An idea for a new app based on relational stuff

[Fredo Corleone]

Hi guys,

today I've read something in a couple of paper about how important is for intelligence to relate things to each other. I thought it would be cool to make a game in which there are sentences which are based on references from/to subjects which are relative, and not absolute, to each other. And I was thinking about a peculiar property of family trees: they are possibly hard to reason about in terms of relation.

Look at this phrase:

This man's father is my father's son.

Who is the man?

- Me

- Son

- Uncle

- Grandson

- Not identifiable?

It's my son, but it's not obvious at the snap of two fingers.

Or see this:

The grandgrandfather of this man's son is my father.

Who's the man?

- Me

- Son

- Someone else?

Still my son, was it all this obvious?

OK, last one:

This man's father is my son's son.

Who is the man?

(And who am I? Haha)

Do you think having this kind of question computationally generated in a app would prove a legit relational-ish training?

I think I could make an app that generates sentences like that (when I'll have time!!!), but if you guys have something to say against the legitimacy of it tell me so that I won't waste time.

Thanks 

[ticktock man]

I have been writing a lot about RFT here. I think that your idea is interesting, but far too simple and already exploited in aptitude tests.

Here is part of what I wrote:

brain training game ideas

Visto 372 veces

Suscribirse 

Foto del perfil de itrn...@gmail.com

itrn...@gmail.com

22 abr

Hello, I've been learning to program games for a little while and at this point I can make some

Foto del perfil de itrn...@gmail.com

itrn...@gmail.com

22 abr

I am also considering a variation where each trial would consist of a short series of transformations

Foto del perfil de fT3g0

fT3g0

28 abr

I think adding a 3D element to it would be cool. There is this application https://4skinskywalker.

Foto del perfil de itrn...@gmail.com

itrn...@gmail.com

10 may

https://drive.google.com/drive/folders/1jOs2pwFn9mydpYka2nqFd_AjnqIyOGqt?usp=sharing Above is the

Foto del perfil de Leonardo

Leonardo

10 may

Does not run in Windows? Bummer. El lunes, 10 de mayo de 2021 a las 13:35:16 UTC+2, itrn...@gmail.com

Foto del perfil de itrn...@gmail.com

itrn...@gmail.com

10 may

windows: https://drive.google.com/file/d/1Ej5gHbn98gjPOWBdbrsGUMR80wIA0CdM/view?usp=sharing On Monday

Foto del perfil de Leonardo

Leonardo

10 may

a Dual N-Back, Brain Training & Intelligence

Something both easy and interesting would be a syllogimous with more than two premises: 

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

Relational Frame Theory offers the best approach that we currently have to potentiate mental capacity. I believe that any training must be focused on relational ability, which indirectly will also train working memory. 

Read the comments here, there were a lot of crazy ideas: 

https://groups.google.com/g/brain-training/c/UPOBpFBSH6g

"N sets of syllogisms can be situated in Y number of places, altering the combination between premises. For example, Toro is hotter than Chivo. But places 1 and 2 have opposite qualities, therefore, if Toro is placed on P1 and Chivo in P2 (we can use even names of cities), the relationship is inverted and and Chivo is hotter than Toro. The training may follow this steps:

- You learn the characteristics of the objects in the fashion that I have explained before. With the multiple categories and all. 

- You learn the map and the relationships between points. 

- Then, you are asked: Given that object 1 is situated in place 2 and that object 2 is situated in place 1:

 And then keep asking all the questions about the two objects. Including positions, if you want. 

I hope that I have expressed myself clearly enough."

"If the two places or cities are connected by multiple steps, you can have multiple modifiers. 1 opposite to 2, 2 same as 3, etc. It is a cool way of adding a same-opposite layer of logic to a more-less layer. I am sure that we can then create above that a third layer with all-no-some. We will have then the perfect relational system. "

"Leonardo

14 oct 2020 14:06:36







a Dual N-Back, Brain Training & Intelligence

Every place can have a all-no-some syllogism attach to it. Then, its sign will be determined by its own syllogism, no by their connections. That is the answer to have all the layers at once! The connections will still be needed for the journey, but the same-opposite status of a place would be determined by its internal logic. 

