Syllogimous v3

[Fredo Corleone]

Hi guys I've just released Syllogimous v3.

Much better than the v1 and v2, as syllogisms now are multi-premise and much more difficult than a poly-syllogism (like in v2).

I've also put same/opposite/less/more questions that are of interest to many of you (to know more https://raiseyouriq.com/the-science);

I don't know if you played with MGS when you were younger,  I made this app with a MGS-like theme which I think it's quite cool.

https://4skinskywalker.github.io/Syllogimous-v3/

I could make a slide menu (left or right) and put a leaderboard with lets say top 100 players if you think it's worth doing it. Let me know!

[Uroš Lukić]

What are the effects of training syllogistic logic?

[Fredo Corleone]

That you get better at it, people here also think that by becoming fast at these kind of questions can improve IQ substantially (google Relational Frame Theory).

[green]

Thank you, I will be using this a lot.

[Fredo Corleone]

Just updated it with a bigger list of nouns and with improved typography.

[Leonardo]

In the original RaiseyourIQ, True are False switched places. In the original assessment of your mental capacity there, the only mistake I made was forced because of that. I think that it is a cool add that force you to pay attention to what you are clicking to. 

Anyways, I have done a few and they are really easy. Does it get harder after 100? 

[Leonardo]

I have solved 100 ones and I see that the number of premises has increased, but my mouse is broken and sometimes double-click, making me fail. It would be nice to be able to control the answers with the arrows in the keyboard. 

[Fredo Corleone]

You discovered it xD

It becomes somewhat challenging above 300.

By the way:

Now you can use tab to move through buttons! (Confirm with enter).

[Leonardo]

Premises

• All TIME is COMPLAINT
• No COMPLAINT is LAKE
• All CASE is LAKE

Conclusion

• Some TIME is not LAKE

Those ones tend to be slightly broken. 

[Fredo Corleone]

I'm having a terrible headache right now because I'm sick, so I might be wrong. But I've solved it like this:

It looks true to me.

[After a while... xD]

I've checked the code and this seems to be a valid one:

I didn't stop here and checked Wikipedia too:

Why do you think this is broken?

[Leonardo]

I solved it correctly, but it was so strange and the inclusion of Case seemed so unnecessary that I just assumed something was going wrong. I hope you get better soon. 

[Fredo Corleone]

Ah, yeah, the fact is any new premise in the mix is just noise, basically the algo randomly choose either Subject or Predicate and make new premises, and the more the premises the more the noise as it can choose among multiple Ss and Ps.

If you can figure out a better algo for making it more difficult just tell me, I'll be more than glad to implement it :)

[Leonardo]

I wish I was half as able as you to come up with that stuff! 

[Leonardo]

I think that just deleting de combinations with pointless stuff will suffice. 

[Fredo Corleone]

I don't think there's a problem here, as you won't know what stuff is useless until the end (until you look the conclusion).

You'll have to draw the full picture as you go through premises, if you want to cheat and go straight to the conclusion and work backward is up to you, I don't see it as an intelligent tactic as you are using the game to train and not for the ego of getting score.

Those additional premises are connected to other premises, but just two of to the conclusion. It acts in a similar way to N-back noise, not all stimuli are signal in N-back.

Once you get into the 400-500 score I'm sure you'll have some hard time if you stick to reading premises one by one.

[Jack]

This is very interesting, does anyone have any results to report from using these relational frame games? Also, is there a way to download this to have it permanently? Thanks!

[itrn...@gmail.com]

Cooool will you add an option to show premises one at a time? I feel like that would for sure make it easier for players to use it to the fullest!

[itrn...@gmail.com]

Also, how many premises is the maximum? Just curious

[Alex Bateman]

Hi.

Very good game IMHO.

I got the following 411/500  syllogism problem:

HOPE is PRESIDENT

RELATION is POLITICS

PRESIDENT is RELATION

IMPORTANCE is OPINION

POLITICS is opposite PROFIT

The correct answer according to the APP is:   " IMPORTANCE is PROFIT"

In the premises IMPORTANCE is not "connected" to PROFIT  by a chain of relations.

