you do fairly frequently, where you could try playing the game really hard for a few days and then see if you notice a difference? As you can see from the list above, the things where its most useful tend to require you to take in information and put it in your working memory very rapidly, if you want to suggest some possible things to test it out on I'd be happy to give my thoughts on if they'd be good testing grounds.

Finally, as the researchers suggest, I do think it will have a positive impact even if only slightly, on most things that require NVWM, which is everything from driving to call of duty to chess to math and many other things. 

[jotaro]

"By the way, this next game I'm working on is pretty much entirely focused on the CDA, and it makes me noticeably better at chess, but thats a sample size of 1."

PLEASE I AM CURIOUS ELABORATE

will it be on mathtwitch available like the first one? with a link?

cute? i like the idea of holding information for extended period of time

so i would apply more reasoning to it so i will come up with a better way.

my max level in your game standard is 7.

and brandon is a black sheep, the dude outperforms a lot of people i wouldnt call him average at all.

so before you take his tip take into account that u wont reach his score anyway through your score might increase with quick attention freeing

and switching .

as for marc if you play call of duty your only way to measure progress

is to know if you can track more dynamic variables in game at the same time, of course you would account for more information and should more decision that covers more possibilities and allow less blind spot, not sure about faster shooting and faster recognition of threats through. for example the field of vision could have more players shooting each other and u able to navigate in that better or driving with alot of obstacles to track at the same time and u will run to less obstacles cause u will not loss track of them .

could be usefull if there is a transfer in a point of view of positions.

[jotaro]

i would also will rant that 500 speed and 100 speed feels the same.
?

[maxim...@gmail.com]

Increasing the speed multiplier scales the threshold at which the velocity is "damped", the balls are allowed to go faster. They accelerate at the exact same rate. It could feel the exact same if you have seconds tracking set to three or four seconds, as it is in fact the exact same for the first second or two, and then it begins to diverge from what you would see when the speed multiplier is set too 100.

In short, the speed multiplier increases the top speed, but not the acceleration. I'm potentially open to modifying this however, that's just how its set up now.

[jotaro]

i tested it in 30 seconds duration with 10 false balls.

500% speed and 100% speed feels the same to me.

[jotaro]

or more like after u set it to 500 and then again to 100 it doesnt go back in practice to 100.

only when page refreshed

so its a bug?

[maxim...@gmail.com]

Hey thanks very much for bringing this up! It was a bug and I fixed it, its working well now. 

[jotaro]

if you wanna thank me elaborate on the CDA

this sounds fascinating and same with the new game.

[maxim...@gmail.com]

For the next game which will be online in a day or two I'm planning to collect research that gives a more detailed explanation of CDA just like I did for the first game, but here's a short intro: 

CDA stands for contralateral delay activity. When researchers take eeg recording while subjects perform non verbal working memory tasks, they consistently observe "a sustained contralateral negativity during the retention period", that is, the time that subjects are actively maintaining information in working memory.

Also, "This contralateral delay activity (CDA) has previously been shown to increase in amplitude as the number of memory items increases, up to the individual's working memory capacity limit."

In short, it appears to be an absolutely key piece of the non verbal working memory and overall intelligence puzzle. 

source: http://www.ncbi.nlm.nih.gov/pubmed/20147415

[maxim...@gmail.com]

I wanted to let everyone know that I added sublevels to the object individuation game. This is something that people asked for.

Sublevels can make it easier to gradually progress and build ability once you reach your limit, since there is a big gap between 4 and 5 for example mentally. However, you can switch them on and off with the button just below the game. I wanted to make it switchable so that you don't have to progress through a lot of levels that are easy for you to reach your limit. When you do a level perfectly the score counter only counts it if you are on the top sublevel, which is the exact same as the standard.

Please let me know if you can find any bugs, I've tested it quite a bit don't think there are any left but let me know!

Also, please let me know if you have any additional feature ideas.

[αrgvmziΩ]

Are you aware that your program is very similar to Target Tracker ( http://www.brainhq.com/why-brainhq/about-the-exercises/attention/target-tracker , also formerly known as "Jewel Diver")?

Incidentally, I can't get yours to play in a browser on a Nexus 7 tablet completely; part of the screen gets cut off.

argumzio

[jotaro]

sub levels is a very good idea

the increase in much difficulty comes mainly from

the high jump in false balls with each additional ball.

lol at this game argumzio posted looks exactly like yours in the picture!

only blue screen and ball color different!

but i dont care, its free and its all good.:)

--

[maxim...@gmail.com]

Hey yeah I am aware of that. I wanted to make one that isn't so expensive and has more features such as those I've added. Thanks for letting me know about the Nexus 7. Its hard to test on too many devices, it works well on my android, I'll try to get it to work on more devices over time. For what it's worth, my understanding is lumosity and posit science games ONLY work on computer, (though lumosity has a very few games available on a separate app). At least its a step up from there :) it should work on a lot of phones.

[Palguay]

For what it's worth, my understanding is lumosity and posit science games ONLY work on computer, (though lumosity has a very few games available on a separate app). At least its a step up from there :)

For what it's worth, brainturk has this game implemented as tracking objects , our free app has 20 games . I am planning on adding this game as well as space fortress to the mobile app in the next two weeks

[maxim...@gmail.com]

Hi Argumzio, how does it look now? I've reduced the width and height of the game slightly, also, are you loading the page in landscape mode? 

[whoisbambam]

i believe there is no cognitive benefit from games for healthy ppl without existing deficits that is transferable to cognitive tasks in non-game situations

am i misinformed?

[Ron Williams]

maximumlabs - Thanks for the adjustment - I found that I could follow up to 6 disks, with 4 distractors, so it's definitely the number of interactions with distractors that poses the greatest challenge for me, at least. The most difficult scenario for me is when all 6 immediately start to spread out to the four edges of the arena, and bump into distractors on the way. Even in that sort of scenario I was able to follow five out of six of them sometimes. 

Would it be possible to make it so that if you do manage to follow, say n-1 of the disks, you don't get dumped to a lower level and have to waste time going up again?

.....

I'm of course impressed by Brandon's abilities in this area - I wonder if practice at the enhanced Image Streaming techniques has improved the capacity for this sort of attention locking?

Also of course impressed by those who can keep track of six in standard mode.

impressed == envious...

[maxim...@gmail.com]

Hey I think it's a good point, dumping to a lower level on n-2 is probably a bit too low of a threshold for having to go down a level, and I definitely understand it can be a pain to go back up again. I made it that if you get n-2 of the balls or more you get to stay on the current level. I've tested it, but let me know if you can find any bugs!

[jotaro]

here is also idea if the player, get n-2 but he made the mistake ono the first ball, let the player atleast complete all the balls in question (n)

before judging his mistake instead a mistake on the first try.

i mean if i knew n-2, but 2 of them are a mistake i dont know what ones but if i press a mistaken at first i get fucked.

so let him finish all his n guesses first.

[maxim...@gmail.com]

That's true and a good point, I think you guys are both absolutely right that it was sending you down to a lower level too often before. I modified it so you keep selecting circles until you've either you've selected the same number of circles as the level you are on or you selected two incorrect balls. This way you aren't punished for happening to get the first one wrong. Since this makes it much less likely to go down to the lower level, I've made it so you must get n-1 correct to maintain your current level as it was before. Let me know what you think and if you can find any bugs!

[Dorso Lateral]

Hey Maxim,

When you posted this thread I immediately thought of the following, but only today got a chance to put together this pitch:

I’d like to suggest games based on the relational integration tasks described in papers by Klaus Oberauer and Adam Chuderski.  I’ve listed some of the relevant papers below with annotations and excerpts (tasks that could become games described in 3&4).

 

1. When are GF and WM Isopmorphic and when are they not? (Chuderski, Adam 2013) http://ecfi-group.eu/download/papers/49.pdf

This paper found that relational integration tasks (which the authors classify as a WM task) along with a STM task, were isomorphic with fluid reasoning measures when those tasks were administered under severe time constraints. (A good article in Scientific American about the paper: http://blogs.scientificamerican.com/beautiful-minds/2014/01/22/working-memory-and-fluid-reasoning-same-or-different/)

It seems reasonable to assume that improved performance on these relational integration tasks ( in the dimensions of accuracy, complexity and speed) could transfer to more complicated and novel fluid reasoning tasks (verbal and figural analogies, etc.) if relational integration were a bottleneck for some people.  The WM-Gf correlation declined when time was less of a constraint, so it would seem that training relational integration (and improving both speed and the number of objects that can be integrated) could be especially useful if one wanted to improve the ability to reason very quickly on novel complex problems.

2. The relational integration task explains fluid reasoning above and beyond other working memory tasks. (Chuderski, 2013)  http://link.springer.com/article/10.3758/s13421-013-0366-x/fulltext.html

This paper explains the relational integration tasks in basic terms and presents new variations of them that could provide options for introducing adaptability into a game version.

Excerpt from the paper: 

“In contrast to the aforementioned views about the proper measurement of WMC [the view most of us have been reading about for some time], Oberauer et al. (2007) proposed that the driving force of strong WMC–Gf correlations is neither the sheer storage of information, even in the context of processing, nor the executive control of that storage. In line with analogous theorizing … Oberauer et al. (2007; Oberauer, Süß, Wilhelm, and Wittmann 2008) proposed that the fundamental mechanism that determines both WMC and fluid reasoning is the human capacity to set and maintain the flexible, temporary bindings between chunks held in WM, or between them and their respective positions within some mental structure. For instance, these positions can constitute concrete coordinates like serial positions during recall, or they can be abstract placeholders in some schema required in a reasoning task (so-called role-filler bindings). Due to temporary bindings, a person is able to integrate elementary relations into novel arbitrary relational structures. Creating such structures is the essence of relational thinking—thinking driven by the way objects are assigned to certain roles in situations, and not by objects’ intrinsic features.”

