RaiseYourIQ.com impressions

[RonW]

Score at raiseyouriq.com:

On the pre-test I scored 54, fluid was 140+.  I would have scored 55, however was distracted by the TV downstairs at a crucial moment. Fair enough, though, I suppose.  I wasn't really trying to go as fast as I could, although it wasn't far off.

I don't think these numbers have much of a relationship to an IQ score, and only a limited relationship to actual thinking ability, because:

• Those of us who are good at picturing simple same/opposite and greater/less relationships have an enormous advantage over purely verbal thinkers.
  
• The test items had virtually _no_ complexity. Essentially the 'hardest' ones were ( '> > > >' <? yes/no) pr ('s o s o' o? yes/no) and variants on those (s == same, o == opposite). I had expected some tricky items, at least... . It seems the time constraint/number of axioms was meant to contain the difficulty. I solved many of the items in 5 seconds.
• Due to the limited range of test item types, it was easy to 'game' the system by noticing the pattern in the item wording, the question in the item resolution and applying yes/no automatically thereafter. I really wanted to have to hold all of the wording stages in mind at once, and usually manged to do so, however, under time pressure, being competitive as I am, I simply noticed the 'style' of the axioms/question and automatically applied yes/no. This, I'm sure, is _not_ the intention of the test, or at least should not be it's intention. And certainly should not be the actuality or the intention of the training. The training is however similar to the test.

Greater effort should have been put in to create a structure to force the test taker to hold all of the conditions in mind, as mentioned, and then decide yes/no by applying the question to the 'axioms'. That combined with trickier combinations of 'axioms' might have made a worthier test. One I'd spend my time training for.

They cannot now do this, though, because their test has been 'normed' on the test-takers who've already taken the test. As corroboration I note that their pre-test information revealed that the average of my age group who've taken the test was 35 raw. How that's possible given the simplicity of the items, even under time pressure is difficult to understand, unless of course some of these were 'therapeutic' patients with an obvious deficiency.

As an aside, I suppose that this 'report' was produced automatically, and that therefore there might only have been one other test-taker in my age group, however, srsly? 35? 

Note that I'm not bashing 'relational frame' theory or therapy. I'm sure it works well for those who find real difficulty with the concept of  'bigger than', such as children, but while this 'basic training' may have a good effect in these cases I'm not so sure that it would significantly help anyone who scored 50 or over on the pre-test.

I did email them with my thoughts but have had no reply.  It seems they're really only interested in therapy, not enhancement.

[Alex]

Hey,

i agree with your opinion  that this training doesnot at all increase iq, i have scored even high by up to 50 on the pre assessment test (average was 41) and on the post test 54.
I am sure that chess can improve also this skills, because you learn the same rules for example relational thinking (even complex moves)  and even more (remembering patterns) and training working memory.

As i have just said, i dont give a coin on it.
best wishes,

alex

[maxim...@gmail.com]

Does anyone know of any scientific evidence that shows a correlation or lack of relationship between working memory training and gains in chess performance? My google search returns nothing.

Anecdotally, if I haven't trained in a while, play chess, then train hard for a couple days on the maintenance stage of non verbal working memory with things like lumosity's follow that frog or http://twitchmath.com/enhancedvisualnback.html, then play chess again after allowing my mind time to recover, my chess performance tends to jump in a very noticeable way.

This intuitively seems quite reasonable because I strongly suspect chess and the training rely on at least one or two or the same neural networks, but that is n = 1 of course.

[jagunit]

Do you guys think that this article is BS?  I've been intrigued by RFT for a while but haven't been able to justify paying for the site as of yet.  Thanks for your review of the experience.  Might check it out one day just for the heck of it.

[jagunit]

Regarding chess, I have not seen any study to that effect.  Until you get to the mastery level, chess performance seems more related to one's ability to recall predetermined strategies and pattern recognition; perhaps more taxing on long-term than short-term working memory.  Anecdotally chess does seem to tax WM, but I remember looking through a bunch of studies a while ago and wasn't able to come up with anything.  The only notable study I recall is that chess players seemed to perform better on the Tower of London Planning task, but even that effect was later attributed to the longer planning time that chess players took to complete the task compared to the average population.

[Ron Williams]

I didn't quite say that I don't think the training can raise IQ. What I do think, however, that if you're anywhere near the high end of the intellectual ability spectrum, the current implementation of relational frame training on this site almost certainly won't help you much.

It's currently like a gym where none of the weights is over 50kg - you can clean and jerk 50kg after the training? Well done...

But if you could already do 100kg before the training...

I.e. They need to devise a training for the top end of the scale using relational primitives that will require mental logical operations of some complexity at speed.

(I do realise that 100kg is nowhere near the top end of that discipline)

I'm certainly not dismissing the concept of ingraining basic essential operations as a means of facilitating their use as readily vailable thinking tools. I simply think that the basic set should be much, much more extensive that what they've chosen.