Seems like IMPORTANCE and OPINION stand isolated together.

Could someone please explain why  "IMPORTANCE is PROFIT" follows?

[itrn...@gmail.com]

Hello Fredo! I really love syllogimous v3, but having to use an index card cut-out to make sure I only see one premise at a time is very uncomfortable... could you perhaps make it so that it can show only one premise/conclusion at a time and I can go back and forth between premises as well?

[Positive]

maybe it could start with 1 premise and could show answer after click ?

[Fredo Corleone]

Everybody asked me that, therefore I'll change the game to show one premise at time. Not immediately but in the near future. I'll let you know when it's there :)

[Luke]

Is it possible to change the game to allow us to choose how many premises we want to use during a session? If I want to play for say 15 or 30 minutes per day I may not want to have to play for 100 syllogisms just to advance by one premise. Is it possible to make the game more customizable? 

[Fredo Corleone]

Sure, the game is open psource https://github.com/4skinSkywalker/Syllogimous-v3 if you want to make a pull request. By the way if you want to hack your level you can open the LocalStorage of your browser and change the level manually! :)

[CLJACKSON04]

Would it be possible to add a timer like RYIQ? I noticed that it is a lot more challenging with time constraints.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/cec3df0d-e78d-4641-84ee-e2f8f0823630n%40googlegroups.com.

[Fredo Corleone]

Sure it's possible, but I'm no more maintaining it. I gotta have a strong reason to work on it, could you provide it? :)

[CLJACKSON04]

In order to create fluency, the reasoning has to be done at speed. That would help us all, yourself included, increase our ability, to increase our ability.

If that’s not reason enough, I believe it was Rand who pointed out that; it’s not enough to seek it, perfection must be attained

To unsubscribe from this group and stop receiving emails from it, send an email to brain-training+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/9a65980c-f512-4555-a0fb-ca52147b8a5an%40googlegroups.com.

[Fredo Corleone]

So the idea is to configure a timer that automatically moves forward (to the next question) after a certain amount of time?

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/9a65980c-f512-4555-a0fb-ca52147b8a5an%40googlegroups.com.

[Bo T]

Yeah , this is how RYIQ.com does it. BTW , i feel like the exercises in this game can be solved by only using working memory ( placing items in an ascending line and holding it in WM , splitting items in two sides , etc..) rather than using "relational skills" , what do you think about that ? 

[Ian Vasquez]

I was also interested in having a timer that skips questions. So I made it happen:

https://iovasquez.github.io/Syllogimous-v3/

Let me know if you are welcoming a PR. 

[CLJACKSON04]

The idea would be an adjustable timer that resets for each correctly answered question. Each question answered correctly, advances you forward as you currently have it. Similarly, for each question you get wrong or fail to answer within the allocated time frame, you should go down one to five questions. 

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/e9c514f4-d410-497d-bb83-0fce6a84005cn%40googlegroups.com.

[성문규]

I’d say the only problems that require relational reasoning in this game are the all/some/none type problems. The less/more and same/opposite type problems can be finessed using visual strategies.

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/8e8f96f1-4c42-47db-af4a-7cf9bb9cdc92n%40googlegroups.com.

[CLJACKSON04]

Nice work Ian! Any way to make it go down a level if you fail to correctly answer the question correctly within the allotted time frame?

To unsubscribe from this group and stop receiving emails from it, send an email to brain-training+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/29976932-a2d7-420f-88b3-10497849eda0n%40googlegroups.com.

[CLJACKSON04]

That may be true but it turns out that more basic relationships are more effective then the more complex relationships like analogies and metaphors.

To unsubscribe from this group and stop receiving emails from it, send an email to brain-training+unsubscribe@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/8e8f96f1-4c42-47db-af4a-7cf9bb9cdc92n%40googlegroups.com.

--

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-training+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/CAHAbeka%2Bk7YMgXspyK_zou6mCem%2Bws695F3hKUTrZK5JdV4vbQ%40mail.gmail.com.