 And, getting to the tasks: 

“For the purpose of measurement of the effectiveness of relational integration, Oberauer and colleagues (Oberauer, Süß, Schulze, Wilhelm, and Wittmann 2000; Oberauer et al. 2008) have developed versions of a so-called relation-monitoring task (henceforth called the relation integration task). In such a task, a participant observes a constantly changing pattern of stimuli that is available perceptually (no need for storage in WM), and detects stimuli matching a simple rule. For example, the task may consist of the presentation of a three-by-three matrix of words, and may require the pressing of a button if and only if three words in a row, column, or diagonal line rhyme. Other versions require three numbers that end with the same digit to be found, or recognizing four dots that form a square within a pattern of several dots. A few studies showed that the latent variables loaded by the relation integration tasks are at least as strong predictors of fluid reasoning as are complex spans (Buehner, Krumm, and Pick 2005; Buehner, Krumm, Ziegler, and Pluecken 2006; Krumm et al. 2009; Oberauer et al. 2008; Süß, Oberauer, Wittmann, Wilhelm, and Schulze 2002), and much better predictors than both STM and executive control tasks (Chuderski, Taraday, Nęcka, and Smoleń 2012).”

 And: 

“…though performance in WM may rely on multiple mechanisms and processes (see Conway, Getz, Macnamara, and Engel de Abreu 2011), the link between relatively simple WM tasks and much more complex abstract-reasoning tests may be primarily driven by the relational integration component of WM.”

 

And, concluding: 

“… the study provided new evidence that clearly supports theories … that have proposed that human intelligence may reflect the domain-general ability to construct higher-level relational structures that bind a certain number of more atomic representations (e.g., perceptual or memorial), are extremely flexible, and can be effectively abstracted from any intrinsic features of the low-level representations. This line of research—explaining intelligence as the ability to conduct role-based relational reasoning based on the processing of relational roles explicitly and separately from (perceptual or semantic) features of entities that fill these roles, and that involves coding the bindings of entities to their specific roles—is in a way a revival of Spearman’s (1927) classical idea of the eduction of relations. This relational-reasoning account has recently gained substantial attention within psychology, and seems to be a very fruitful framework for future studies on fluid reasoning.”

It seems that these relational integration tasks developed by Oberauer (see below) and Chuderski could be testing a cognitive capacity (or skill) lying below the level of reasoning - and below the math and logic relational games Maxim recently created - and that these capacities or skills could be bottlenecks for some people in regards to relational/analogical reasoning and fluid intelligence. 

3. Which working memory functions predict intelligence? (Oberauer, et al. 2008) http://diyhpl.us/~bryan/papers2/neuro/working-memory/Which%20working%20memory%20functions%20predict%20intelligence%3F.pdf

This paper describes four relational integration tasks in detail that were “all constructed to tap the ability of mentally building and integrating multiple relations between given elements  These tasks not only require the detection of pair-wise relations between given elements, but the integration of several relations into the representation of a new configuration.”

1. A verbal task where the relation is rhyming.

2.    2. A numerical task

3.    3. A spatial task: “…flight control, involved monitoring the trajectories of five to nine triangles (representing airplanes) moving in different directions across the screen. Whenever one airplane was about to crash into another plane or a mountain (represented by brown patches) participants were to stop the video and redirect one airplane. Each stop came at a small cost, but each lost airplane incurred a large cost, so that a good score could be obtained by intervening if and only if necessary to prevent crashes.'

4.   4. A second spatial task

As a group, these relational integration tasks correlated with the reasoning measures of the Berlin Intelligence Structure at .94 (and with the Creativity measures at .47).  These correlations were greater than any of the more common storage and processing WM measures. (And far greater with regards to creativity).

Actually, there were eight relational integration tasks, as all four were administered in versions that required holding the items to be integrated in memory and also in versions where the items were continually visible and therefore did not require memory to be engaged at all.  Surprisingly, the correlations to the reasoning measures were nearly identical in both the memory and no-memory versions! (The no-memory version was the one used in the first study (WM-Gf isomorphic) in this list.)

At this point, you might be wondering whether relational integration tasks are WM tasks at all!  This paper’s reviewers wondered the same thing and there is a very interesting part in the discussion where these questions were addressed. For starters, the authors note that the Relational Integration (RI) measures correlated strongly with the more traditional storage and processing (SP) WM measures, so there is that. Additionally:

“These findings have far-reaching implications for our view of working memory and intelligence. The traditional interpretation of the relationship between working memory and fluid intelligence or reasoning is that working memory provides resources for simultaneous storage and processing, that is, the ability to remember information not currently present in the environment, and to manipulate this or other information at the same time.

 

Both abilities are arguably required in many complex tasks — for example, remembering intermediate results while carrying out further operations in multi-step mental arithmetic tasks (Hitch, 1978). On this account, however, it is hard to understand why relational integration [RI] tasks without any demand on storage should predict reasoning so well. The main difference between specifying the RI factor through memory task versions and specifying it through no-memory task versions was that in the former case, the factor correlated more with SP, confirming that the variation of memory in the RI tasks was effective. This variation had little effect, however, on the RI factor's correlation with reasoning or the other intelligence factors. We conclude that a demand on short-term storage is not a necessary feature of a good measure of WMC [working memory capacity]. Other research (Colom, Rebollo, Abad, & Shih, 2006; Oberauer et al., 2000) has already shown that a processing component is no necessary feature either. Thus, “simultaneous storage and processing” is a good description for one effective and very popular class of tasks used to measure WMC, but it should not be used to define WMC as a construct.

 

This is not to say that our RI tasks capture all there is to the construct WMC, and that SP tasks are redundant. Rather, we argue that the construct WMC should be conceptualized in a broader way than before, and operationalized by a broader set of tasks. The present RI tasks were intentionally designed to be different from conventional SP tasks, with the goal to establish a separate factor of RI besides SP, and to test the hypothesis that despite their dissimilarity with SP tasks, RI tasks predict reasoning ability. The finding that both our RI tasks and the SP tasks, despite their superficial dissimilarity, share a large amount of variance and account for large amounts of variance in reasoning, raises the need for a conceptualiza- tion of WMC that covers both kinds of tasks.”

 

The authors go on to defend against some specific arguments that RI should be considered a component of reasoning itself, rather than WM.  I’m not entirely convinced by their arguments, but whether RI should be classified as a component of WM or of reasoning (or both:) may not be important to the question of whether training such skills, which seem to underpin much more complex reasoning (figural analogy tests, etc.), may be worthwhile.  We get into questions of near vs. far transfer, of course, but personally I’m not opposed to the notion of training basic underlying skills and capacities that might make me better at reasoning, even if that improvement represents only near transfer.

However, I’ll add the following because it helps clarify exactly what these relational integration tasks are meant to be testing (and what, converted into adaptive games, they might be used to train): 

“It can be argued that the relational processing involved in our RI tasks is very similar to the relational processing necessary in analogical reasoning tasks such as those used by Sternberg (1985) to measure the “inference” component (i.e., discovering the relationship between the first 2 terms of an analogy)… In response … we fully agree that discovering the relation between two terms in an analogy is the same kind of relational processing as is measured in our RI tasks. Yet we insist that WMC is not a limit on processing individual relations but on integrating relations, and this is what is measured by the RI tasks. Analogy tasks also require integrating relations, but this is necessary only in the next step, where the relation between the first two terms is applied to the third term to complete the analogy — Sternbergs components “mapping” and “application”. This is why, in our view, analogy tasks correlate with WMC.”

 

4. What is working memory, and how can we measure it? (Wilhelm, Hildebrandt, Oberauer 2013) http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3721021/pdf/fpsyg-04-00433.pdf

A number of novel working memory binding tests are described in this paper that could be made into adaptive tasks to train the building and maintenance of bindings (but stopping short of integrating bindings as in the relational integration tasks) in WM.  Chuderski suggests that these tests could be tapping a lower-level capacity upon which relational integration relies. 

From the abstract: “The findings support the hypothesis that individual differences in WMC reflect the ability to build, maintain and update arbitrary bindings.”

From the paper: 

“In our own view, working memory is a system for building, maintaining and rapidly updating arbitrary bindings. For instance, items in a list are bound to list positions, objects are bound to locations in space, and concepts are bound to roles in propositional schemata. The capability for rapid formation of temporary bindings enables the system to construct and maintain new structures, such as random lists, spatial arrays, or mental models. Working memory is important for reasoning because reasoning requires the construction and manipulation of representations of novel structures. The limited capacity of working memory arises from interference between bindings, which effectively limits the complexity of new structural representations, and thereby constrains reasoning ability .”

A surprising hypothesis from the conclusion of the paper: 

“According to the binding hypothesis, high WMC reflects the ability to establish robust bindings in working memory, which in turn support encoding of those bindings into SM. Therefore, high [working memory capacity] might be a cause, not a consequence, of a well-functioning [short-term memory]. 

Maxim, I'm not aware of any neuron-level views on these relational integration/binding tasks, but I think they would be worth a shot.  It would especially be interesting to see if training on such basic relational tasks would help a player rapidly improve on the more complex math and logic relations games you recently created. Since there are verbal, numerical and spatial versions, it should also be possible to evaluate whether improvement in spatial relational integration could improve verbal or math relational ability, or the reverse.  

What does everyone think?  Have you seen anything out there like this?  (There is the German NeuroNation, maybe they relational integration games?) 

[maxim...@gmail.com]

Hey Dorso Lateral, 

Thanks very much for sharing this, this is extremely interesting! 