For example:

Chivo is hotter than Toro.

Seattle connects with Paris.

Paris connects with Madrid.

Madrid connects with nothing. 

Seattle:

All flowers are plants.

Some plants are green. 

All flowers are green. 

If that is true: Seattle is same.

If that is false: Seattle is opposite. 

Seattle is opposite. 

Paris:

Some dirt is blue.

Some cars are blue.

All cars are dirt.

Paris is opposite. 

Madrid:

All humans are mortal.

Aristotle is human.

Aristotle is mortal.

Madrid is same. 

Given that Chivo is in Seattle and Toro in Madrid: Is Chivo hotter than Toro?

First modifier in Seattle: Opposite: Chivo is not hotter than Toro

Second modifier in Paris: Opposite: Chivo is hotter than Toro.

Third modifier in Madrid: Same: Chivo is hotter than Toro.

Answer: Yes. 

The interesting part is that you can have any none number of syllogisms in a city, and then determine the sign of the city in function of the majority of them. For example:

Berlin:

If the majority is true: Same.

If the majority is false: Opposite.

No Door is War.

All War is Pore.

No Door is Pore. 

False. 

All Entity is Death. 

No Death is Rock.

No Entity is Rock. 

True. 

Some Beatle is Train.

No Train is Volcano.

Some Beatle is not Volcane. 

True. 

Berlin has a Same sign. "

"We can have categorical and conditional syllogisms. By definition, syllogisms must have 3 terms. I wonder how can we made them more complex... Maybe a chain of syllogism forming one bigger argument? 

A polysyllogism (also called multi-premise syllogism, sorites, climax, or gradatio) is a string of any number of propositions forming together a sequence of syllogisms such that the conclusion of each syllogism, together with the next proposition, is a premise for the next, and so on. Each constituent syllogism is called a prosyllogism except the very last, because the conclusion of the last syllogism is not a premise for another syllogism. 

Maybe we can determine the sign of a city with chains of syllogisms. Or we can try probabilistic reasoning, which is how humans think in order to determine the sign of the city. If 0.50 to 1=same if 0 to 0.49=false. 

https://en.wikipedia.org/wiki/Probabilistic_logic

Introducing Bayesianism to this will be fucking nuts and advanced. Should we start a new thread to discuss all our crazy ideas? "

"Examples of pollysyllogisms that we can use to determine the sign of our cities:

https://www.memoriapress.com/wp-content/uploads/TL2-Ch10.pdf

Example of the possible advanced bayesian approach for an additional level of complexity (only speculation, I will like to see the other things applied first):

Bayes' Theorem is a means of quantifying uncertainty. Based on probability theory, the theorem defines a rule for refining an hypothesis by factoring in additional evidence and background information, and leads to a number representing the degree of probability that the hypothesis is true. To demonstrate an application of Bayes' Theorem, suppose that we have a covered basket that contains three balls, each of which may be green or red. In a blind test, we reach in and pull out a red ball. We return the ball to the basket and try again, again pulling out a red ball. Once more, we return the ball to the basket and pull a ball out - red again. We form a hypothesis that all the balls are all, in fact, red. Bayes' Theorem can be used to calculate the probability (p) that all the balls are red (an event labeled as "A") given (symbolized as "|") that all the selections have been red (an event labeled as "B"):

p(A|B) = p{A + B}/p{B}

Of all the possible combinations (RRR, RRG, RGG, GGG), the chance that all the balls are red is 1/4; in 1/8 of all possible outcomes, all the balls are red AND all the selections are red. Bayes' Theorem calculates the probability that all the balls in the basket are red, given that all the selections have been red as .5 (probabilities are expressed as numbers between 0. and 1., with "1." indicating 100% probability and "0." indicating zero probability).

https://whatis.techtarget.com/definition/Bayesian-logic

[ticktock man]

Nowadays I believe that logic gates are the most interesting way of connecting places.

