|
| Thread: Brain Mapping |  This thread is pages long: 1 2 3 · «PREV / NEXT» |
|
Kareeah_Indaga
Responsible
Undefeatable Hero
|
posted February 04, 2010 05:34 AM |
|
|
|
Mytical
Responsible
Undefeatable Hero
Chaos seeking Harmony
|
|
posted February 04, 2010 05:43 AM |
|
|
A long time ago in a galaxy far away, me and somebody were discussing something that led to something I like to call 'thought police'. How somebody can be 'punished' for thinking something. No doing something, just thinking about it. For me, that would be beyond just 'bad'. We've ALL had stray thoughts we are not proud of.
Now I don't know how that this technology will lead to that, or even if it could. The thought is a scary one to me however. Oh you had a dirty thought about that girl you had no way of knowing is underage..your going to jail for pedophilia. The social, legal, and personal (for everybody) ramifications are very very scary.
____________
Message received.
|
|
DagothGares
Responsible
Undefeatable Hero
No gods or kings
|
|
posted February 04, 2010 07:44 AM |
|
|
Quote:
Quote:
A muslim isn't any more likely to be a terrorist than a budhist monk, a christian, a atheist, Dagoth, a communist and what-not.
Dagoth is a terrorist!! 
A poetic terrorist!
____________
If you have any more questions, go to Dagoth Cares.
|
|
Fauch
Responsible
Undefeatable Hero
|
|
posted February 04, 2010 05:27 PM |
|
|
you fill your bombs with flowers?
I never have bad thoughts, it's ok. Can I go to jail, because someone who is too good seems suspicious?
|
|
Mytical
Responsible
Undefeatable Hero
Chaos seeking Harmony
|
|
posted February 06, 2010 07:53 AM |
|
Edited by Mytical at 08:11, 06 Feb 2010.
|
As for lie detecting, the theory is since lieing uses more imagination that the brain needs more bloodflow when lieing then when telling the truth. So, theoretically it would indeed make the 'perfect' lie detector. The only pitfall would be one similar in the traditional lie detector..if the person was convinced they were telling the truth even when they were not.
Edit: I have just learned that I had two things mixed up. Brain Maps, and Brain Fingerprinting. Which Brain Fingerprinting is the actual subject that I am more interested in. So this discussion will not help my term paper at all, but is an interesting discussion anyhow I feel.
____________
Message received.
|
|
TheDeath
Responsible
Undefeatable Hero
with serious business
|
|
posted February 07, 2010 10:04 PM |
|
Edited by TheDeath at 22:05, 07 Feb 2010.
|
Quote:
Right Death it's called "Racial Profiling" and it's amongst the most stupid and ignorant things a educated person can do.
A muslim isn't any more likely to be a terrorist than a budhist monk, a christian, a atheist, Dagoth, a communist and what-not. The problem with your logic is that humans aren't points in a statistic that draw some bigger curve for you. They're actual humans. There is no room for acting on hunches and suspicions.
Well you're good at words, not good at facts though.
How hard can it be to count all known terrorist bombings/hijackings/whatever and assign percentages based on the simple result of this? It's called statistics.
Let me ask you something. Go in Iraq. Which one of the two soldiers are more likely (read: count, hypothetically of course) to kill a terrorist there: a north korean or american?
Besides the fact that the sheer number of american soldiers deployed there makes them more likely than any other soldier (nationality) to kill a terrorist... and of course north koreans aren't even deployed there against 'terrorists'.
But for more realistic question would be between an american and let's say, a french (which are much fewer deployed).
Both are humans right?
I don't remember asking that though. You might need to read up on statistical facts better.
____________
The above post is subject to SIRIOUSness.
No jokes were harmed during the making of this signature.
|
|
ohforfsake
Promising
Legendary Hero
Initiate
|
|
posted February 07, 2010 10:43 PM |
|
|
Are you suggesting a measurement of the amount of muslims, then a measurement of the amount of muslims that turns out to be terrorists and then finally comparing these results with results of other religions (i.e. comparing the amount of buddists that turn out to be terrorists, the amount of dagoth gares, etc.)?
If you suggest to validate suspicion based on such a measurement I'd say that you've taken correlation to imply causation.
If I should have a guess, I'd guess it's completely independent of your religion if you end up as a terrorists or not, or at least to such a large degree that other environmental factors are much more important.
My reasoning behind my guess is now, that a larger percentage of muslims than people of other religions (if you want to seperate this on religion, which I see no point in doing) is likely subject to methods of "brain washing" to make them into potential terrorists. However this is not because of them being muslims, it is because of other environmental effects, i.e. the "brain washing". Therefore it might make much more sense to make statistics over regions of areas, because it seems more likely that the organizations behind terror would be some what stationary in their area of operation, than that they are focusing on people of a specific religion.