It appears to hold real promise in pushing the field forward by providing a superior tool. I think this might be quite related to what I've been trying to figure out: what is the mental process that is occurring when a pattern strikes you as obvious? For example, sometimes when you do a Raven's test, you just immediately see the pattern, what is going on? It sounds like this could potentially be very related.

I would be entirely open to making a game, or several games, based on these findings. 

First, however, I need to understand what you've discussed fully. (It has exceeded my capacity for immediate complete relational integration haha) 

I've read your entire post and bits and pieces of the papers, but I want to gather additional background for a day or two and then hopefully I'll be able to respond with something intelligent. (and hopefully start building a game of some kind) 

Also, to help me get background, you mention neuronation games as potentially having to do with relational integration, I was wondering if you could elaborate on this a bit? To they have games that have any similarities to the games discussed in the papers or are stressing relational integration?

Thanks again, I'm going to spend a bunch more time understanding this!

[Dorso Lateral]

Hey Maxim,

My pleasure:)  First, regarding NeuroNation - I have no idea if they have similar games, or if similar games are out there in some other place.  I recently got the NeuroNation app but haven't used it yet.  The reason I thought to ask the list if NN might have something similar is that NN is a German outfit and since the originator of this relational integration paradigm is German, well, I thought that NN would be onto this if anyone was.  

So, I guess the question for the community here is: has anyone seen such games on any of the brain training websites out there.  (The games are described under papers 3&4 in my post, above, dated 4/26).

Yeah, Maxim, it is a lot to get one's head around:)  Take your time.  In the next couple days I mean to put together a little description of the stages of analogical reasoning and identify the various tasks that correspond to those stages.  I first need to get a copy of Sternberg's book (cited in one of the excerpts from paper 3) so that I can use his stages of analogical reasoning as a model.  This would, hopefully, help us all get our heads around what each task is testing/training and therefore how such tasks could possibly benefit those with a bottleneck at a particular stage of reasoning, hopefully to widen that bottleneck and improve their reasoning ability.  

In general, as far as I can tell, this relational integration paradigm could offer a completely different way of improving working memory and fluid reasoning, one that does not even rely on short-term memory, as all four of the relational integration tasks described in the papers exist in two forms: 1) a form that require holding the relations in short-term memory and 2) a form that does NOT require memorizing the relations at all.  Both versions correlated equally well to Raven and other figural analogy tests. 

I'd like to note here that I am not a professional in any field related to this.  I just happened to run into these papers and spent a bunch of time reading them.  So this is totally amateur hour on my end:)  I would be very interested to learn what other members of the list think of all this.

[jotaro]

i gotta love academics complicate stuff with their language, guess its a from to introduce barriers so to protect their authority.

in truth this is all could be written in much more simple terms, alot of vague stuff going on.

lolol.

[Jabba Dabba]

From my cursory reading I think I'll begin work on the following games , which are relational integration tasks. Also as per the above paper , they should be done under a time constraint :

> Word matrix task : In an n x n matrix of words try to guess as fast as possible if the words along a diagonal or row rhyme with each other.

> Numeric matrix task : In an nxn matrix identify as quickly as possible if the leading or ending digits of the numbers on any diagonal or row are the same.

The third one is a spatial task quite similar to the game maxim has already developed.

[maxim...@gmail.com]

Hey Dorso Lateral,

To answer your question, I have not seen such games on any major brain training web sites out there, there, and I've checked a huge number that offer games via both mobile and internet, if you can think of it, there's something like a 90 % chance that I've checked it, though I could always be missing something, because there are quite a few at this point. I also gathered some info at NeuroNation, but didn't subscribe. It looks to me like neuronation has no evidence of having these games based on youtube videos and what they have in the beginning stages of their website.

It's a great point about improving working memory while maintaining all of the stimulus in clear view. I think it may be the case that what has been called "working memory" traditionally is really many subprocesses, some of which, possibly the most interesting of which, do not depend on maintaining information that has been removed from the field of view.

[maxim...@gmail.com]

Also, do you consider it preferable that relational integration games be as simple as the games discussed? Maybe this would do a better job of pinpointing the skill? On the other hand, like the raiseyouriq training, it may be very constrained to someone who is focused on enhancement rather overcoming a deficit.

Chuderski's definition of relational integration appears to simply be: "the process that binds mental representations into more complex relational structures." Though if there is a more detailed definition I would love to see it!

...

[maxim...@gmail.com]

some neuroanatomical research on relational integration: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.28.3703&rep=rep1&type=pdf

...

[maxim...@gmail.com]

Okay so to get back with you as promised:

I think the theoretical premise is very very compelling. I'm not completely confident that the games they used in the paper labeled 3) are the right place to start. They only spend a sentence or two discussing why they think they are a good choice of game, and unless I'm confused there are many games that rely on relational integration.

However, if no one has any other ideas, maybe it is as good a start as any. 

Jabba, maybe it would make the most sense if I make one of the ones that you haven't started?

...

[Dorso Lateral]

Hey Maxim,

I'm going to have to back out of my earlier promise to peg tasks to stages of analogical reasoning.  I don't have a good enough handle on it to make a positive contribution at this point.

But I do have a couple more thoughts about relational integration (RI) in relation to cognitive training:

1. I agree that the theoretical premise is compelling, if for no other reason than that Chuderski found that the correlation of the four RI tasks with Gf overlapped most all of the Gf-WM correlation as well as predicting Gf test performance above and beyond what WM tasks are able to do.  It seems that if one could improve RI skill or capacity that something very similar to WM would be trained along with well as something else (?) very much related to Gf.  I also have the feeling that such training would be more pleasant and less effortful than something like n-back.

In addition, if Oberauer happens to be correct in writing "“According to the binding hypothesis, high WMC reflects the ability to establish robust bindings in working memory, which in turn support encoding of those bindings into SM. Therefore, high [working memory capacity] might be a cause, not a consequence, of a well-functioning [short-term memory]," RI tasks may turn out to be the "missing link" between effort and results in cognitive training.  

2. I think it's intriguing how similar your object individuation game (which, by the way, is similar to task used by Oberauer in a 2000 paper) is to the 3rd RI task, "flight control," with the latter being a more complex version of the former. What if object individuation and relational integration represent two parts of a chain of abilities that enable, at the top of the chain, higher-level behavior such as analogical reasoning?  I've mentioned before that I like your focus on cognitive bottlenecks.  However, what if we took the bottleneck one-step further, within the framework of working memory capacity limits?

I recently a paper (and can't find it now!) wherein the authors describe how subjects trained in various working memory tasks exhibited near-transfer to other measures of working memory but not far transfer to reasoning tasks.  Searching for a reason for failure to far-transfer, the authors argue that, since the near-transfer effect was large, it seemed unlikely that the working memory improvement was not significant enough to transfer to reasoning.  They therefore hypothesized that it may simply take time for a subject to use their improved WM capacity on higher-level tasks.  After all, we learned to reason, in all the ways that we currently know how to reason, within our current WM capacity limitations.  If we improve our WM capacity at this point, it seems plausible that we might need to re-learn how to reason using our new capacity, to learn how to leverage the processing advantages a greater WM capacity offers.  

For example, what if a person where to significantly improve their object individuation ability?  They may, at that point, find that their scores on STM measures (spatial span) also improve, suggesting that they have, in the lingo of the RI theorists, increased the number of temporary relational bindings they can establish.  But what if they also find that their scores in RI tasks have not improved?  An explanation could then be that they had widened a bottleneck at the lower, object individuation/STM-level, but now find that RI is a new bottleneck, insofar as it relates to their performance on the lower-level tasks.  What then?  Well, train RI tasks until their performance on those tasks reaches the same binding capacity (chunk capacity, etc.) as they demonstrated on the object individuation/STM tasks - in other words, widen the next bottleneck in the hierarchy of abilities/skills that make up reasoning ability, and go all the way up the pipe until reasoning skills have improved in proportion to the new capacity (WM capacity, basically) demonstrated in lower level tasks.   Chaining near-transfers to effect far transfer.

If such a scheme were to work, we'd expect improvements in object individuation to lead to more rapid improvement in RI, and improved RI to then lead to more rapid improvement in whatever task lay ahead of it in the chain or hierarchy (assuming there is such a chain linked in this way), all the way up to reasoning.  We would further expect that improvement may take a sudden leap when a new binding/chunk capacity were reached in the lowest task in the chain, which would lead to leaps in tasks higher up the chain, leap-frogging.  

As you've noted, WM capacity seems to be domain-specific, and if that is the case we probably couldn't expect such chain-effected reasoning improvements to generalize into other domains (spatial to verbal, say), but maybe that's just what we have to live with.  The early returns from the n-back studies seemed to promise wide generalization, and settling now for a series of near-transfers that add up to an effective far transfer confined to one domain may seem a bit of a bummer.  Of course, all of this could just be pure bunk in any event.  

Also, here is a classic paper that some may have read that puts forward a model of working memory that is similar to the relational integration/binding hypothesis, in that Halford argues for WM capacity to be viewed as the number of elements (variables) that the mind can simultaneously form relations between: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.357.7406&rep=rep1&type=pdf

BTW, I joined NN and found no trace of relational integration-type games.  Although, as you note, relational integration is required in many games.  I'm not smart enough to predict if training it in "isolation" would really be advantageous.  I hope you make a game or two, though, and we can try and figure out experientially if it is.  

[maxim...@gmail.com]

Hey Dorso Lateral,

This is another very interesting paper! It's interesting that this paper concludes that the rule of four that we've seen in things like object individuation applies to here as well.

