Comment author: Gunnar_Zarncke 14 May 2015 08:09:00PM *  1 point [-]

I took your survey and recommend the following changes:

  • Allow zero as an allowed value. I tried to set 0 and failed.

  • Consider allowing non-integer values (I considered 0,5)

  • Add a text field for comments (I wanted to give the above feedback there; it also allows simple anonymous feedback in a way LW doesn't - beware trivial inconveniences)

  • Consider to add a field to give personal estimate of crush-ratios - you can use the to report calibration

Note that you could much increase participation by adding a LW poll of the same. I can do if you don't know how.

Comment author: necate 15 May 2015 06:03:48PM 0 points [-]

I dont know how, so you do it

Comment author: sixes_and_sevens 14 May 2015 03:14:42PM 0 points [-]

There are various things wrong with this reasoning, but I don't think you're getting my general point: this entire approach is misguided and it will not lead you to good outcomes.

Comment author: necate 14 May 2015 03:45:31PM *  1 point [-]

I accept that you and most people here think this aproach is not helpfull. I will therefor abandon it. However you said that there are various other things wrong with my reasoning even if the aproach was not generally bad. Is my general process of asigning probabilities to believes wrong? Would the thing I did in the following paragraph also be wrong in an abstract scenario, where I would for example want to differentiate between blue and red balls instead of someone having a crush on me or not?

" If however this person gives the response with 100% Chance to a Person he/she has a crush on, but also with 50% chance to a person he/she just likes as a friend, then this signal will only help me differentiate between the states "friend" and "love intrest" with 50% probability. And now it becomes relevant on how many of his/her friends this person has a crush.

Let us say the person has 10 friends and has a crush on 5 of them. Than on average he/she would give 5 correct positive signals 2.5 false positive signals and 2.5 correct negative signals. So if I get a positive signal, than that means that with a probability of 2/3 that person would have a crush on me and with a probability of 1/3 he/she would not."

Comment author: ChristianKl 14 May 2015 02:13:16PM 1 point [-]

I don't think I need to have spend 5 hours in private conversation with a girl to be interested in her to the extend that I want to ask her for a date. I don't think that "know well enough that you could have a crush on them" is a meaningful category.

However I think this part is less important than finding the prior, because most people have at least a general idea about what certain signals mean from personal experience

Physical intimacy is a signal that a person likes me, but without knowing the baseline for the other person it means little.

Comment author: necate 14 May 2015 02:57:03PM 0 points [-]

Have you got any advise on how to rephrase tis definition? The Idea was that everyone knows best for himself how long it takes do develope a crush on someone. Hence the category "know well enough that you could have a crush on them" . Hovever i was not sure that everyone would be able to define this for himself therefori addet the backup category. If you dont need 5 hours to find out if you would want to go on a date with a girl, than you know what "know well enough that you could have a crush on them" means for you and you should go by your personal definition.

Comment author: sixes_and_sevens 14 May 2015 01:47:40PM 4 points [-]

It's important to note that the base rate of people finding other people attractive is different from the base rate of other people finding you attractive. You're way more interested in the second question than you are the first question, but no amount of polling people on the internet can answer that question for you.

It's a bit like learning to juggle. You can't learn to juggle just by reading books and imagining how balls get thrown and how fast they fall. To learn how to do it, you've got to throw some balls up in the air. You've got to figure out how your body and brain deal with throwing and catching, and then you've got to learn how to control it. To begin with, you're going to drop a lot of balls, but that's not the worst thing in the world that can happen.

Comment author: necate 14 May 2015 02:49:11PM 2 points [-]

That is indeed another problem I did not consider. It definitely decreases the value of knowing how many people others have a crush on in general. But still, the fact how many people have a crush on me in particular should be somewhat correlated to how many people they have a crush on in general. Since, as you pointed out, it is impossible for me to get the specific percentage for myself I’ll have to go with the general one.