[성문규]

Really? How so? Do you have any sources?

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/8e8f96f1-4c42-47db-af4a-7cf9bb9cdc92n%40googlegroups.com.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/CAHAbeka%2Bk7YMgXspyK_zou6mCem%2Bws695F3hKUTrZK5JdV4vbQ%40mail.gmail.com.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/CACezZ7PAN%2B9k6sMiU%2Bujgm1GiQ6Uf3t%2BYUF3grA%3DsWWQjphwmQ%40mail.gmail.com.

[Bo T]

I think the time restraint might solve this problem at higher levels , since you need way more time to solve if you are only relying on working memory. 

[Fredo Corleone]

I've just added the countdown time in MGS style. Just tell me if something is wrong...

[Fredo Corleone]

Started training again with this and I find it rather mind bending. What I find most difficult is to depict images of words like WHILE in mind head, these kind of words always slow me down.

I will try sticking to the training for at least 3 weeks and see what happens cognition-wise.

Do you guys just read sentences and try to make sense of them or do you use your mind's eye?

I rely heavily on my mind's eye.

If you guys want I could make so that more difficult syllogisms are picked more frequently than trivial ones.

[CLJACKSON04]

Yes, I do. I also have a couple of quick methods that can be used to evaluate IQ gains from relational training. I will send something out as soon as my schedule opens up...this weekend at the latest.

Cheers 

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/CAHAbekZ8FNo06GGKo1v%3DxzQ3H8eMPWoc3Bj-EK3y9B5ivV%3De0g%40mail.gmail.com.

[성문규]

I think having some sort of percentage-style way to modify the frequency of different types of questions(in this case 3 types) would be nice. Also the option to show one premise at a time(without seeing the question ofc).

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/04214498-18eb-4914-9036-2a0654303bden%40googlegroups.com.

[Bo T]

Well it is stated in the SMART studies (the studies which shows that relational training increases IQ ) that opposite/same and less/more questions are used , so I don't think it is wise to decrease the frequency of those questions because they are "easier".  The addition of a time restraint makes the game identical to RYIQ.com, so it should provide the same proven benefits regardless of how trivial we think the exercises are.

[성문규]

Hmm. Imo it is still useful to have the option of just the syllogism questions.

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/d6035b50-f513-4a22-9786-92098d6d5e22n%40googlegroups.com.

[Fredo Corleone]

Would you, everyone, like to have premises with both opposite/same and less/more? 

[Bo T]

Yeah that would be great 

[CLJACKSON04]

Nice Job Fredo! Your modifications are really making a difference.

As far as sources for the more fundamental relationships having a more significant effect on IQ gains, please refer to this article:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7198658/

Cheers

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/2d6c2aea-aba2-4472-8991-feb564f99324n%40googlegroups.com.

[Fredo Corleone]

My Syllogimous is lacking the following kind of questions.

Which I think I can add. The thing is to take any 4 items anywhere in the premises chain and to ask if the relation between a pair and another pair of items is the same or different.

Same/different is kinda cool here as you don't just need to visualize but you also have to see if relations are the same which is a step higher IMO.

I'm adding this to Syllogimous v3! :D

[Fredo Corleone]

Did it! Now there are these new kind of questions in Syllogimous v3, but those are available only if you solved correctly at least 100 syllogisms.

This new class of questions is the fourth one in the game.

https://4skinskywalker.github.io/Syllogimous-v3/

If you have any other kind of questions that might be easy to implement don't hesitate to tell me, we are going to make this togheter! :D

[성문규]

Oh wow these look great!!!

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/n7lUYs307qU/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/0c37e43a-de12-4bec-8078-f98fdb5fe026n%40googlegroups.com.

[Bo T]

Thanks a lot Fredo! 

[The Realest Ever]

Hey Fredo, 

Can you create the ability to review the last question after answering?

It really bothers me when I get one wrong and I would like the ability to go back and reason it out. 