I actually think the hypothesis that it may take a while to learn to use your increased working memory may be correct based on my n=1 experience. I also agree that training that is less effortful and more pleasant is important. Finally, I agree that sorting out how exactly these different components of the "intelligence system" relate to each other, especially things like bottlenecks and progressive stages, seems to be key at this stage in the development of neuroscience.

I'm almost done with a version of the relational integration game that is a slightly modified version of the numbers game! I'll release it soon and I'm very much looking forward to everyone's feedback. I'm very much looking forward to making it more complicated and adaptive. I chose the numbers one because it seems to me like the best platform for adaptivity and increasing complexity among the ones that were listed.

Lastly, I just wanted to reiterate my impression that this approach of focusing specifically on the ability to integrate relationships and understand many variables is probably addressing intelligence much more directly than working memory training.

[maxim...@gmail.com]

Also, is there anywhere where I can try some of the problems discussed in that paper?

[Dorso Lateral]

The mechanics of this forum suck.  Why can't we quote and respond like in other forums?

Anyway,  very glad to hear you're tackling the numbers version of RI, Maxim!  Looking forward to trying it out!  

BTW, what's the rule of four in regards to object individuation?  Is that discussed in one of the papers you linked to below the task on yours site?  (Still haven't had time to read those papers.)  Yeah, that Halford paper is very cool - it illustrates very clearly exactly what happens when we try to reason above our working memory/relational integration capacity - relational bindings between variables start to slip out of the mind's workspace and we have to start all over again.  Very frustrating experience.  Reminds me of the frustration I felt as a child learning to tie my shoe - I knew how to do it ,but the damn lace just slipped out of my fingers as I tried to slip the loop 'round it:)  Is that the paper you're asking about whether there is a place to try out its problems - the bar graph ones?  

If so, I don't know, although Halford cites a few papers near the front as sources for the development of those tasks.  Practicing those tasks as an exercise would be a bitch, but they would be very useful as measures of one's working memory capacity (or RI capacity) and legitimate improvement of same.  They also illustrate clearly how someone with a very high working memory capacity - 5 or 6 chunks - would seem head and shoulders above someone around the average (3 or 4 chunks) when both were confronted with novel material to process.  

[jotaro]

and still even if your short term memory improved it still has high limits, maybe there are tasks that are too complicated for short term memory alone.

[maxim...@gmail.com]

Yep I meant the Halford one, its something that keeps coming up in many studies, its more of a "common finding that has led to a rule of thumb" than a scientific result, here's some discussion:"It has long been noted that individuation is limited in capacity: we can quickly and effortlessly perceive that there are exactly two items but not that there are exactly eight items (Jevons, 1871; compare Figure ​Figure11 left panel upper row with lower row). Enumeration is equally quick, accurate and effortless within a narrow range of one to four objects."

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4145468/

However, many papers that discuss non verbal wm and/or object individuation capacity mention it, though I could be over generalizing.

[maxim...@gmail.com]

Okay Version 1 of the relational integration game involving numbers is at http://twitchmath.com/relationalintegration.html

It's a very bare bones initial version, and I'm planning to use it as a launch pad towards more complex games. That said, I do find myself playing it a bit addictively even though its so simple.

It is is slightly different from the game described in the paper. You have a 3x3 grid of squares that contain two digit numbers. Your job is to identify when three vertically or horizontally aligned squares have numbers with the same final digit. When you identify this pattern, click on any of the three squares. 

The game runs for about 40 seconds, and your objective is to correctly identify as many of the patterns formed as possible. It's actually suprisingly difficult to not miss patterns because you are actually dealing with a fair amount of visual information and the speed is fairly fast. I'm also ironing out the counting mechanism to make sure it has the correct total count, I think there may be some kind of bug, but either way at this point its definitely playable and a platform to start experimenting more with relational integration.

Finally, its quite different from any other game I'm familiar with, it takes some getting used to. I haven't been able to get a score above 10 or so yet. It's possible that the denominator is too high due to a big though.

[maxim...@gmail.com]

Hey I just wanted to let everyone know that I'm experimenting with a a logic relational ability game that is non verbal and has a much higher density of relationships. 

It also has compressed time and requires you to answer the question twice, the second time a few of the relationships change. It has higher arity, in other words it requires you to understand relationships between a larger number of items. 

It's still very experimental, I'm eventually hoping to get it much better engineered over time. The idea behind changing a few of the relationships and prompting the second time is to prevent the process of calculating step by step and then forgetting the earlier steps as was the dominant strategy in http://twitchmath.com/logicrelationalability.html

[jotaro]

isnt it better to do introduction , when you actually gonna put the link to the game?

[maxim...@gmail.com]

hahaha wow ... emabarassing ... here it is haha http://twitchmath.com/logicExp.html

[Ron Williams]

Great - more or less what I asked for (I didn't go past level 3 so far as I'm busy).

I noticed that there seems to be a dichotomy between the 'axioms' and the final question, as it appears to be 'programming-based' - i.e. == vs =.

Could we have a switch in settings to either be consistently '=' or have '=='? Actually the use of '=' is sort of 'wrong' in a programming sense, in that in the initial rules it appears to mean '==', anyhow. It surely can't mean 'assignment'? 

What would  'CF=-Zq' _mean_ anyhow? 

Aren't we saying that CF is equal to or interchangeable with -Zq? 

You probably didn't mean that a variable CF is being assigned the negative value of what the variable Zq contains?

If so, then it should be CF==-Zq.

But if so then why bother with ==? It seems to be only out of programmer habit... 

I'd rather have '=' everywhere, as in math, or for a programming style, '==' everywhere if it's always a comparison and not an assignment operation.

But thx anyhow - I like the look of it - it might make a better training for math and programming than the original

Ron

[Psionic]

Hey, nice game for programmers, although I didn't unserstand what stand ":" for since the 3rd level.

Maxim, did you tried something yet with the Game I mentioned in Object Individuation Capacity Game thread - the Polyrhythms? Thanks for letting me know, I just submitted the concept for realization to neuronation and waiting for an answer from them

Dne pátek 5. června 2015 5:52:35 UTC+2 RonW napsal(a):

[cev]

Got 16 on relation integration - cool game!

Add a penalty for incorrect identification?

Interesting that still no nine achieved on the object individuation task; I've had eights but it is immediately apparent to me that nine circles are too many to handle.

[maxim...@gmail.com]

Hey Ron, interesting, I hadn't considered it this way. I was considering it an assignment. I'll have to think a bit more about that. I think it could be an assignment and still be consistent with programming, but its not in the order that would be most intuitive with programming

[maxim...@gmail.com]

Hey Psionic,

: is just a way to separate visually the end of the if condition with the beginning of the next step. I got the idea from python syntax. So for example

pl = -ha

ha = -yH

if pl == ha : lB = -ha else lB = pl

if ha == yH : vJ = -pl else vJ = pl

if yH == lB == ha : Vk = -yH else Vk = lB

if !vJ == yH == lB : tp = -yH else tp = yH

the last line means if it not true that vJ, YH, and IB all have the same value, then tp is equal to -yH, otherwise it is equal to positive yH

I didn't make the polyrhythms game yet I'm sorry! I'm just doing this for free and not planning to make a business out of it so I unfortunately can't do everything, I wish I could! I still want to hear all suggestions though, I'm really interested in augmenting human cognition in any useful way.

...

[αrgvmziΩ]

After blazing passed level 30 in the Math Relational Ability game, I am beginning to think it is either too predictible or there is indeed a higher level pattern to it, such that difficulty follows a U-curve (seemingly centered around level 8) instead of an upward slope. In either case, it may be a good idea to mix things up... if only to prevent me from being able to do this. ;)

argumzio

[maxim...@gmail.com]

check out the experimental relational ability game, I bet that you won't be too easy for you :)

http://twitchmath.com/logicExp.html

[argu...@protonmail.com]

Yes, that is actually trickier, but for a surprisingly different reason. I decided to make a few minor changes to the source so that the relations would be more salient. Since I've done that, it's been easier to read off the relations faster. My scores hover comfortably above 10 now fairly consistently. I still struggle with taking it a little slower to correct careless mistakes, though.

Example:

⟪⧲⪩ ≡/≡ ▩Å⟫

⟪▩Å ≡/≡ Å℉⟫

⟪▩Å ≡≡≡ ⧲⪩⟫ ≡>> ⟪⫷⨂ ≡/≡ ▩Å⟫ 
ELSE ⟪⫷⨂ ≡≡≡ ⧲⪩⟫

⟪⫷⨂ ≡/≡ ⩙⪩⟫

⟪⫷⨂ ≡≡≡ Å℉⟫ ≡>> ⟪↺⧱ ≡/≡ Å℉⟫ 
ELSE ⟪↺⧱ ≡≡≡ ⫷⨂⟫

⟪↺⧱ ≡/≡ ⧑⟡⟫

⟪⩙⪩ ≡≡≡ ↺⧱⟫ ≡>> ⟪⬱⧳ ≡/≡ ⩙⪩⟫ 
ELSE ⟪⬱⧳ ≡≡≡ ⩙⪩⟫

⟪⬱⧳ ≡/≡ ⨊⩩⟫

⟪⬱⧳ ≡≡≡ ⧑⟡⟫ ≡>> ⟪ℨ▩ ≡/≡ ⧑⟡⟫ 
ELSE ⟪ℨ▩ ≡≡≡ ⧑⟡⟫

⟪ℨ▩ ≡/≡ ↺⟃⟫


⟪⧲⪩ ≡≡≡ ↺⟃⟫ ? 