All in all, if the people behind the various terrorist organizations had lived in places where other religions would be the common religion among people, and if these peoples power of persuation is independent of the religion, I'd find it very likely that we'd actually see many buddhists terrorists or similar.
Of course this is just guessing, but it'd explain the correlation, as these people live in the area where the terrorists lives and it is also the area where said religion is most common.
I used " " around the term brain washing, as I'm not certain we have a common understanding of that word.
Edit: Also making a statistic as you suggest, does not tell anything about the spread (or uncertainty of the mean, if you want), which is a very important factor, I'll give an example of why to those who are interested:
Example:
Let's say I measure the "average" height of all the people in the world. Simply by measuring all heights, sum it and divide with the amount of people.
Now I get it as 1 meter and 70 centimeter, but is that my expected height then?
Well actually there's a rather specific spread of height over the entire globe, we so there's a rather big uncertanity of this mean value depending on where you are born.
If you are born in the northern area of Europe, I believe 1 meter and 80 centimeter is much more likely as the average height, so for a dane it is more reasonable to expect 1 meter and 80 than 1 meter and 70.
Likewise if measurements of likelyhood of becoming a terrorists dependent of the religion also had taken account of the area where you are born, I am certain that you'd see a very high spread, which indicates it has not very much about if you are born in a muslim familiy, or something else, but it has to do with were the people who train terrorists operate.
Now I hope I did not make any blunders in this, if so please notice me of it, and thanks for the debate.
____________
Living time backwards
|
|
TheDeath
Responsible
Undefeatable Hero
with serious business
|
|
posted February 07, 2010 10:55 PM |
|
|
Good job putting words into my mouth.
Simple example to show you what I mean:
100 airplane bombings total
75 by muslims
25 by others (doesn't matter for this example)
Assuming you use a non-deviating rule (which is kinda true on a small time period over average), you'd expect after the next year in total to be 200 bombings, 150 by muslims and 50 by others. This is a simple example of doubling the numbers.
Now that makes +75 muslim bombings and +25 by others.
You see a muslim in an airport among "the others". As a person with the above statistics in mind, which one would you suspect more? In other words, where would you dedicate more RESOURCES to deal with the bombings problem, towards muslims or towards everybody? (which means less efficiency obviously, not focused on a less number of people).
If you suspect wrongly 30 muslims, for instance, you still achieved a good -45 bombings! Quite an achievement I must say. Why are people so concerned with 100% probabilities or "all or nothing" approaches? That's impossible, realistically speaking.
Please repeat this until you get it: it's not PROOF, it's not "guilty until proven innocent", it's SUSPICION and allocation of resources based on statistics. It lets you be more efficient, OVERALL.
____________
The above post is subject to SIRIOUSness.
No jokes were harmed during the making of this signature.
|
|
ohforfsake
Promising
Legendary Hero
Initiate
|
|
posted February 07, 2010 11:09 PM |
|
|
My problem with your method is that you do not take into consideration the uncertainity of the mean value you get, i.e. the variation, or spread if you want.
Let's just for the amusement assume we do as you suggest, and often we inspect the wrong persons, but in average we get the numbers you suggest.
Now I claim that what we do is inefficient, because to me the logic behind the theory that assumes muslim background and terrorists are correlated does not seem logic to me.
However the measurements speak for themselves, we see a clear correlation, though maybe we only see this correlation, because the causations arrives from what I guessed in my previous post.
I do not intend to waste anyones time, but I'll just shortly write my guess.
I guess that the correlation is due to many muslims live in the area where terrorists operates and therefore it is more likely for a person of muslim background to become a terrorist.
Therefore, to increase efficiency I'd want us to, when we measure how many of the muslims who were terrorists, to also measure what area they come from.
If my guess is correct, we'd see another correlation, and through that we can now choose to focus our investigation even more precise thereby making it even more efficient.
That is, if I understood your example in your latest post correct.
Edit:
Just as an example of variation: All muslims from Europe that we measured, amount who showed to be terorists, n.
All muslims from middle east that we measured, amount who shoed to be terrorists, N.
N>>n, a clear variation independent of belief, but of area in stead of.
Now I'll go so far and say that all muslims from various regions of varoius countries will show much larger amount of terrorists than of other.
That means in stead of, maybe 45% (or 3x the amount of people, if we find 15% for other religions) we might find 90%, which now is double as many. However the real problem is that in the real world it is very hard to hold all variables steady while measuring only for a single variable, that means you can find all sorts of correlations without there being any kind of causation.
____________
Living time backwards
|
|
TheDeath
Responsible
Undefeatable Hero
with serious business
|
|
posted February 07, 2010 11:11 PM |
|
|
It's pure statistics. Correlation and cause are irrelevant. Only numbers speak for themselves. Explanation is irrelevant.