Plus,  

If anyone gets stuck they would be able to get help

and also if there's any bugs they'll be able to bring it to your attention

Thanks for the app man this is a great one!

[Fredo Corleone]

Sure I can create a history. I can also put a link to a GH issue page in which people can post and discuss bugs.

[CLJACKSON04]

How hard would it be to have each premise directly related to a prior or following premise as in raiseyouriq.com?

I'm finding that as I get up in levels, having to search for the premises and sort them detracts from the reasoning required to answer the question.

--

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 view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/63e15dab-9c68-4562-a582-8fc0f5765146n%40googlegroups.com.

[Fredo Corleone]

Not hard, I did it in the past. But using the conclusion as a new premise poses few problems...

I think it's fine as it is :)

[Tofe]

has anyone seen benefits from this game?

[Fredo Corleone]

I noticed an increase in awereness, especially while dreaming.

[Hak Fin]

I'll be happy to paypal you some money for the work you did and if you can make the latest version offline with the ability to choose stages/types of syllogisms + timer, I'll be happy to send you more.

Please send deets, cheers!

On Friday, September 10, 2021 at 11:00:00 AM UTC-4 stopchemt...@gmail.com wrote:

Hi guys I've just released Syllogimous v3.

Much better than the v1 and v2, as syllogisms now are multi-premise and much more difficult than a poly-syllogism (like in v2).

I've also put same/opposite/less/more questions that are of interest to many of you (to know more https://raiseyouriq.com/the-science);

I don't know if you played with MGS when you were younger,  I made this app with a MGS-like theme which I think it's quite cool.

https://4skinskywalker.github.io/Syllogimous-v3/

I could make a slide menu (left or right) and put a leaderboard with lets say top 100 players if you think it's worth doing it. Let me know!

--

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 view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/cec3df0d-e78d-4641-84ee-e2f8f0823630n%40googlegroups.com.

[Fredo Corleone]

No need to have money for that, but thanks :)

Anyway, would you like to be able to use the web app offline through the browser? That's easy I can make it a Progressive Web App, this means your browser will cache files and even when you are offline the app works as usual.

The next update will contain the following features:

1. A left slide-right menu with settings (turn on and off categories, number of premises, reset score, ...) + restore default settings button;
2. A right slide-left menu with the history of problems/answers (with corrections);

3. Make the web app work offline (PWA).

For the future I'll consider adding an additional sliding panel bottom-top that's shows in-depths stats such as avg response time, right and wrong totals, right/wrong overtime (line chart), total play time and other metrics like these, but I have to think about this.

[Uros Lukic]

Could we have an option to change the terms into alphabet (A instead of Tomato, B instead of Energy) or have entirely custom lists of terms?

[robert chalean]

The App is online https://coloralphabet.herokuapp.com/em/ (you have to change berre mode to derecho) and (new sys to normal sys) .... My idea is that the app have for example 2000 images.... but now i can not do the modification

El jueves, 12 de mayo de 2022 a las 14:56:39 UTC-3, The Realest Ever escribió:

I second this, but as an extremely visual person I would like to do these exercises with small thumbnail images as opposed to letters / words. 

Robert Chilean had that option when he had his syllogimous up and I found it really beneficial, although he had a image of a dildo in there for some reason, lol lets leave that out.
(on a side note its interesting how I can still remember most of those images even though I haven't done that exercise in months.)

[Fredo Corleone]

I didn't forget about making the requested changes, tomorrow those will be presumably online for the testing!

I'm currently scripting the UI for the various settings and the history.

Stay tuned...

[Fredo Corleone]

Syllogimous v3 is online at https://4skinskywalker.github.io/Syllogimous-v3/

I hope you enjoy this!

Changelog:
- Number input for changing the timer time;
- Checkboxes for choosing which categories to play;
- Number input to set the quantity of premises;

- Carousel to play one premise at time;
- Right pane to see your whole history with corrections;
- Feedback overlay.