Very cool game. It's actually more pleasant when done with abstract symbols, I think. :)

argumzio

[☉]

Before I get too carried away, I just remembered that not every browser will be able to render all of these spectacular Unicode symbols that I used:

⧑⧒⧍⧎▩⬱⬤↺⟰⟱⪨⪩⩩⩙⨁⨂⨃⨄⨅⨆⨇⨈⨌⨊⨋⫷⫸℥ΩℨÅ℉℃⟡⧰⧱⧲⧳

argumzio

[☉]

After adding charset="UTF-8" to the meta H.T.M.L. tag, the inconsistent rendering issue I noticed with the Unicode characters was fixed. In case anyone is wondering, I've attached what level 19 looks like. (I did not use the experimental code, because that would seem to "morph" when entering yes or no.)

argumzio

P.S. I hope you don't mind I took the liberty of modifying your code. Maybe you can use it as an "alien" version for more exploratory users?

[maxim...@gmail.com]

Hey Argumzio, Awesome idea I agree the alphabetic characters can be distracting! 

Wait did you try out the new logic game with url ending in logicExp.html? I

[☉]

Yes, actually, I gave that a go and got up to level 10 quickly enough.

The A == B == C is a nice wrench you threw in there, but nothing a little Unicode grease can't fix. :)

argumzio

[maxim...@gmail.com]

That's ridiculously impressive, is there a way you found to make shortcuts or something? You have just over 1/3 of the time of the previous game, you have to answer the question 2x, and the "arrity" is significantly greater, so I think it would be reasonable to say that getting to level 10 on the experimental game would be something like 6x + harder than getting to ten on the original logic game.

[☉]

Well, I don't feel too impressed. After all, I had to change out the A-z characters to the more visually distinct Unicode symbols for the basic version - but I did that to take advantage of what I thought I was seeing in the MRA game. There seems to be some high-level patterns on a level of generality that may more often than not facilitate visual inspection. The method is by no means perfect and requires ample use of pattern detection (i.e., intuition); it doesn't seem to be clearly communicable, at any rate. I don't entirely think I've gotten lucky since I'm able to do it consistently, but I still wonder. :/

However difficult the tasks are, one can still guess 50/50. So to counteract that, it might be a good idea to add more complexity at the higher levels. Perhaps you could use other logical concepts like modality and temporality? A third condition (in addition to yes and no) may be "not relevant" or "unknown" (down arrow), in the case that the puzzle asks about a statement that is not resolvable based on all of the given premises.

Aha! How about this:

Instead of checking for only one statement, we can progressively be required to evaluate more than just one at a time! So instead of just "yes" or "no", we could have the following setup:

A = B

B = C

(A = C) => (B = D)

C =/= D

1) A = C

2) B = D

3) B =/= D

For the ones that are true, we enter 1 = left, 2 = down, and 3 = right. In this case, we would press left and right and then hit enter to confirm our selection. (Repressing the keys prior to pressing ENTER can deselect them.) NOTE: I slipped in an explosive statement, so they're actually all true. ;)

I'm not sure how one might want to scale this with levels, but I think it would make sense, just to eliminate (lucky?) guesswork. This may also be scaled up with a number key (on the keyboard) for checking up to ten separate propositions.

I still would like to have an "Alien Logic" version with Unicode symbols, of course. :D

argumzio,

throwing pebbles at a boundless ocean

[maxim...@gmail.com]

Interesting, I'll have to consider this, I was actually having this exact thought about 50/50 when I designed the Experimental Logic Relational Ability Game at http://twitchmath.com/logicExp.html game, I think you actually have only a 25 % chance of moving up a level if you just guess because it asks you two different questions each time. (BTW, if you are wondering about whether the statement with XX == YY == ZZ has a non 50 % chance of being true, I designed it so that it always has exactly a 50 % chance of being true)

[☉]

Yeah, I noticed that "morph" of the question. I think it would be better to ditch that kind of serialized processing, because that's like being asked the same question twice, and keep the basic game's format (I like how propositions are intermingled with conditional statements) but then just ask multiple questions simultaneously. Lucky guessing should easily be minimized by a factor of 2^q for each q question added. I'm thinking scaling could be linear until more information clarifies that. And since we'll be asking more than one question as things get more difficult, we don't necessarily need to increment the length of the query. After all, how often is it in, say, programming that you have a series of statements like that? Sometimes you have multiple statements that might mean something in many places at a time.

So it could go like this (L# is level, S# is number of statements, and Q# is number of questions):

L1 = S1 & Q1

L2 = S2 & Q1

L3 = S3 & Q1

L4 = S3 & Q2

L5 = S4 & Q2

L6 = S4 & Q3

L7 = S5 & Q3

L8 = S5 & Q4

L9 = S6 & Q3

L10 = S6 & Q4

L11 = S7 & Q4

L12 = S7 & Q5

L13 = S8 & Q4

L14 = S8 & Q5

L15 = S9 & Q5

This might need some adjusting, but you get the idea. This assumes we'd ideally want no one to break L15, since I believe that S9 & Q5 would be extremely difficult to guess. As for mixing things up, you can have "is opposite to" be replaced with "== ~", or per my Unicode "≡≡≡ !!!" (i.e., something that it is visually distinct and still obeys the usual meaning of bang).

argumzio

[maxim...@gmail.com]

It's a very interesting idea, I may code it up sometime soon! I have a lot I'm working on though so I doubt it will be within the next day or so. If you have to serially answer each different one it would reduce the chance of guessing your way to a high level quite a bit. Did you see the notation I used in the experimental game ? it doesn't use words at all. Lol I still can't believe you got to level 10 in it, extremely impressive, I wouldn't have thought it was humanly possible for anyone in the world to do that without a week of practice at least.

[jotaro]

what? he already took pic of him getting to level 19, and its only a second or a third day.

give him a week he will reach level 30. lol

--

[maxim...@gmail.com]

I'm just making sure he's tried the one at twitchmath.com/logicExp.html :)

[jotaro]

yea u are right the level 19 pic is from a different game.

[Hi]

I recommend others change the variables to actual imagined scenarios, it makes much more for a work out, at least for me, more than increasing relational ability there are definitely aspects of creativity (or to be more specific divergent thinking) which are improved in conjunction with relational ability, which of course is in part an aspect of and or a catalyst to forms of creativity, more precisely or more definitely, those that specifically relate to the cognitive domain, especially in the realm of forms of cognitive flexibility, a correlative of divergent thinking, and likely as well to a different but in some aspects related end, relational reasoning.

I try to make my examples much more elaborate but:

NW (i.e. bank robber shoots gun in bank) = vm (i.e. increase in heart rate)

vm (increase in heart rate) = VB (i.e. flight or fight response)
 
if vm (increase in heart rate) == NW (bank robber shoots gun in bank) : PF (child falling asleep while mother reading him/her a bed time story)= -NW (no bank robbers, everyone using and working in the bank is pleasant and calm)  else PF = NW

if vm (increase in heart rate) == NW (bank robber shoots gun in bank) : pB (taking a quiet afternoon stroll in a park) = -NW (no bank robbers, everyone using and working in the bank is pleasant and calm)  else pB = vm

is NW == pB ?

So far I've been using and even combining:

- Metaphorical

- Spatial/other Physics/Engineering

- Psychological/Social 

- Physiological (partly physics)

- Time

- Other

Also with the:

"else pB = vm"

Where you're already aware of the flow of the answer, I still make myself come up with an alternate "hypothesis", making myself having to momentarily switch sides in order to encourage more cognitive conflict/stress.

What are your own insights and adaptations? I'd love to hear them :) .

[maxim...@gmail.com]

Oh man this is super creative! I wish I had thought of that :)

...

[☉]

I recently found an exploit with MRA and LRA. ELRA doesn't have the issue (due to the compound nature of the test). Feel free to reach me at argumzio @ protonmail.com if you want to know more. I've already gone beyond level 100 on MRA and LRA using it.

I don't think the exploit will exist once LRA has multiple queries per question.

argumzio

[maxim...@gmail.com]

Does it involve the probability being slightly not evenly distributed? 

[☉]

Yes, and for the record I responded off list. :)

argumzio

[Hi]

Thank you, I appreciate that as well as the fact that you've created a platform for training such faculties which is steadily increasing in sophistication, if you'd like me to share any more thoughts, perhaps more potential rule manipulations as opposed to or including simply ways to interact with the rules themselves, let me know :) .

Again, anyone please share any insights that you'd like to make as well :) .

Kindly, 

Hi

...

[maxim...@gmail.com]

Hey Everyone, I changed the character set in http://twitchmath.com/logicrelationalability.html to the one Argumzio recommended, thanks for the promising idea!

Re: most recent comment, sounds interesting and thank you for your thoughts, any chance you could elaborate a bit? I'm intrigued.

...

[maxim...@gmail.com]

Hey I just wanted to let everyone know I removed some unnecessary parenthesis and most importantly added color coding the http://twitchmath.com/logicExp.html game.

...

[☉]

Unfortunately, not all browsers will render the characters unless you add charset="UTF-8" to the meta tag.

argumzio

[Ron Williams]

This is what I'm seeing in Chromium:

⌠is opposite from âˆ

∠is equivalent to ©±


is ⌠equivalent to ©± ?

is that what you intended? :)

Certainly more symbolic, in any case.

RonW

[Hi]

Hi maxim,

Check out this thread created by argumzio, there's a well to tap for nearly endless creative oil for cognitive potential for sure!

https://groups.google.com/forum/#!topic/brain-training/jdCC14Ob3CM

[maxim...@gmail.com]

Hey thanks for the pointer, are you suggesting a game based on the wason task?

It's an interesting task, I tried it and got it right, but I think its just because I've been doing a lot of programming. By the way, if anyone wants to build logical skills while also learning programming I suggest checking out codefights.com, it has micro competitive programming challenges that only last a few minutes each time.