I do understand that what they did choose was probably chosen because they may be the essential tools that poorly performing children are missing, so I don't fault them for that.

I do fault them for expecting that a slightly expanded, but still simplistic training will help already high functioning adults.

--
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]

I've actually had a very rough version of something like this for a while, I'm going to add it to the other games I've posted in about a week or so. It adds if then statements and it is adaptive. The complexity grows very quickly. I think you'll find its massively better than raiseyouriq.com, and quite difficult.

[maxim...@gmail.com]

actually I'm just going to put up the rough version on the site tomorrow morning and then update when I make a better looking one.

[jotaro]

lol i look forward to your new game, you should break down complexity,speed, and duration (should differentiate btw them in customization)

i am not sure what are the operations ron talks about.

however 50kg DUMBBELL is quite the undertaking i am in a gym a few months already i have never seen anyone 50kg dumbbell in one hand

i am not even sure my gym has 50kg dumbbell.

[Ron Williams]

Did I say Dumbbell? I don't think so. I said weight.  Specifically re. 'clean and jerk'

My actual gym as a vast range of different weights. There is, of course a greatest weight in the gym. The hypothetical gym in my post has a maximum constructable and manipulable weight of 50kg. 

This was to model the less-than-maximal intellectual 'weight' in the 'gym' which is 'raiseyouriq.com' and which some of us enter, already beyond what it can do to help us. (do I really have to explain a metaphor that was intended to make my meaning clear? In that case I guess I failed...)

I'm pretty sure my gym does have 50kg dumbbells, but that wasn't what I was referring to... I've never seen anyone use the heavier dumbbells, but when I go there, sometimes the weight machines are on enormous settings.

....

The 'operations' I was talking about would consist (potentially) of operations basic to propositional logic or other mental skills, if those skills aren't based on any type of formal logic. Take a look at the game WFF n Proof to see what I mean to be an example. But not necessarily exactly those.

If there were some basic operations like greater/less-than and same/opposite, which RYIQ.com uses, but which could lead to strengthening more complex thought elements and fluidity in them, then those are the operations I mean. 

I had no fixed idea about that, just a thought that there might be other basic 'operations' than those put forward on that website which might be useful to fix into automatic, fluid usability.

Ron

[jagunit]

So I didn't realize one could take the assessment for free.  I completely agree with you on the questions being rudimentary and not indicative of IQ at the higher levels (I scored 53, while trying to go as fast as possible) but I CANNOT tell you how much RFT reminds me of the LSAT Logic Games section.  

Essentially, it's a MUCH more simple version of the LG section, and rightly so given that on the RFT site you're supposed to perform all the problems in your head (you're free to sketch diagrams for the LSAT).  Given that training for the LSAT improves reasoning ability, I wouldn't be surprised if the RFT site does the same on a much lower scale.  

I tutored the LSAT for a couple of years, and I can say that if you really want to challenge yourself, try doing any of the LSAT logic games without scratch paper, or just doing some of the more difficult games with scratch paper.  After studying for the exam and later tutoring it, I can tell you that my logical reasoning ability has without a doubt improved.  Now I just need to find some more material to go on to next (old GRE analytical section? suggestions welcome.  That WFF game looks interesting).

[maxim...@gmail.com]

So I'm actually going to post two RFT games. A math one and a logic one. I've posted a very rough version of the math one at http://twitchmath.com/mathrelationalability.html. It may take a bit to get the hang of it to get the hang of it and once you have the hang of it I think that you'll find it gets quite difficult after a few correct answers. It's highly adaptive.

I'll post the logic one later today or tomorrow and I'll be making much more sophisticated versions of both. I thought I may as well post the rough version now since you all are already discussing the need for a better RFT game. 

[jotaro]

can u make customization without time limit to use as a training wheel.

umm .

[maxim...@gmail.com]

Good point, I think lost sight of how hard it is to start because first I made the logic one and practiced that a lot, then I made the math one and practiced that a lot so now its hard for me to know how it would look to someone just starting out. I'll modify it to make it easier to get the hang of.

[maxim...@gmail.com]

Okay I modified the starting place to make it easier to get the hang of.

[jotaro]

lol i also have no idea if i get it correct or wrong. should give the user feed back.

and it says u reached level 3 when in fact it is level 1?

no matter if i do yes yes yes

or no no no, as a series to the x=-(y)

x positive i stay at the same place.

x>y?

 i do yes yes yes

and no no no a few times and i still stay at the same placE?

[maxim...@gmail.com]

Also, I added some additional time, you now have 120 seconds. As always, everyone please feel free to suggest changes!

[jotaro]

well 30 seconds in level 3 to increase level is too quick no?

besides, if i dont get it correct i fall to level 1?

this should be a level without time. or something.