Todo:

1. Bottom pane that shows various statistics (play time, accuracy, speed, preferred category, weak category, ...);
2. Proxy continuos measure for QI (gotta figure out a formula (maybe with machine learning???)).

If you have anything to say just ask. :D

[Fredo Corleone]

Sorry to bump this thread up once again, but this is the best exercise I've done in ages for my imagination.

[robert chalean]

Ithink is very good a big list of words or images to no repeat or make confusion between exercises . Or use letter because is ver abstract and not produce confusion for me every color is a letter

[Fredo Corleone]

Hi Robert! I wouldn't advise using colors because some users have color blindness, but it's a cool idea for those which are not affected by it.

About the letters instead of words: it was initially with letters but it wasn't that much stimulating to me so I decided to go with words as they tickle my brain much more. :)

It might be better to edit the wordlist to remove all very abstract words and keep, or even enrich it, with more concrete words.

[robert chalean]

I use colors alphabet since 2012 I learned showing my apps to other people that my ability to recognize colors. Is below average but with training is better. 

[robert chalean]

... is better in reading letter but for example to my see the circle pupil in the eyes is very difficult I see al eye almost the same color I thing that color recognition or sensitivity improve sociability or extrovers ion

[Fredo Corleone]

I thought about this last two messages for a while before answering and I have to say I don't really understand them. Is it a feature request for Syllogimous or some personal experience you wanted to share (or both)? Feel free to ping me privately if you need to. :)

[Leonardo]

Wow, man. With the timer on this can get really, really, really intense. It is impossible that this doesn't make you faster on the long run. 

What it was cool about Robert's is that you can configure how much time any single premise is shown, so it forces you to internalize it in, for example, half a second. Which was insane and pretty hard, and I think that was the reason why I became dramatically faster when I trained at Robert's Silogismos daily. 

Another interesting feature that both RYIQ and Robert's have is the random positions of True and False, which, in mi opinion, serves as a another way to train opposition. 

[Leonardo]

Oh, nevermind, the timer doesn't work right at all, at least in my PC. It keeps running while it shows me that I was right, immediately adding me an undeserved mistake. 

[Leonardo]

I hit reset to clean all the mess by the timer, but now I see that some answers considered as Right still add up as Wrong. Maybe my PC is fucked up or something. 

[Leonardo]

"Although the RAI demonstrated considerable predictive validity across the test battery, closer investigation indicated the RAI’s relatively limited utility in discriminating performance for high IQ participants due to a potential ceiling effect. As such, the authors concluded that the inclusion of a wider range of relational tasks, such as temporality and analogy, may be beneficial in parsing out individual differences across a greater diversity of trial types and providing a more comprehensive account of relational ability and how its various aspects relate differentially to various aspects of intelligence, as typically assessed."

It would be nice to include temporal questions. As someone with a +140 IQ, I confirm that it is too trivial to train with a limited set of frames. There are much more to employ, and I have developed other ways to train them outside of this, but I think that everyone would benefit from the inclusion of more frames in your software. Particularly you, whom I have in high regard, and to whom I wish the absolute best. 

[Leonardo]

Such a development should improve the utility of a relational abilities index in providing a more sensitive and nuanced differentiation of performances, in particular at the higher end of the performance spectrum (Colbert et al., 2017; Gore, Barnes-Holmes, & Murphy, 2010). As such, the Relational Abilities Index+ (RAI+) has been developed for this purpose, and assesses performance across five modules of relational responding (Same/Opposite, More/Less, Same/Different, Before/After, and Analogy) in comparison to the two modules assessed by the original RAI (Same/Opposite and More/Less).

There is where we all want to be. So it makes sense to develop strategies centered around that. 

[Fredo Corleone]

1) "What it was cool about Robert's is that you can configure how much time any single premise is shown, so it forces you to internalize it in, for example, half a second. Which was insane and pretty hard, and I think that was the reason why I became dramatically faster when I trained at Robert's Silogismos daily."
I could do it pretty easily but my idea for the timer was different: I wanted the time to be tied to the whole question (premise + conclusions) and the carousel to be manual not automatic.
I don't think it changes much in the end.