[☉]

Very cool. Never heard of codefights.com before.

argumzio

[Hi]

Yes and no, maxim :) . 

I will elaborate later in the day. 

Thank you.

[Hi]

Sorry maxim, things have been x. I promise to be able to reserve a few moments to explain myself properly by the end of the day.

Its 12:57 am at the moment. 

Latitude: 13 degrees 3' North

Longitude: 144 degrees 4' East

Otherwise I hope in the meantime you check out argumzio's suggestion as well as Brandon's categories, both of which I'll likely share some thoughts on in the evening as well, both likely in combination and separately for greater measure. Regardless as to what I decide in the end I imagine that the discussion won't be as dangerous of a thought adventure as venturing through the mysterious Bermuda Triangle (Bermuda triangle has multiple "pun's" (I'm counting at least three, oh there's another one haha!) here if anyone would like to have fun working them out!).

Kindly,

Hi

...

[jotaro]

lol first brandon have talked in riddles, now hi has even more riddles.

when i thought people here cant get more unexpected, HI comes along.

dude i will seriously give you a golden medal just for being unexpected.

--

You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brain-trainin...@googlegroups.com.
To post to this group, send email to brain-t...@googlegroups.com.

[Hi]

Sorry for the delay, apologies.

Two quotes taken from this message:

1) "I'd like to build a basic foundation of communication here that defines the purpose of increasing the level of differentiation [in the logical relations task] so we can understand the reason for submitting without refutation to its ultimate & highly consequential why, making any alternative direction incredibly secondary."

2) "When we implement a differentiated enough set of categories into the logical relations task [or its equivalent] it opens the beginnings of a portal to mastering the movements of the human mind, in so saying this, upon remedying a final solution and its training a humans psychological power will have a platform to potentially become boundless and therefore profound. Ultimately any potential we have for contributing to humanity will be provided with the opportunity to become sufficiently optimised, mentally at least. Our mental universe is the universe of our mental experience, in so saying this, optimising anything outside of this should obviously be of secondary importance to this Ultimate."

This is the beginnings to that (1 & 2) attempt:

First of all, I need you maxim, argumzio and Brandon to make these ideas and implementations better if you're able to. Our consciousnesses combining ideas enables the greatest innovation to take flight. 

Secondly, I'm going to take an incrementalists approach with suggestions (which are many - most of which serve the primary cause such as enhancing qualities of cognition otherwise the secondary elements are mainly indirect considerations of this purpose) I'd like to see reflected in the exercise with firstly focusing on the categories being utilised in the game of logical-relations to begin with (argumzio seems to be intrigued with how dual-n-back and the wason task can be conglomerated anyhow, something which I'll personally reserve discussion for another time).

Thirdly, I'm going to be using level 15 of logical relations as an example of the limited dimensionality associated with the tast itself as it concerns creating a mental map of a logical set that underpin something such as a mental/other experience and its connected constituents, which could otherwise be better achieved with the inclusion of more if not better relational categories (opposite and equivalent simply don't cut it as it concerns noticing and defining all fundamental areas of and a perception, only multiples of the same property) which enable expansion on microscopic to macroscopic lenses, as well as other levels of phenomenology, such as different expressions of MULTIPLICATION (expansion) and DIVISION (contraction), all of which are understood (principle of containment) through the lens of ENCOMPASSMENT. These three (see capitals) of course taking residence in Brandon's illustation and elaboration of ontological categories and the Jewish script.

Moreover, I would argue that as intelligence increases the greater the level of sensitivity there is to differentiation and in so saying this, choice as to what relations to attend to due to being aware of more relations to begin with. 

Also, I doubt that an organisms level of differentiation, as it concerns abstract intelligence would not increase more than or anything at all for that matter in the case of the basic form of how the logical relations task stands at present (only the two categories - opposite and equivalent) if we're comparing it to the individuals working memory, which is obviously correctively enhanced certainly through an attentional mechanism, both related and not related to the obvious selective improvement in working memory. That is, unless we're referring to the organisms sensitivity of already acquired levels of differentiation, which can increase with an increase in working memory capacity but not level of differentiation. Working memory would just be an upgrade from a hand to a baseball glove but the catching and base ball following (through the air) ability would remain relatively the same.

One simple example of a persons average level of differentiation, as separated from working memory, is if we have:

Person A sees person B sitting on the floor against the wall in the corner of a room instead of sitting on one of the comfortable sofa's.

Person A decides to say to person B, "are you sitting against the wall in the corner because you just want to be in your own box, your own world away from people?"

Person B responds, "Not really, I just like it here where I am. No reason in particular."

Person A responds, "These are four comfortable chairs that you could be choosing between, I find that really weird. You're keeping away from people." 

Person B calmly responds, "Oh yeah, no special reason. I find your observations kind of weird as well but somewhat explainable."

Person A begins to get angry, "Your weird, you're trying to stay away from people." Person A leaves the room. Of course, a small irony on the leaving in light of the statement, "you're trying to stay away from people" as well as the fact that person B was quite accessible for communication relative to their normal whereabouts, out of the house or in their room.

Person B's ultimate reason for choosing to sit in the corner against the wall as opposed to sitting in one of the supposedly comfortable chairs:

- Sitting on the floor in the corner of the room allows him to receive the greatest amount of warmth that he would find comfortable from a wall heater close by in order to counteract the cool weather which is growing disproportionately larger than the heat in the space occupied by the chairs.

Now although there are several more layers of reasons ergo differentiation, such as the greater structural support the wall provides (while using the laptop such as I am now) as opposed to the very soft couches (water bed style), this simple example illustrates very clearly that although person A could correctly identify through perception where person B was in the room, as well as, remember this detail during the conversation as well as the conversation itself (working memory), he was unable to create further differentiation both as to what he would communicate and how he would communicate as well as building sufficient answers as to why something might be so. He was only able to use the most available response to him, which was created on the first level of differentiation, "this is not normal", however he was unable to go beyond the first level of differentiation, therefore his brain was only capable of creating an equivalent explanation, again, on the first level of differentiation: "that which is not normal is confusing and if it is a human then this human is weird in order to help my brain explain the disorder and therefore explain away my own inability to properly organise a suitably approximated explanation or even something that still allows for an open ended enquiry into the matter," but even this is still probably generous. 

All in all he could only move (at a very basic level) in either one of two directions, opposite or equivalence, one of the very reasons why we should increase the number of categories the task has so if he were to ever play it (unlikely) he would have the opportunity to develop greater levels of experiential discrimination-differentiation. We should be looking to create activities that aim to multiply one's ability to differentiate more than working memory, setting up working memory enhancement as a secondary condition which is amplified as an indirect/direct consequence of increasing differentiation, at least those activities whose working memory load can be compared well with the logical relations task which naturally increases in WM load (not to mention other recognitive abilities which warrant more elaboration on another time, both how they relate to intelligence, as well as for this purpose, how they relate to the task itself)

as one increases in level (however the main purpose here is to illustrate that when one reaches a certain point as the difficulty level increases in the logical relations task at present, working memory load increases without differentiation load (limited to the combinations of equivalent and opposite), with the secondary not primary focus, increasing disproportionately), as previously implied and now explicitly stated.

A more simple and explicit example of lower-higher levels of differentiation includes the following:

Lower = The jacket that I am wearing is blue and the shorts black.

--> Correct identifications in an un-to-low-differentiated state.

Higher = The jacket is likely going to keep an organism more warm during the winter compared to a pair of shorts, however not all organisms seek to be warm in the winter nor do all organisms have similar annual biological seasons/rhythms.

--> Correct identifications in an un-to-higher-differentiated state.

And we can go higher still, moving from microscopic to macroscopic levels of abstract (not necessarily sensorial) differentiation meaning higher in this context is of course not synonymous with geographical elevation, in fact, simply noting the geographical location of an organism is likely much more peripheral compared to more pertinent items of prediction, such as why the organism is there and therefore what is it doing, unless of course we're merely looking at practical considerations, but this comes after figuring out whether the answers to this abstraction need following up with this practical consideration such as physically moving to the related destination to retrieve more information or even the organism itself. 

Not making the necessary realisation of abstraction however, even if physical transportation was necessary, that they could use other means of transportation and retrieval without needing themselves to physically travel is an admission of abstract failure as well, such is the case with many social interactions where one may feel they need to say something to someone to defend themselves against a threat in order to protect their own ego. I'm personally not completely insusceptible to the latter however I've so far intuited that the application and practicing of higher order thinking is to the negation of the more reptilian set of responses, which is what encompasses the related mental artifact/s and the practicing of its negation obviously one of the main purposes of this discussion.

One of if not the ultimate goal is to make systems (accurate detection and reasonable response to a set of interacting inferred consequence from lower to higher levels of abstraction) thinking one's default mode, something which doesn't come naturally to human beings, something which Brandon's contribution (regarding "A Correspondence of Ontological Categories and the Jewish Script") most definitely makes a step towards and what we can do here with greater actualisation of not only the interacting expressions of the ontological categories themselves but including the subforms associated with the different categories (such as multiplication and division as a depiction of expansion/contraction, increase/decrease; both of which reside in different contexts or the same context if we're referring to the increase or decrease in expansion/contraction, and its connection with limits), as well as the real nitty gritty of emphasising the discrimination of interaction and simultaneity more than simple bi-directional thinking (i.e. equivalent or opposite) which in ultimate conglomeration speak of systems within systems, above, below and between all in the unified All (ontological category (macro-micro-micro-macro)). I think we will be able to implement Brandon's categories among other related variations here however for me at the moment its an incremental build up as I'd like to build a basic foundation of communication here that defines the purpose of increasing the level of differentiation so we can understand the reason for submitting without refutation to its ultimate & highly consequential why, making any alternative direction incredibly secondary, for now at least.