[maxim...@gmail.com]

oh sorry I hadn't seen this when I sent my message about adding time.

Thanks for pointing that out about level 3, I changed it.

Can you elaborate a bit? One thing is that in order to get to the next level you need to answer in 30 seconds or less.

[jotaro]

lol i reached level 5 once , too many computations in the head.

i looked at it and didnt even try to do it. :)

i think the logic one should be more interesting .

[jotaro]

actually i was more curious about the logic one from the get go.

[maxim...@gmail.com]

I modified the game to go one level at a time, like the object individuation game. This increases the time spent at the limit of one's ability.

If you are like me, this game can be a bit stressful, especially at first, because it rapidly brings you right to the limit of your relational ability, but I found it was definitely worth it because I noticed big benefits. (n=1 of course)

[jotaro]

i reached level 6 and maintained however the questions seems easier now, there are certain things that stopped popping up that were difficult.

anyway  i am less tired now also.

[maxim...@gmail.com]

I didn't change the questions, maybe your relational ability is increasing :) ?

[jotaro]

i reached level 8.

honestly the computation became automatic, as i thought there is too much predictability in the structure, so i play the same algorithm each time and it works, i just keep one item in head at a time though.

the problems with such games there is no way to introduce novelity on a consistent basis. since its a program. and the statements are the same pattern. umm maybe i will be better at computing equations.

so it feels very task specific.

[jotaro]

or rather i dont need to keep more then a thing at a time in my head.

my favorite game of yours is non verbal n back. its enjoyable.

[jotaro]

so i am looking forward for more balls per time .

[Ron Williams]

Thanks, maxlabs for the math one, looking forward to the logic.

I'm currently I'm at level 7, after about an hour. But I would like to be able to resume...

I know I can leave the browser open, but do you think you could add a cookie to save state? Or maybe just let us start at whatever level we like? Ambition would just get kicked downstairs anyhow :)

And it looks like one might need levels 10,11,12... sometime soon, unless a plateau forms, which would mostly be because of the time limit - i.e. speed of processing will be the limiting factor, at which point, I suppose the real training kicks in. 

There would likely be a speed bottleneck at 1s/manipulation, so, perhaps at that speed, assuming you have to do the substitutions in sequence, at 30s, a maximum level of about 25 might be reasonable, allowing for a glitch or two.

As Jotaro has said, when speed picks up, one simply moves down the list, substituting, so really after a while, what we're doing is honing the speed of application of a few math skills - /2, x2, - , pos>neg, = is != to >. Which is fine, but there are a number of other skills that are worth training, although it might not be obvious what they are. They would be skills that underlie understanding of different branches of mathematics, like set theory, geometry, and other things like tensor operations, quaternions, etc.

Hopefully the process that the brain goes through to speed up the current list of operations might be of assistance in the other areas, but maybe not, too. Let's see...

[maxim...@gmail.com]

Thanks for the feedback Ron. To address your immediate question about saving, my temporary solution to this (which I'm about to replace in the next couple days) is that it saves it in the browser's local storage, so you should be able to open up the game in the same browser tomorrow or the next day and it should maintain the level you were on.

I'm going to replace this with a login system that saves all of your game data so that you never have to worry about this kind of thing and can easily play on any computer or mobile device, in any browser, and regardless of various settings by just signing in. This will also have the benefit of allowing for analysis of performance over time, relationships between the games, etc.

I agree with you guys that the type of relations is very limited. I wanted to get this version up as soon as possible and plan to add additional relations such that the game has more depth. With the game as it you start to get diminishing returns relatively quickly.

I'll post the logic game tomorrow morning.

[Ron Williams]

Ok, another 30 min or so, and I've got to level 9. Picture or it didn't happen ...

I did this by 'cheating', in a sense, though - I'm not sure whether it's entirely kosher vis-a-vis the training, however I found myself subvocalising the operations as I moved down the list, as:

half, quarter, minus half, minus quarter, eighth, etc etc - about 1 or 2 sec per row. Then you end up with, say, -16Fi = (-2Xu)   =>   8Fi = Xu   =>   press left 

It seems that the brain will always find the easy way out, the lazy s.o.b. - it's always looking for simplifications that turn every attempt at stressing it into a nonsense. I think that's why children, after a certain point (at least most of them), stop increasing in intelligence. The world-as-we-know-it runs out of complexities that stress their cognition.

I'll try to do it entirely by visuals alone and suppress the subvocalisation as much as possible. I'd guess that utilises a much more basic brain operation, as no muscle movement and communication to/from is involved. 

Also doing the exercises without subvocalisation should end up being faster - i.e. it should take much less than one second per row, thus the highest level attainable in 30 sec could be significantly more than 30.