2) "Another interesting feature that both RYIQ and Robert's have is the random positions of True and False, which, in mi opinion, serves as a another way to train opposition."
What should this be other than noise?
I mean the crux of the exercise is to understanding the premises and verify the conclusion, not really to be careful on where the buttons go.

3) "Oh, nevermind, the timer doesn't work right at all, at least in my PC. It keeps running while it shows me that I was right, immediately adding me an undeserved mistake. "
I just tried it and it works, could you please provide a video of it?
Thanks

4) "I hit reset to clean all the mess by the timer, but now I see that some answers considered as Right still add up as Wrong. Maybe my PC is fucked up or something. "

If you open the right pane you can see the history, is it showing correctly?

5) " It would be nice to include temporal questions"

What's the difference between more/less and before/after?
IMO there is no difference in the way you solve them.

6) " "[...](Same/Opposite, More/Less, Same/Different, Before/After, and Analogy) in comparison to the two modules assessed by the original RAI (Same/Opposite and More/Less)." There is where we all want to be. So it makes sense to develop strategies centered around that. "

Also don't forget that my syllogimous implements syllogisms which IMO are way harder than the rest.

[Leonardo]

I think that, despite being apparently the same thing, your brain employs another set of neurons to visualize temporal relationships. Additionally, when I train with pen and paper I employ time as a modifier of the rest of the relationships, which holds a truth value which is dependent of the temporal location. 

[Leonardo]

I must say that your dictionary is top notch. Really hard and diverse words. Way better than Robert's. 

[Leonardo]

LOL. You must find a way to avoid this. It is very common. 

[Fredo Corleone]

I can avoid this really easily but it never bothered me, it also happens less and less as you progress.
If you guys really dislike this I can make so that the conclusion is never a premise.

[성문규]

I do have to say, I was forced to train analogy extensively as a child, and it worked wonders for my brain even if I hated it. If more complex analogies can be made in an automatic fashion, it would be extremely beneficial. But I suppose that's near impossible to do unless we have something akin to AGI.

2022년 6월 5일 (일) 오전 7:34, Fredo Corleone <stopchemt...@gmail.com>님이 작성:

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/e2a330b3-aeae-4844-afd2-ac2086c86874n%40googlegroups.com.

[성문규]

Once again fredo, thank you so much this. It is the only logic training app that is  extensive in variety and accommodating for individual training purposes and actually scales in difficulty very well. Amazing contribution.

2022년 6월 5일 (일) 오전 8:15, 성문규 <itrn...@gmail.com>님이 작성:

But yea if there's also an option for just syllogisms in settings somewhere you can toggle on and off.

2022년 6월 5일 (일) 오전 8:12, 성문규 <itrn...@gmail.com>님이 작성:

[Fredo Corleone]

Do you have any idea? If you have templates for questions I will be more than willing to see them.

Thanks

[Fredo Corleone]

One idea I have is introducing a class of puzzles like the following:

Premises:

1. A says that B is lying

2. B says that C is lying

3. C says that both A and B are lying

Conclusion:

B is telling the truth

How to solve this:

"If C is a truth-teller, then his claim is true that both A and B are liars. But A would be telling the truth by claiming that B is a liar, so A can't be a liar. So C can only be a liar and it is not true that both A and B are liars. At least one is a truth-teller. But which one?

If C is a liar, which we’ve established from the opposite case, what are the other two, by necessity? B would be a truth-teller, since his statement that C is lying is a truth. Now that B is a truth-teller, it follows by necessity that A is lying about B being a liar, so A is a liar.

C can only be a liar, and from that, A is a liar, and B is a truth-teller."

[Engin Dinçer]

With what did you compare in order to say it worked?

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/CAHAbekZ2XjCQeCL3xihu29CSBv0ZuDGMdJCAXoH_Zp0QpOsiQw%40mail.gmail.com.