____________
The above post is subject to SIRIOUSness.
No jokes were harmed during the making of this signature.
|
|
ohforfsake
Promising
Legendary Hero
Initiate
|
|
posted February 07, 2010 11:14 PM |
|
|
But explanation helps us determine measurements for finding even better correlation, correct?
____________
Living time backwards
|
|
TheDeath
Responsible
Undefeatable Hero
with serious business
|
|
posted February 07, 2010 11:19 PM |
|
|
No doubt a correct explanation may provide, at worst, an insight, and at best, drastically increase the overall efficiency and decrease "errors"... but I was talking about pure statistics that a computer could do.
____________
The above post is subject to SIRIOUSness.
No jokes were harmed during the making of this signature.
|
|
Fauch
Responsible
Undefeatable Hero
|
|
posted February 08, 2010 03:11 AM |
|
|
Quote:
100 airplane bombings total
75 by muslims
25 by others (doesn't matter for this example)
Assuming you use a non-deviating rule (which is kinda true on a small time period over average), you'd expect after the next year in total to be 200 bombings, 150 by muslims and 50 by others. This is a simple example of doubling the numbers.
Now that makes +75 muslim bombings and +25 by others.
You see a muslim in an airport among "the others". As a person with the above statistics in mind, which one would you suspect more? In other words, where would you dedicate more RESOURCES to deal with the bombings problem, towards muslims or towards everybody? (which means less efficiency obviously, not focused on a less number of people).
ok. let's say there were 5000 people in those planes. 4000 muslims and 1000 other people.
75 of 4000 muslims are terrorists : less than 2%
25 of 1000 other people are terrorists : 2.5%
|
|
ohforfsake
Promising
Legendary Hero
Initiate
|
|
posted February 08, 2010 11:45 AM |
|
|
For the pure statistics a computer could do, would it not be for the best to take advantage of the computers calculation power and let it go through statistics for all variable possible (at least what we can imagine)?
If one could program a self learning mechanism in such way, that the computer would, for a given correlation set a likelyhood on causation, this way the computer should be able to optimize its own procedure automatic.
Let's as an example say N people shows to be terrorists out of M people.
Of these N people, an amount, A, comes from a specific area, an amount, B, is at a certain age group, etc.
The computer will then, based on the measurements be able to place likelyness and thereby decide how important the different variables are. This should result in more and more accurate detecting, thereby actually finding the cause in the end.
Now I do not know, though I do naively assume you can measure everything, that there might be a set of circumstances of which it is impossible to find the governing law behind a phenomena through experiment.
If that is the case however, I might be so bold and claim that such a law is impossible to find, though I hope not of course.
____________
Living time backwards
|
|
Mytical
Responsible
Undefeatable Hero
Chaos seeking Harmony
|
|
posted February 08, 2010 12:23 PM |
|
|
Well..lucky for me statistics can fail. Where I grew up according to a statistical study they did (during the time I was growing up), I would have been 78% likely to be married by age 16, 83% likely to be pregnant (which would lead to a lot of the marriages...) and a 80 (ish) chance I would never rise above almost complete poverty.
____________
Message received.
|
|
ohforfsake
Promising
Legendary Hero
Initiate
|
|
posted February 08, 2010 12:43 PM |
|
|
Statistics can be wrong (most often interpreted wrong when medias tries to bring sensational journalistic, in reality selling pornography based on emotions), but statistics does also never say with absolute certain this will happen, or this won't happen.
According to statistics the law of gravity at the surface of the Earth is very very very very likely to hold true when considering if a rock will fall to the ground or not. It does not tell however that it is 100% certain.
It is certain enough for us to grant it as knowledge though, that is knowledge is independent if you are right or wrong, it is all about how well justified it is.
So if there's 80% chance of something happening according to statistics, and you look back to the past and check if it really were like this, you'd either find that
a) What you'd been told did not match reality to the uncertainity of the statistic observation, i.e. whatever had been measured and informed to you, had been interpreted wrongly.
b) What you'd been told does in fact match reality, so you'll see to the degree of uncertainity of the statistic observation, that a sufficient large part of the population have been triggered to said events, whereby 80% chance is an accurate measure. You were simply in the 20% group it did not happen to.
Assuming something is impossible because it is unlikely is one of the worst mistakes you can ever do, but act upon it as not important is however what is to be expected.
I have often heard it as an argument for God that due to life being so unlikely that God must have decided it.
I don't know how unlikely it is for life to start, but I'll come with another example, that shows that no matter how unlikely it is, as long as it is possible, it cannot said to be impossible.