Perhaps a game in which the question only makes sense if you hold all of the rows/axioms in mind at once might remove the simple speed determiner.

Ron

[Ron Williams]

I said, "The world-as-we-know-it runs out of complexities that stress their cognition."

Perhaps, assuming this to be true, using this kind of relational training or at least a training that forces us to make something of complex elements cognition might be enhanced by increasing the complexity we can see in the surrounding world - such as by using the Quantum Image Streaming techniques reported by Brandon, assuming that that's what it does (I haven't personally tried it, due to laziness).

[Ron Williams]

Or course, mixing up the rows, so that even if there is a simple substitution path through to the conclusion, you have to find it, would slow us down, and force keeping the last result in memory as you scan for the next step. 

I predict that the maximum level in that case would be not much more than the nine you already have, due to losing at least a second per scan. Of course some intrepid soul might visualise a map of next steps quickly and still run through pretty fast.

It would need some mental power to do that, though. However, would that training be useful in increasing cognitive skills - i.e. would there be any transference to general cognition? Dunno... (Further research is needed?)

[Ron Williams]

To maxlabs

Hey! it seems there is level 10, even though the 'number of successes' at the bottom doesn't show it.

a couple of hours of working and taking a break to have another go ...

After a while, I'm afraid it gets a little like adding up a column of figures. So as you said you're introducing more complexity, that would be a good thing...

[Ron Williams]

Ok, I'm afraid I 'wasted' an evening just doing this over and over, and managed to get to level 15, and maintain it a couple of times, but mostly I seem to stick around levels 13/14, which I guess represents my current speed limit.

I found using just the visuals to be quite hard, and these scores don't represent those attempts. I seem to need to subvocalise while I calculate.

As I said, I look forward to more complexity, however there are aspects of even this game that could be beneficial, such as my apparent ineptitude at not using spoken words. It feels infantile, like the child who sticks their tongue out as they write, forming the letters with its tip. 

It's quite difficult to stop the subvocalisation, though.

[Dorso Lateral]

Ron, what is wrong with subvocalization?  I kind of get why you think this is "cheating," but I also kind of don't.  Are you saying that you'd rather be using your visa-spatial sketchpad?

BTW, my experience with RYIQ was almost the same as yours.

Both you and Joky have made comments suggesting that Maxim's new relational abilities math game doesn't stress working memory enough.  I know where you guys are coming from - when going through RYIQ I definitely had the same thoughts.  However, after thinking and reading about working memory and n-back for several years, are some of us now stuck in a working memory-centric perspective on intelligence?  Does relational frame training need to stress working memory to improve cognition?  I don't know, just asking.

Ultimately, though, it seems inevitable that processing speed will be stressed if the relational chain is not too easy - as you say in your most recent post, RonW, about hitting your speed limit.  Don't know if anyone saw this one yet, but here is a paper that might be of interest: http://www.sciencedirect.com/science/article/pii/S019188691400573X

Do processing speed and short-term storage exhaust the relation between working memory capacity and intelligence?

Abstract

The roles of processing speed (PS) and short-term storage (STM) for explaining the relationship between working memory capacity (WMC) and intelligence are analyzed at the latent variable level. 253 Chinese college students completed thirty-two measures from different content domains tapping the cognitive constructs of interest. The key findings showed that (a) PS accounts for the relationship between WMC and fluid intelligence, (b) STM and PS are required for explaining the correlation between crystallized intelligence and WMC. Therefore, this study provides support for the view that PS underlies the correlation between WMC and intelligence, yet with the nuance that its relevance decreases when cognitive tasks rely on crystallized knowledge and skill.

In looking for more difficult RYIQ (relational frame) exercises, are we looking, essentially for those that force us to process relationships or chains of relationships as quickly as possible?  Is it as simple as that?  Jagunit, what do you think?  You say that LSAT Logic Games are very similar to RYIG, although obviously more complex, do you have any suggestions how to make Maxim's math relations game better?  Having trained Logic Games extensively, you that your training transfers to Maxim's math relations game?  (BTW, Do you think the Logical Reasoning section is also similar, or something else entirely?)

Maxim, the three games you've put up so far are great.  I really like your approach of targeting aspects of cognition that might be bottlenecks for some people.  It seems that RYIQ, with it's low ceiling, targets relational skills bottlenecks among certain populations.  As RonW and I found out, it doesn't seem to be very effective for higher functioning people that have already acquired those skills.  I'm very curious to see how you will make your math relations game more difficult.

BTW, I think one of the points of the Relational Frame Training was to use operant conditioning to train these relational operations - to make them an automatic process of system 1, rather than an effortful process of system 2 thinking (to use the jargon of dual-process theorists).  RYIQ uses a happy "ding" sound to positively reinforce successful trials.  Will you consider adding reinforcement to the game?