[Moon Kyu Seong]

My iq tests and general ability to study in school. But I have to say, I was between the ages of 10 to 12 while I did that training, and so it is also likely that my brain's development also contributed to the effect.

2022년 6월 5일 (일) 오전 10:01, Engin Dinçer <edinc...@gmail.com>님이 작성:

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/CAG0o83cpDZUvGFaLMN6BUo%3DZHNXjV3QhtddnJ9akU1MwzoUsZg%40mail.gmail.com.

[Moon Kyu Seong]

I was just thinking of general analogies based on commonly accepted meanings of words. Your analogy problems have the analogical relationships themselves limited to same/opposite, more/less, etc. But in a case like: space : astronaut :: biology : biologist, the relationship is more like B explores/investigates A. 

Obviously such relations are not formal and depends heavily on meaning which is subjective, but that doesn't mean the relations themselves lack any logic. More complex/detailed relations can be like: space : astronaut : astronomer :: marine biology : diver : marine analyst, where the relation is something like B explores A up close while C explores A from afar. Obviously not perfect, but that doesn't mean it's completely incorrect either.

2022년 6월 5일 (일) 오전 9:00, Fredo Corleone <stopchemt...@gmail.com>님이 작성:

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/9117a9d0-fdcb-4be1-9330-f7382096726an%40googlegroups.com.

[Moon Kyu Seong]

Now, as you can guess, the more terms are involved, the more specific and complex the relation itself would be. Fill in the blank problems automatically require the player to be able to put all those terms together in a sentence that demonstrates their interrelations, of which there are possibly many that work. But with more words, it can be narrowed down and the player has to be able to keep all of those meanings and possible relations in their mind to be able to narrow it down correctly. Idk this is just my take. Look up Center for Talented Youth analogy exams. I trained for those as a kid. I have yet to find an automatic software that scales in difficulty in such a comprehensive problem type. I assume it requires something like AGI or by the very least some very advanced NLP softwares.

2022년 6월 5일 (일) 오후 12:15, Moon Kyu Seong <pseo...@student.kis.or.kr>님이 작성:

[Fredo Corleone]

Or maybe making a function that picks and arrange things from a semantic relations database such as WordNet.

[Fredo Corleone]

I was wondering if it's OK to introduce a new class of problems based on semantic in Syllogimous, because it will require knowledge to answer.

[Leonardo]

It would be nice, yes, if you do so. As you said, I don't think it must be particularly difficult. Thanks, man. You are great. 

[성문규]

for relatively simple ones sure but for more complex and nuanced sentences/relations we may see errors. Idk only one way to tell lol.

To view this discussion on the web visit https://groups.google.com/d/msgid/brain-training/be788cc0-a954-45c1-b664-108911b9260bn%40googlegroups.com.

[Leonardo]

There is already a lot of apps for Android that make you train this kind of analogies. I didn't find them very useful for increasing my cognitive abilities. 

Also, you can add any kind of option you find interesting, as long as it can be deactivated in the settings, who cares? 

Anyways, the software is pretty good at this point. Maybe the best for training. 

[Leonardo]

With:  "It would be nice, yes, if you do so. As you said, I don't think it must be particularly difficult. Thanks, man. You are great. "

I was answering to: "If you guys really dislike this I can make so that the conclusion is never a premise."

[Leonardo]

I find it weird that you got subtracted one point when you fail. For example, if I'm 6/6 and I fail one I get 5/7. Shouldn't it be 6/7?

[Fredo Corleone]

I agree it would be more intuitive to put correct/played, but this way I can introduce, and I think I will, the missed question as a 0 score, so that correct is +1, missed (to slow) is 0 and wrong is -1.

If I remember correctly at the moment it's -1 for both missed and wrong, but I want to change it to the above.

[Leonardo]

That makes sense. Thank you. I will try to record the issue with the timer. 

[Leonardo]

Sorry about the million opened tabs. 

https://1fichier.com/?t88kr9exafe6ljgob2cf

[Leonardo]

As you can see, the timer keeps going on while the 'right' screen or the 'history' screen is showing.