When all humans have evolved to a mature enough level of systems thinking operating as the default, violence against humans overall culture and its progression becomes impossible which meets ultimate optimisation or complete which meets ultimate mass suicide. If its the former, even if undesired human features become obvious to simulation, no purposeful destruction of related humans will occur only action towards the destruction of the undesired features (i.e. which naturally becomes more possible with the optimisation of resources to optimise genetic engineering and any form relating to human potential). In so saying this, one obvious "if then" is, if cultural evolution is equivalent to technological evolution then technological enhancement is equivalent to global enhancement. The realisation of the former will have to come through those that are already evolved enough to create persuasive enough artifacts, mental and otherwise, that encourage this travel however people cannot make the realisation by themselves, it is bi- directional+, they need to also be manipulated into realising it. But hey, we can't let this conversation get too off track now can we? Ha ha.

Back on topic... The aim is for the clearest perception of reality and as far as we humans are so far aware, understanding phenomena from inwardly (i.e. mental experiences, perceptions, etc) to outwardly (i.e. how inward experiences, perceptions, etc are organised to reflect knowledge of the outside) in terms of a system which is the conglomeration of interacting subsystems is the best way to fulfill this duty, such as Brandon's mention in relation to the ontological categories he's described (as already previously mentioned above) and there, in part, account for the observer (subjective) aspect of all cognition, systems thinking does this as well but more thoroughly if the correct measure of thoroughness (i.e. number and type as well as sub-type of categories) is applied.

Systems thinking might indeed be synonymous to systems 2 thinking, of course however there are deeper levels as it concerns *system/s as an abstract quality to contemplate, specifically and or especially as it pertains to previous mention on simultaneity and interaction. Perhaps others here who are more knowledgeable on System 2 thinking might like to enlighten this relationship better than I can at present.

Within the confounds of a system, which everything is and is a part of, simultaneity and interaction seem to be not only the highest possible forms of reasoning about a related system but the only real way to accurately reason about a system, where x as an individual and/or set is put into context with y as an individual and/or set. One thing cannot be understood without something else also existing, even with thought itself, there must always still be an observer.

We are increasing both inference of logical consequence and inference/detection of simultaneity difference and therefore, a greater recognition of cause and effect, feedback loops, and even, how to remedy them by virtue of increasing logical consequence of applied simulations, for example.

As intelligence increases its not only the detection of categories via differentiation but the interaction of their logical consequence held simultaneously with a clear representation of simultaneity, which is both an act of working memory (holding multiple interactions (of logical consequence) and or simulations at once) and detection (similarity<---->difference). 

The above sentence is where the updating feature can come into play which is what shall be described in one of the examples. This feature is analogous to the operation of the "Enhanced Visual N-back" however instead of updating *positions* we're updating *rules* through change instead of space, for example, where one newer rule works to the negation of the older rule therefore updation (sounds better) of the "changing dynamic of the environment (C.D.E)" needs to be reflected in one's response before one can move onto higher levels of differentiation, something of which is highly cited in the intelligence scientific literature as being a basic definition of intelligence itself (responding optimally to the C.D.E. Dynamic which is akin to *interaction and *simultaneity). With intelligent perception one perceives the necessary level of differentiation, with working memory one can effectively organise the perceptions for optimal response, which might for example involve comparing it with past experience however effective comparison is to the limits of differentiation and final execution of the code, working memory. 

To simplify, the detection of change is to differentiation and the (effectiveness of) response to what is detected, working memory capacity. These are the primary limits of optimised adaptation (internal/external). Of course, one might indeed have the capability to differentiate a change but be limited as it concerns the organisation of an equivalent response, the role of working memory. Otherwise our computer systems (at least mine for now) are very good at the inverse of this. The Terminator movie series would of course be an example of these two melodies (literally speaking as it concerns the related movement of the electrical waves and impulses) being richly identified and actualised as it concerns the demonstration of sophisticated artificial intelligence.

Firstly, let's go back to the original agenda in relation to providing an example of the limitations associated with only having two (limiting) categories (equivalent and opposite).

We'll take the example of level 15 for example and use a basic example involving the invention of a television.

⧠is opposite from ³©

Inventing the television is opposite to having no electricity.

³© is opposite from Œ‘

Having no electricity is opposite to having house lighting

If ⧠is equivalent to ³© then ©¨ is opposite to ⧠otherwise ©¨ is equivalent to ³©

If inventing the television is equivalent to having no electricity then lighting candles is opposite to inventing the television otherwise lighting candles is equivalent to having no electricity

©¨ is opposite from „â

Lighting candles is opposite from sunlight

If ©¨ is equivalent to Œ‘ then Ÿº is opposite to ©¨ otherwise Ÿº is equivalent to ©¨

If lighting candles is equivalent to having house lighting then getting burned is opposite to lighting candles otherwise getting burned is equivalent to lighting candles

Ÿº is equivalent to ˆ‰

Getting burned is equivalent to First aid kit

f Ÿº is equivalent to „â then ¤§ is opposite to „â otherwise ¤§ is equivalent to „â

If getting burned is equivalent to sunlight then universal expansion is opposite to sunlight otherwise universal expansion is equivalent to sunlight

¤§ is opposite from ƒâ

Universal expansion is opposite to rigid mentality

if ¤§ is equivalent to ˆ‰ then ‹™ is opposite to ¤§ otherwise ‹™ is equivalent to ¤§

If universal expansion is equivalent to first aid kit then technological innovation is opposite to universal expansion otherwise technological innovation is equivalent with universal expansion

‹™ is opposite from ˆ§

Technological innovation is opposite from cultural de-evolution

If ƒâ is equivalent to ‹™ then ââ is opposite to ‹™ otherwise ââ is equivalent to ƒâ

If rigid mentality is equivalent to technological innovation then not growing as a person is opposite to technological innovation otherwise not growing as a person is equivalent to rigid mentality

ââ is equivalent to â«

Not groing as a person is equivalent to not living a full life

if ââ is equivalent to ˆ§ then °© is opposite to ˆ§ otherwise °© is equivalent to ˆ§

If not growing as a person is equivalent to cultural de-evolution then constantly looking to improve things is opposite to cultural de-evolution otherwise constantly looking to improve things is equivalent to cultural de-evolution

°© is opposite from â¤

Constantly looking to improve things is opposite from the North Korean government

⤠is opposite from Ÿ°

The North Korean government is opposite from the standards of entrepreneurs

is ⧠equivalent to Ÿ° ? 

Is inventing the television equivalent to the standards of entrepreneurs?

Yes

Is the North Korean government equivalent to lighting candles, having no electricity, not growing as a person, burned skin, cultural de-evolution, rigid mentality, international relations first aid kit?

Yes

---------------------------------------------------------------------

Of course some of these real examples ended up simultaneously being opposites and equivalents however this still doesn't disagree if one just follows the logical organisation of the rules being generated, although there can be disagreement with something by relational flow of content separate to its order relative to the logical organisational flow of the automatically generated rules. Nevertheless I suppose one could argue that its a form of cognitive interference which is worth training anyhow. 

This can be ameliorated with the implementation of greater differentiation required in in the early stages.

Another limitation is when something is meant to be an opposite or equivalent but by story line contradicts or vice versa, again, something which could likely be avoided if there were several more levels of differentiation required early on seeing the mind is forced to create several more layers of distinction-differentiation (see previous reasons for the encouragement of this). Naturally, the more similar something seems the less differentiation is being made, vice versa. And, the more distinction that is made the more room there is for differentiation to expand most prosperously, vice versa, however each with notable similarities and differences, with enough of the latter to deserve small recognition here.

-------------------------------------------

Moving to higher levels of differentiation

-------------------------------------------

So what I'd be ultimately aiming for is:

* With each level we increase the number and difficulty (i.e. increase is meaningfully more difficult than multiplication and conversion to multiplication) of different categories simultaneously.

We start with one "if then" and one category. Firstly, none of these levels so far take categorical difficulty into consideration, only their quantity.

Naturally questions should be asked in terms of category, meaning any category that's explicitly or implicitly in the description can or should be in the question that follows, as is pretty much the case now anyhow.

I'd like to have more examples up by tomorrow to show increasing levels of differentiation but I have no doubt, at least hopefully, that Brandon, argumzio, yourself and even others will be able to help out with elaboration on this strategy to prioritize differentiation over working memory (i.e. number of different categories preferably of greater abstraction over the number of the same categories), with the latter implicated increasingly anyhow as the levels progress higher, as previously mentioned.

You may even wish to create a terminology set for people.

The purpose shouldn't be to use terms that people know more than it should be using terms that accurately describe reality in as much detail and optimisation as possible, the latter of which is a control for getting stuck in detail that is not serving a purpose more than a hierarchically more general feature of reality, especially when its substituted as the first feature used to describe something. For example, calling a chair instead by its timber type to describe what you're sitting on to someone over the phone (or electrical device - again an error in communication if one feature is the only feature you can use to describe something) instead of saying a chair. The statement is of course still correct however its a failure in communication as its obviously not the most optimal choice if its the only choice one can make to describe how they're sitting to someone, unless of course there is some guessing game going on, in which case it is likely a correct choice as it slightly (all things being equal) increases the predictibility of a correct answer without sacrificing it, depending on the context. See previous examples of lower-higher levels of differentiation for further explanation, at least, other than the following proposed direction.

Here is the first very basic and preliminary example of increasing levels according to the number and type of different categories rather than the number of the same type of categories.

Organisation presently, as implied, in an un-to-low-differentiated state:

X is opposite to Y.