[maxim...@gmail.com]

Hey Ron, 

I definitely agree about the need for more complexity, more coming soon. Also, I modified the score distribution display to go up to 15. The good news is it was recording your scores. Also, this game I'm putting up in a bit stresses working memory significantly more.

[maxim...@gmail.com]

Hey Dorso Lateral,

Thanks for the feedback! Thats exactly my approach. My hypothesis is that brain training works well at making you better at specific but important skills like math, engineering, etc, if you are able to identify the neurological component(s) that is(are) holding you back and invest a significant amount of time into training.

Also, thats a great point about positive reinforcement, I'll likely add that into more sophisticated versions.

...

[Ron Williams]

I don't necessarily think that subvocalisation is inherently bad, however, I believe that the truly automatic and basic mental operations do not have a vocal component, simply because they happen too fast for that.

Therefore, I'd like to automatise operations I'm training, so that they'd become more like, to use a programming term, built-in functions, than add-on scripts, which usually run orders of magnitude slower.

If I could do that, as I wrote previously, I should be able to get through levels faster, and thus attain higher levels, but more importantly, potentially have an expanded basic toolset. Which is also why I'm interested in a training based on the elements of formal logic.

[maxim...@gmail.com]

I wanted to let everyone know that I've posted the logic game at http://twitchmath.com/logicrelationalability.html

Unlike the math game, it has greater than linear increases in difficulty as you progress through levels. 

PS. I know what you mean about subvocalization, also I'll explore using the elements of formal logic in a more sophisticated version. 

...

[hallu]

20, subvocalized the intermediate outcomes.

[jotaro]

ROn , verbal information like in math realtion is processed in the same part of the brain then use suv vocalizing

your visual part is unfit for this task, this why u use subvocalization because its the part that process this kind of information. they are at the same place int he brain visuals are useless here.

also LOGIC:), eh in math i reached level 11 in 2 minutes game after i kept going from where i stopped last time

level 8.

[jotaro]

NONVERBAL n back and inviduation game both use visuals, as opposed to the RELATIONAL GAMES,

hey i wonder about logic games though, can it be used to improv the game?

[jotaro]

the logic game, maybe when if statement increase by one, you should increase time allowance to increase a level more then 30 sec,

in level 6 i am short of 20 seconds to go up to level 7 lol.

[maxim...@gmail.com]

It's an interesting question about whether visuals can be used in the logic game. Something I'm going to keep in mind as I make a more sophisticated version.

[maxim...@gmail.com]

I could potentially make the time allowance for a level increase a function of the level. How do other people feel about this? 

[jotaro]

its simple make it optional like the sub levels in the individuation game

and maybe make an other optional mode of "no time limit mode"

if u get it right u go up

if wrong go down,

why is this good? people will be able to exprience 7-9 if statements and in math will be able to reach level 40 with time limit its impossible to exprience this.

:) of course those 2 modes iwll be optional.

[maxim...@gmail.com]

ah true I'll definitely add these dimensions as part of the customizations :) I meant in standard mode as it seems quite reasonable for the time allowed for a level increase to scale up with the number of statements. 

[hallu]

Last score after correcting for extraneous operations, no point in chasing it further.

[jotaro]

lol level 51 why do i disbelieve you? hi you should change stats for math game as its more reasonable to go further lol.

my proposition involves 3 options , the player can choose and switch in btw the game

1)standard->optimizes effeciency 30 sec for level increase

2) training wheels-> with each if statement time increase or time allowance iwll be function of the level including level increase

3)no time limit mode-> this allows to maximize load of complexity.

the second is to train for the first.

and the rest are obivious.

its good for both games.

On Sat, Apr 25, 2015 at 10:33 AM, hallu <hall...@gmail.com> wrote:

Last score after correcting for extraneous operations, no point in chasing it further.

[jotaro]

also what "extraneous operations" means?

[Ron Williams]

I've been playing the logic for a short while , and have found that, since the nonsense letter pairs fall as either opposite or equivalent, I can picture them as falling on either side of a whiteboard with a line down the middle, thus transforming a logical relationship into a basic locational one.

When I did this my level jumped from 4 to 6 immediately. Then the 'logical relation' test becomes more of a test of how many two letter pairings you can remember the location of.

I haven't tried to up the level over 6 because of fatigue.

A difficulty is keeping the locations in mind while doing the if-thens.

Well worth attempting, though.

Prolly hannu could do umpty-squillion with a similar technique...

There are probably other simplifications that yield better results.

I'd have preferred symbols rather than the words opposite/equivalent etc.

I really do advocate looking at WFF n Proof for inspiration. It's apparently popular in at least one ivy-league law dept as a  'fun'  logic training. Google it. They also claimed (c 1970 - ?now) that if you master the game your IQ will improve.