You can have something like the Silogismos training of Robert, but with interconnected nodules or cities whose characteristics are determined by the Boolean signs they carry. 

For example: You have modules A, B and C. 

Inside A you have your relational proposition:

Car is bigger than Fox

Fox is bigger than Dove

etc.

Is Dove bigger than car?

If true then 1, if false then 0.

If you have a logic gate that modifies the output from A with, for example, a NOT, then you send a 0, given that the output must be the opposite of what it would normally be.

That 0 means that what you send down the next module alters the output of this new one. For example, imagine than A outputs 0 and B outputs 1, but before C there is an OR logic gate. Then, the result from it is 1, and than means that the result from C would be a 1 if C is an AND gate and the answer to its own syllogism is yes.

Except for the fist step, every new module will be a logic gate determineb by the external input and the internal solution.

You can add difficulty by adding propositions inside the nodules and my making extensive webs of modules. If you only show one piece at a time and force the user to rely on memory, that training would be the hardest possible.

[ticktock man]

If you have issues understanding it, I can draw it for you.

Excuse my poor English. Please, make questions. 

[Fredo Corleone]

I have to say that I disagree with how other's have tried to implement relational training so far.

The focus seemed to be more on the amount of the relations rather than the difficulty of the relation itself.

With what I'm suggesting I should be able to shift the effort from memory to understanding.

Let's take two example:

a) All V is O, all V is X, some O is not X. (which is one of the difficult ones)

b) This man's father is my father's son. (which is one of the easy ones)

They are, for my brain at least, on par in terms of difficulty.

The only way to make "a" more difficult is by including more premises and this doesn't change the nature of the relation, it just makes it more intensive from a memory standpoint.

Instead with "b" you can increase the difficult by changing the quality of the relation, for example by using the man in relation to his grandgrandfather or his wife or his uncle etc...

 

[ticktock man]

I have played thousands of syllogisms in your app.

It is not just about memory, which is crucial anyways, but about layers of modificators above layers of modificators. The relationships of RFT are same-opposite and more-less. People have tried to practice them separately, but there are ways to force yourself to resort to several convoluted ways of interacting with all kinds of relationships simultaneously, which is what you are going to find in the real world when trying, for example, to understand a book or a complex situation.

A good training must, in my opinion:

- Present all kind of relationships. (So no ability is left behind)

- Potentiate memory. (Given that memory is crucial in life for learning and understanding)

- Potentiate speed. (Given that in life one does not have infinite time to respond and sometimes must act in a blink)

- Work with recursivity and loopliness. (Given that are essential qualities of human languages)

[Leonardo]

And also have a spatial component. 

[Leonardo]

The cool thing about Silogismos by Robert was the use of regular words instead of mumbo jumbo, something that forced you to visualize extremely complex scenes once the number of propositions was high enough. 

[robert chalean]

I think relational game of syllogismos is better with less elements then you can do the history in your mind automatically in less than 4 seconds.... or something like that. 

I am working to upload the games in other free server.

[Leonardo]

Speed is fundamental. That is why I used the 0,5 or 0,4 seconds option. I was able to do 10 silogismos with 0,4 each. 

[itrn...@gmail.com]

If you use android, did you try this? I made it, and I'd like some feedback.

https://play.google.com/store/apps/details?id=org.godotengine.Categorical_Logic

[Fredo Corleone]

I see my Syllogimous game (that was to my knowledge the first) has started a good trend of apps! xD

Guys if you happen to copy any of my code please put a notice somewhere.

[ticktock man]

The only thing Syllogimous is lacking to be perfect is:

a) more premises.

b) a better way to track progress. I have lost my progress to many times no not be frustrated with it. I have never been able no reach 1000 for that reason alone.

[Fredo Corleone]

I can address that by:

a) Implementing polysillogisms;
b) Making so that you register an account to play OR making you download a file that acts as your savedata.

[ticktock man]

Haha

That would be neat.

Thank you.