Inventing the television is opposite to having no electricity.

Is X opposite to Y?

Is inventing the television is opposite to having no electricity?

Yes and yes.

X is opposite to Y. X equivalent to A.

Inventing the television is opposite to having no electricity. Inventing the television is equivalent to inventing something that is electrical.

Is A equivalent to Y?

X is opposite to Y. X is equivalent to A. X is contained by B.

Inventing the television is opposite to having no electricity. Inventing the television is equivalent to inventing something that is electrical. Inventing the television is contained by past human inventions.

Is A contained by B?

Is inventing something is electrical contained by past human inventions? 

Yes.

X expands to the contraction of A.

A certain period of time has elapsed and humans have figured out a way to invent televisions that don't consume electricity.

Is A equivalent to Y?

Is inventing televisions equivalent to inventing something that is electrical? 

No.

Is B expanding, contracting or neither and is it expanding, contracting or neither A?

Are past human inventions expanding and are past human inventions expanding the invention of televisions?

It's expanding B and it is contracting A.

C secretes D which contains X which is to the inverse of the creation of C.

Intelligence secretes innovation which contains inventing televisions which is to the inverse of the creation of intelligence.

Overtime will B expand or contract?

Contract.

^^ So as we can see here, as the rules progress and change overtime to sometimes over-write the old rules (i.e. X is equivalent to Y) due to for example the principle of contradiction, its sort of like your enhanced version of spatial n-back. 

So the terms we used in this example were:

- equivalent/opposite

- contained

- expanding/contracting

- increasing

Naturally we can see that by the time we arrive at "North Korea" we will find not only the final conclusion to be much more differentiated but also much more nuanced to the expression of understanding rather than the expression of prejudice, something which only reasoning in *equivalence and *opposite leaves an organism more susceptible to creating. Sufficient ontological realisation is to the realisation of the illusion of separation and the very real reality that everything is connected, one thing or country cannot happen without affecting the creation of another country and another thing and/or country the creation of it.

Apart from Brandon's contribution regarding ontological categories, categories that likely should definitely be implemented each of which have their own dynamic are:

* Time - Before, during, after

* Volume - more/less

* Change - increasing/decreasing

* Relationships (equivalent to the nature of the already used *equivalence and *opposite) - inversion, mirror, conversion, feedback loop, reaction

In a much more differentiated and programmable form, some of the kinds of questions that I'd like to be asking in the logical relations task when it finds a more differentiated state:

Does system on level one which harnesses system level a b c create a compatible complex for system two to occur?

Does system x compete/complement with system y?

If yes does it compete/complement in this way? 

Analogously, what is the electrochemical potential of this set and is it comparable to this subsystem of system b?

At least partly understanding implementing the nature of the term system is equivalent to chemical terminology like:

- Entropy

- law of multiple proportions

- law of radioactive decay

- molecular orbital theory

- reaction quotient expression

- redox reaction

- resonance

- signal splitting

- symproportionation

-- spontaneous

-- exo/endo-thermic

-- critical point

-- cell, nucleus, element

-- coordination 

-- forumula

-- DNA of reasoning

-- collision

-- enthalpy

Terminology which they themselves have come through an organism inferring interaction in/and simultaneity, vice versa in higher/lower forms, cause-effect correlational inference. Which is systems thinking, how one thing relates to another even though it might not be directly physically connected to it. Which otherwise cannot be understood without the context of simultaneity and interaction being accounted for. Naturally the higher levels one reaches in interaction simultaneity, which could otherwise be referred to as "sensitivity to context" (in short and therefore without adding a because, which is already as aforementioned (i.e. sensitivity to increasingly higher levels of differentiation)), in the game of logical relations, the easier it is to not only infer new and old (already recognised by humanity) systems but on the topic of old systems learning things like chemistry and physics would become a complete breeze because your awareness and sensitivity to disconvering all elements that encompass problems is there. Chemistry and physics might even actually become common sense.

Overall to sum, when we implement a differentiated enough set of categories into the logical relations task it opens the beginnings of a portal to mastering the movements of the human mind, in so saying this, upon remedying a final solution and its training a humans psychological power will have a platform to potentially become boundless and therefore profound. Ultimately any potential we have for contributing to humanity will be provided with the opportunity to become sufficiently optimised, mentally at least. Our mental universe is the universe of our mental experience, in so saying this, optimising anything outside of this should obviously be of secondary importance to this Ultimate.

More examples will be provided, especially as it concerns rules that emphasise *simultaneity and *interaction, something yourself, Brandon, argumzio and even others might like to add to.

Kindly, 

Hi

Otherwise:

https://www.youtube.com/watch?v=12u1nA7bXzc

On Tuesday, June 23, 2015 at 7:36:39 AM UTC+10, maxim...@gmail.com wrote:

[Hi]

I look forward to being able to elaborate more on the interaction between four words by the end of the week:

*Simultaneity 

*Interaction

*Reaction

*Differentiation

And any other perceptions that have arose in awareness enough for distinction to be assigned, therefore a word (a finger pointing to the stars "focus on the finger and you'll miss out on the glory of the stars to ignite your inner universe) that follows to allow contemplation of this distinction for others to appreciate, and therefore, neuroplastically modifying the comprehender's mind forever, infinitely and infinitesimally.

We do not share words, we share perceptions. 

If we only ever focus on the words and not the ever unfolding and expanding implications of these symbols we shall only ever be in the thought of the word *universe but never realise that we are sharing and are it.

Kindly,

Hi

...

[Hi]

* Therefore when we use these categories, these words... We do not engage the words we engage their meanings, and therefore activate regions in our brain that are anything other than the language regions, relatively speaking. We engage the "meaning (brain) regions" of the words, at least, to the level of understanding, vision, penetration, insight, intuition, perception, awareness, discernment and imagination of the receiver.

"If I ask you to look at the stars (word) in the nights sky and all you end up seeing is the stars (word), you've missed the point" ;) .

Kindly,

Hi, traveling beyond the bounds of the symbol/word *universe and actually activating it in my mind ;) .

On Monday, June 29, 2015 at 1:46:24 PM UTC+10, Hi wrote:

...

[jotaro]

not necessary you can say the same to two people, and they will process it differently.

eh.

besides the point, knowing what something means doesnt mean you are going to use it or something.

--

You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To unsubscribe from this group and stop receiving emails from it, send an email to brain-trainin...@googlegroups.com.
To post to this group, send email to brain-t...@googlegroups.com.

[jotaro]

basically everyone to understand , need to process sentences in terms of his own, to connect it to his previous memories to make sense.

thus the previous ones will decide how you are going to process it and, what kind of insights you will need to process it CORRECTLY, since the one is determined before hand and, the second is really a bit tough to implement,

while possible, it will really be silly to expect people to take THE PERCEPTION

you take . they are going to make it differently

for a simple example, this one is silly too.

for example lets take the meaning of a ball:

well the first kid learnd about a blue large ones, when mother pointed out when he was a kid

an other one was pointed out and saw a red and tiny one.

the third one learned in a classroom from the board and there wasnt a drawing of its meaning, the meaning was explained though words.

so those 3 will have different meanings, perceptions when you say the word ball to them, plus we need to consider the average ball they see.

so yeah we dont share perceptions at all, this is impossible and most perceptions are not translatable to language. because you are going to compile it with your own dictionary. that is of course different from the person who decompiled it to language.

[Hi]

Your incorrect interpretation is an example of exactly what I meant. 

What I meant by "we do not share words, we share perceptions" was more "we share words in attempt to share perceptions". My explicit use of the de-emphasis of the words was primarily to indicate the importance of delving deeply into the layers of meaning of the categories. Be careful not to see those things called stars as stars now you hear! Look beyond the veil. Seek to find and differentiate the interactions within reactions of simultaneity, of all stars, dots, lights and beyond!

:D

[jotaro]

my incorrect interpretation is also an example of what i meant.

see.

what categories? ehh you should specify,

even if i dig deeply i would still will construct different perception.

this effect takes "effect" anyway.

regardless.

JOTARO, when tit for tat has gone wrong, and recent fashionable trend is crap.

[Jack Walters]

There are previous posts and games that answer your question. 

--
You received this message because you are subscribed to a topic in the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/brain-training/6P6BWXBySXk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to brain-trainin...@googlegroups.com.

[4d8d12d]

Hey I'm loving the logical relation game that you've created and have been getting better at it for awhile. Is there a chance you could expand the game to compare at the end an opposite also instead of just checking for equivalence? 

I think it would add a bit of variation as to not make it so predictable, as it becomes possible to 'game' once you've gotten good enough at seeing the patterns that develop.

On Thursday, April 9, 2015 at 7:54:39 PM UTC-5, maxim...@gmail.com wrote:

I've recently made a game that improves object individuation ability, which I believe is an important component of non verbal working memory that is an untapped opportunity to improve. Game and scientific support here:http://twitchmath.com/objectIndividuation.html

Thanks to those who provided feedback so far :)

I'm also planning on making some additional games relating to both non verbal working memory and relational ability, and I wanted to check in if there was a game that a lot of people wanted to see made. 

If no one makes any other suggestions I'll first make a game that stresses non verbal working memory in a format similar to n back but probably closest to lumosity's follow that frog. I'll also probably add some features like variable timing. I believe the follow that frog type format where the possible position are irregular and distributed differently every time you play MAY be better than n back because n back may begin to rely a bit on crystallized intelligence as you develop familiarity because the grid is the exact same every time. 

That said I think follow that frog is fairly limited because it takes about a minute for the game to start every time you play because it completely halts and provides the steps numbered for you every time the path increases by one, also you can pause whenever you want, which I don't think provides the best training.