Rules and explanation:
https://www.oercommons.org/authoring/1364-basic-wff-n-proof-a-teaching-guide/view

[Ron Williams]

Although I'm not sure that doing what I did conforms to the spirit of the training, as it transforms a verbal memory operation into a visual one, which happens to be my strength. My point is that 'is located near some things and not near some other things spatially' isn't really the same as 'equivalent' or 'opposite' either. So yes, I found a metaphor that works better for me but avoids the core concepts.

sry hallu != hannu. I'd got it into my head that it was a Finnish name, and chose one. I'd be interested in hearing about the simplifications made, to do what you did.

Jotaro, I expect hallu meant that they'd found a good way of simplifying the necessary operations such that it is extremely easy to run down the column of equations. I briefly did such a thing, but stopped because it wasn't in the spirit, I thought. What I did was only operate on the powers of two with full-blown mental operations, but keep track of the sign using my index finger, extended or not.

Then at the end, there are some possibilities.

1) If there's a negative (index finger extended) => press button for => press for left > right

2) if not negative, if power is one                                                     => press for left not> right

3) if not negative, if left-hand-side has integer 2-power multiplier    => press for left not> right

4) if not negative, if left-hand-side has fractional 2-power multiplier => press for left > right

Thus the final operation becomes automatic, with no thought needed.

I could have devised a similar thing using my foot for +-, and my two hands to keep track of the powers of two (except that sometimes you'd have run out of fingers), but to me, tricks like this are essentially infantile. We need to improve mental operations, as counting on our fingers won't help us with complicated, advanced mathematical concepts.

If I tried visualisation and practiced it, I could probably do something like, say keeping track of the progress by having grey tiles above and below a central one, turn the current tile red for positive, blue for negative, and how high above/below centre is the power of 2. Then where the currently coloured tile is and what colour it is tells everything. It should be very quick, ultimately.

You just need very crisp and stable visualisations.

[Ron Williams]

Of course this trickery would probably become impossible to manage for most people if the steps were randomised - i.e. not in a linear progression: ie/dd -> dd/aw -> aw/ff -> ff/er ... etc, allowing the previous step to be substituted in the next.

And so a much, much harder thing would be to have to scan a randomised list for the next step, as I think I noted in a previous post. And maybe have some of the equations be irrelevant to the final question so that you have to spend more time finding 'the one' - distractors in other words.

But would the extra need for scanning provide a better training, even if it is harder? 

Dunno. You'd think so, but maybe not.

[Dorso Lateral]

Hey Maxim,

Great job with the verbal logic game.  One thought on both the math and verbal game in regards to structure.  Joky is urging training wheel stages and adjusting time allowed for more difficult stages.  I suggest switching from a testing focus to a learning focus.  I mentioned this before, but the n-back task was originally a test.  It was hastily adapted into a progressive-difficulty task but the testing focus remained.  When attempting to improve a basic ability (wiring, essentially), this kind of makes sense.  You're trying to stress the system and force it to adapt.  But logical reasoning skills may not be analogous to working memory and may require a different approach.  Right now, a player struggles to get to the next level and once there struggles again to get to the next level.  Essentially, their current abilities and reasoning skills are being tested.  But is that what these games should be doing - testing current ability?  I would think a slower approach, where the player gets more repetitions at a particular level (ala RYIQ), with success being positively reinforced after each successful trial, would be more conducive to improving these skills, and would improve the chances that such learning would generalize to other tasks requiring the same skills (relational frame fluency, or whatever one wanted to call it.)  

As is, I'm not sure that racing through to the stage that represents ones current ability level and/or working memory capacity is going to result in improvement in these skills or generalization of any improvements.  

One of the things I appreciated about going through RYIQ, despite the low ceiling, was the learning focus.  Most cognitive training games feel like competitively oriented video games and I'm not sure that's the right approach.  It's not how one would teach math skills, for example. 

[maxim...@gmail.com]

Thanks Dorso Lateral.

It's an interesting point about testing vs learning, this hadn't occurred to me. Do you feel that adding a setting that removes all time constraints would go a long ways towards this? Possibly combined with repeating the same level several times?

...

[Dorso Lateral]

Hey Maxim,

I think the time constraint is good.  But I think that passing a level more than once should be required to move to the next level, with each success followed by some kind of reinforcement.     The way the other program is implemented is like this: n number of questions at the same level, with x seconds to do each one.  Player must get all n correct, each within x seconds to go on to next level.  One wrong and player must start level over.  Once the level is passed with 100% accuracy then either a) next level becomes available if player sailed through that level on first try, or b) a test identical to the level (but without feedback/reinforcement for each success) must be passed.  This structure ensures that the player gets a lot of reps and masters (is conditioned to perform excellently) a level before the next level is introduced.  Time allowed for each trial stays the same as levels increase in difficulty. Once the player reaches the final level it's possible to repeat the whole program with x reduced to whatever value, or to randomize and get trials from random levels at whatever speed is selected.  These modes allow for distributed repetition, speed improvement and quizzing (which also reinforces learning).