Take 6 different coloured dices (those with 1, 2, 3..., 6 on their sides), now we assume these dices are made so good that the chance of each side showing, that is each number comming up, is equal, 1/6.
After you roll all 6 dices you will see 6 numbers, some alike, some not, tell me, what is the likelyness of exactly this combination?
Well if every single digit only happens 1/6 times then this exact combination happens 1/6 * 1/6 * ... * 1/6 or to say 1/6^6 times.
That is extremely unlikely, it is only excpetected to happens 0,002143% of the times in average.
Because all in all there is 46656 possible outcomes (6 outcomes for each, 6 times, that is 6*6*6*6*6*6 = 6^6). So every outcome is extremely unlikely, but one outcome must happen, so no matter what you get it is something you should not expect to get again in a long time (depending of how often you throw of course).
Now the same goes with life, as long as the chance is not 0% (which it couldn't be as we are here), then it is not impossible, thereby it can happen.
Had it not happened, we'd not be here to question why did life as we know it not happen, and say "oh because it is so unlikely".
Maybe a different type of life, unknown to us, would have evolved a very different place, asking the exact same question.
All in all, the point is just that don't assume that unlikely and impossible is the same, however reacting properly (i.e. prioritating lower on what is unlikely, unless the consequences are huge of course) is the smart thing to do, in my opinion.
____________
Living time backwards
|
|
TheDeath
Responsible
Undefeatable Hero
with serious business
|
|
posted February 08, 2010 10:38 PM |
|
|
Quote:
ok. let's say there were 5000 people in those planes. 4000 muslims and 1000 other people.
75 of 4000 muslims are terrorists : less than 2%
25 of 1000 other people are terrorists : 2.5%
Yes but the "others" are not a single group, you can't focus on them, because by "others" you mean "everyone else", and 'everyone' is a word that signals quite a huge list. It needs to be avoided.
Quote:
For the pure statistics a computer could do, would it not be for the best to take advantage of the computers calculation power and let it go through statistics for all variable possible (at least what we can imagine)?
What?
Quote:
Statistics can be wrong
Only if you count wrong.
Statistics are always "correct" if you know how to count. What do you mean by 'wrong'? Statistics don't say what you should do, they only provide you with information... it's your decision after all to make a choice based on that.
A reasonable person, of course, prefers to 'weight' in his choices according to statistics (and his experience probably) than just going to a blind "check-everybody-equally" which is inefficient to say the least.
I don't know why you want to make this more complicated because it's really simple.
One example where statistics are "of no use" to you, is if you can magically know who is a terrorist, for instance. That doesn't mean the statistics are wrong. If there have been 50 muslim bombings and statistics say there are 50 muslim bombings, does that mean it's wrong? Why?
____________
The above post is subject to SIRIOUSness.
No jokes were harmed during the making of this signature.
|
|
ohforfsake
Promising
Legendary Hero
Initiate
|
|
posted February 09, 2010 11:25 AM |
|
|
Quote:
What?
What what? I'm talking about optimizing in the section you reply to, if that is what you mean.
Quote:
Quote:
Statistics can be wrong
Only if you count wrong.
Okay I'll rephrase if you did not get the point of my post.
Conclusions of a most likely scenario, based on statistics analysis, does not always happen, eventhough it is what one expects.
Most often, when the events in question acts in a different manner than expected, it is due to conclusion based on bad interpretation of the numbers, which medias often seem to do, eventhough I can't come with any examples right here, right now.
Though one must remember, the link between the science community and the public often is the media, however the media is sadly not always (rather seldom actually) the neutral information provider, no they want to sell newspapers, so they'll make it look better or worse than it is, that way they'll overinterpretate on the information, most often statistical, and a huge percentage of the public will then believe it to be true.
Of course it is not, but it is not the medias who gets the blame, it's the scientific community.
The second part basicly said that eventhough the statistic analysis is correct, whereby future prediction actually holds true for x% of the population, then just because it does not happen to you, it does not mean you weren't in the risk zone.
____________
Living time backwards
|
|
Mytical
Responsible
Undefeatable Hero
Chaos seeking Harmony
|
|
posted February 09, 2010 11:29 AM |
|
Edited by Mytical at 11:42, 09 Feb 2010.
|
I guess I should rephrase also. Statistics can be misleading. People are individuals and while it can show the 'average' of what is going to happen..there is too many factors to successfully gage what will or wont be true for each person.
____________
Message received.
|
|
ohforfsake
Promising
Legendary Hero
Initiate
|
|
posted February 09, 2010 11:37 AM |
|
|
I completely agree, that is why averages in themselves are more or less useless, you really need to focus on uncertanity and the logical limits (assumptions) you set on your analysis.
____________
Living time backwards
|
| |
|
|