Concerning your juggling comparison. Maybe I did not express myself clearly. If you want to find out if someone likes you, then of course the most important thing is interacting with him/her. It is way more important than knowing the prior. I do not expect that after finding out the prior and reading about flirting signals I will be able to skip Interacting with people. I believe it will help me interpreting the interaction with people the right way.

The only way I can see the prior being irrelevant is when flirting signals are 100% perfect filters. Let us say I am flirting with someone and giving a signal and getting a response. If that person gives that response 100% to a person she has a crush on and 0% to a person she does not have a crush on than the prior would indeed be irrelevant. If however this person gives the response with 100% Chance to a Person he/she has a crush on, but also with 50% chance to a person he/she just likes as a friend, then this signal will only help me differentiate between the states "friend" and "love intrest" with 50% probability. And now it becomes relevant on how many of his/her friends this person has a crush.

Let us say the person has 10 friends and has a crush on 5 of them. Than on average he/she would give 5 correct positive signals 2.5 false positive signals and 2.5 correct negative signals. So if I get a positive signal, than that means that with a probability of 2/3 that person would have a crush on me and with a probability of 1/3 he/she would not.

If that same person would only have a crush on one of his/her 10 friends than on average he/she would give 1 correct positive signal 4,5 false positive signals and 4,5 correct negative signals. So if I would get a positive signal in this case than with a probability of 18% (1/4.5+1)) that person would have a crush on me and with probability of 82% (4.5/4.5+1)) he or she would not have a crush on me.

So in the first case I am reasonably sure that the other person has a crush on me and can proceed giving more obvious hints, while in the second case I should still gather more information before moving on to the next stage.

I am of course aware that in real live I will not know the exact probability for how often a certain signal is given to a friend or a love interest, but I can still get acceptable estimates for that.

Is there anything wrong with this reasoning or do you just thing that flirting signals are usually 100% perfect filters?

Comment author: sixes_and_sevens 14 May 2015 12:47:42PM 12 points [-]

This post may get downvoted, as I suspect it's of low value and low interest to a lot of readers. You shouldn't take this personally.

For what it's worth, I admire your approach, though it's based on incorrect assumptions. Trying to calculate whether someone is attracted to you will not end well. Researching psychology for romantic reasons will probably also not end well.

People solve this problem by making bigger and bigger signals at each other, until either one side stops making the bigger signals or until the signals are so big you can't ignore them, (also known as "flirting"). If this sounds hard and unreliable, that's because it is. It takes a lot of practice to get good at this. You would be best advised to practice talking to people while trying to figure out how they feel about the conversation instead of carrying out this sort of research.

Comment author: necate 14 May 2015 01:34:31PM 1 point [-]

I have no problem with getting downvoted, If my aproach seems stupid to other people it is good for me to know that. I am also aware that you cannot find exact percentages when it comes to what other peole think about you. Hovever as far as I know in general it is important to know the prior of an event if you want to get a feeling about how probable it is. Or am I wrong here?

I see that this does not fully apply to flirting because there you have signals that are very obvious and cant be misinterpreted. Hovever you also have ambigious signals. It happens that people flirt and one person gets the wrong impression about what the other meant.

If lets say every second person had a crush on you and you flirt with them, then you would not need to do much to be reasonably sure that you understood them right. However if lets say only one in a million people had a crush on you it would be much more likely that signals you got where actually caused by something else. And if this is true the prior would be relevant.

The second part of my post about how signals raise the probability might be a stupid, but I still don't see why knowing a rough figure for the prior chance of someone having a crush on you would not be helpfull. Can anyone please explain to me where the error in my thinking is?

Comment author: necate 13 May 2015 07:54:13PM 2 points [-]

I started a HPMOR follow up fanfic called Ginny Weasley and the Methods of Rationality and finished the first chapter. I am not a native english speaker and have still decidet to write in english. While i needet quite a lot of help with mistake correction I was told by someone that my writing was better than what he had seen from some native speakers.

View more: Prev

Powered by Reddit Powered by Reddit