 That's how they structured it, anyway, and I think this creates a learning environment rather than a testing one. I'm not an educator or anything, so take my thoughts for what they might be worth.  

As a quick tweak to your program, simply requiring multiple successes at the same level before moving on to the next one would reinforce whatever learning is going on, as well as allow the player to relax, which should lead to increased learning and mastery.  I think we want to train ourselves to be able to carry out these logical operations quickly, but also effortlessly.  Constantly being squeezed by the clock may not lead to that state due to stress, although once at the next level the clock again becomes a factor.  

All that said, I haven't spent enough time with it to really judge.  As it exists now, I would want to cruise to whatever level I get stuck at and linger there for a while, continually passing that sticking point level until I feel that I've mastered it, no longer feel the time pressure, and don't have to think about it, then move on to the next level.  But I don't have the option of doing that because one passage of a level automatically kicks me up.  I guess an option to requiring multiple successes at a particular level would simply be to offer a choice.  "You've passed; would you like to practice this level or continue to the next?"

Anyway, I'm very impressed by all your games! 

[Dorso Lateral]

Hallu,

What do you think of these two relations games?  

Any suggestions on how they could be improved?  

[maxim...@gmail.com]

Hey Dorso Lateral,

Thanks again! I think you make a great point and I couldn't agree more with your suggestion of toggling adaptivity.

I'm going to implement it within the next day or so, I think it will fit well.

...

[jotaro]

lol someone reached level 15, in logic

hi man i would love to play logic game "no time limit mode"

just wana have fun without trying to break speed records.

also if i get the level correct i will prefer to go up and not get stuck, if the next level is too hard you fall back again anyway, well thats my take on "no time limit mode"

[Dorso Lateral]

I hope Jokyboy gets what he wants.

I just spent some time on logic, teetering between 8 and 9, then finally made 10.  Each time I messed up 9 I got kicked down to 8 (and was told Nice!!! :), then back to 9.  I guess reworking 8 prepares one for another crack at 9, but I would still prefer to stay at 9 until mastered.  I finally realized I just needed to chill at 9 and work it a couple times correctly without thinking much about the time, and then finally made 10.  Didn't take too long, but I'm now a little intimidated by 10 cuz I know I'll get knocked down to 9 again.  What I would have liked is to stay at 9 even after ducking under the 30 seconds.  I'd like to see the option, once a new level is passed, of staying there for a self-selected number of trials, with right/wrong feedback and neither getting kicked down or up.  

As is, I guess it works, but the ratio of positive feedback (Nice!!!) to negative punishment (knocked down to previous level) is not optimal.  Punishment creates stress and stress is not conducive to learning.  But I'll stop beating that horse.  

Maxim, are you also working on a more/less version of the verbal?  (I guess the math one combines same/opposite and more/less, but the more/less frame seems to have greater play in the real verbal world - bigger/smaller, darker/lighter, faster/slower, etc., etc.) 

...

[maxim...@gmail.com]

Hey I made some changes to both games:

1) It only says nice if you move up now, sorry about that oversight

2) you can now toggle both infinite time and non adaptability on both games

Let me know what you guys think and if you can find any bugs!! Thanks for all the feedback so far :)

...

[Ron Williams]

I still intensely dislike words as mathematical symbols. I'm used to the symbolic form. Try writing Pythagoras' Theorem in words, and see how intelligible it looks, or Maxwell's Equations.

If one is training in a discipline, one should use the conventions of that discipline. If I was training to read James Joyce's 'Finnigan's Wake', symbols might not be optimal...

--

[She will Fly because She is Born]

Maybe you should see if you can add ontological analysis? Perhaps Brandon Wilson has something to say one this subject too?

https://groups.google.com/forum/#!searchin/brain-training/ontological$20categories/brain-training/gh5xgRRwVtc/XiU309FIkLUJ

Also, if you're able to create some variation in the ordering of: "if nF is equivalent to cd then Qy is opposite to nF otherwise Qy is equivalent to nF"

That would be great. Meaning, instead of always having equivalent at the start you could have opposite, or opposite equivalent, further although I'm yet to progress through all the stages, a mixing up with more categories (i.e. like but not equivalent, unlike but not opposite, and even variations of opposite so the brain isn't so pre-preemptive with how it should categories something).

Thank you, what you're doing here is really fantastic :D !!!

...

[Alex]

Hi,

i have tried the logical realtional ability training, and i find it really hard. I am successfully up to level 5-6, but moreover on level 7-9
it becomes more intuitive, i dont know how to overcome limits, without guessing or be intuitiv at all.
But great work anyway, this might real show improvement in logical thinking and might be possible to transfer on math.
This really strenghts my ability and force me, instead of getting boring. But anyway, i have a lots of trouble with the mathmatical one,
instead with the verbal one, but i dont know why.

best wishes,

alex

[jotaro]

lol i doubt this game improve logical ability, simply because it limits it self to one pattern, also most of the difficulity comes from processing if statements.

it takes more proccess it to find the one pattern, sort it to yes or no, here because of the if statments, its more difficult then in the math game to come up with algorithm effeciency.

On Sun, Apr 26, 2015 at 4:32 PM, Alex <ale...@hotmail.de> wrote:

Hi,

i have tried the logical realtional ability training, and i find it really hard. I am successfully up to level 5-6, but moreover

it becomes more intuitive, i dont know how to overcome limits, without guessing or be intuitiv at all.
But great work anyway, this might real show improvement in logical thinking and might be possible to transfer on math.
This really strenghts my ability and force me, instead of getting boring. But anyway, i have a lots of trouble with the mathmatical one,
instead with the verbal one, but i dont know why.

best wishes,

alex

[maxim...@gmail.com]

It will get easier with practice, and you'll find the limit of what your able to do will increase significantly :)

[maxim...@gmail.com]

Thanks for the feedback! Interesting idea about ontological analysis, i agree it would be nice to have more variation, things I'm going to strongly consider for sure as I work on this.

...

[Dorso Lateral]

Thanks Maxim!  Plays great:)

[Dorso Lateral]

The notion that G = Gf = inductive reasoning has been around for some time (decade +).  The notion that G = analogical reasoning is somewhat more recent.  Remember Godel, Escher, Bach?  Its author, Douglas Hofstadter, recently published a book that may be of interest to those interested in relational reasoning: "Surfaces and Essences: Analogy as the Fuel and Fire of Thinking."  

[Freeman]

Hi,

do you by any means still have a harder, more complex alternative to raiseyouriq.com at hand or online? I would be very grateful, and donate to your causes.

Greetings

On Thursday, 23 April 2015 at 07:31:28 UTC+2 maxim...@gmail.com wrote:

actually I'm just going to put up the rough version on the site tomorrow morning and then update when I make a better looking one.

On Wednesday, April 22, 2015 at 10:18:54 PM UTC-7, maxim...@gmail.com wrote:

I've actually had a very rough version of something like this for a while, I'm going to add it to the other games I've posted in about a week or so. It adds if then statements and it is adaptive. The complexity grows very quickly. I think you'll find its massively better than raiseyouriq.com, and quite difficult.

On Wednesday, April 22, 2015 at 7:17:05 PM UTC-7, RonW wrote:

I didn't quite say that I don't think the training can raise IQ. What I do think, however, that if you're anywhere near the high end of the intellectual ability spectrum, the current implementation of relational frame training on this site almost certainly won't help you much.

It's currently like a gym where none of the weights is over 50kg - you can clean and jerk 50kg after the training? Well done...

But if you could already do 100kg before the training...

I.e. They need to devise a training for the top end of the scale using relational primitives that will require mental logical operations of some complexity at speed.

(I do realise that 100kg is nowhere near the top end of that discipline)

I'm certainly not dismissing the concept of ingraining basic essential operations as a means of facilitating their use as readily vailable thinking tools. I simply think that the basic set should be much, much more extensive that what they've chosen.

I do understand that what they did choose was probably chosen because they may be the essential tools that poorly performing children are missing, so I don't fault them for that.

I do fault them for expecting that a slightly expanded, but still simplistic training will help already high functioning adults.

On 22 Apr 2015, at 8:07 pm, Alex <ale...@hotmail.de> wrote:

Hey,

i agree with your opinion  that this training doesnot at all increase iq, i have scored even high by up to 50 on the pre assessment test (average was 41) and on the post test 54.
I am sure that chess can improve also this skills, because you learn the same rules for example relational thinking (even complex moves)  and even more (remembering patterns) and training working memory.

As i have just said, i dont give a coin on it.

best wishes,

alex

[Leonardo]

All the links are dead. Is anyone still trying to create a good relational training? 

[Freeman Isaak]

I hope so!

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/7H3jVG4q_L0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to brain-trainin...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/c4be3d20-8ef9-472b-89df-aa7717b58500n%40googlegroups.com.

[Leonardo]

I have suggested some ideas here, take a look: https://groups.google.com/g/brain-training/c/_yvRKXGlX9E

[itrn...@gmail.com]

I do. I made an app that uses categorical logic(All A is some B, Some B is not C, Some A is the difference of B and C, etc) and conditional logic as well for more complexity. It also has the option of showing each axiom/conditional statement one at a time. Check it out on another post here: https://groups.google.com/g/brain-training/c/_yvRKXGlX9E/m/9m036ThOAQAJ