Cancel
Comment on Plural Thinking

BenRayfield Wed, Nov 3, 2010
Sonicport, your area of research and mine are very similar and we may want to work together on some long-term projects.

The http://emotiv.com  mind reading game controller (which interprets electricity on the skin of your head) does offer the raw electricity data to software, but it costs a lot for the license to connect your software to it. Using a bayesian network (as I'll explain below), combined with the kinds of software we build for musical instruments, I'm sure I could implement "plural thinking" controls. Its just a matter of fine-tuning the timing and preprocessing of the inputs and outputs to a bayesian network. I could make the Emotiv Epoc detect when you're thinking about moving left and right at the same time, or any other combination, but I could do the same thing indirectly as a psychology software that interprets how you move the mouse when controlling realtime generated music (that's what I'm going to build eventually).

If you research deep enough into what you call "plural thinking", you will discover a very important way of thinking thats missing from most people's brains, something that everyone expects is obviously there but nobody verifies it. It is best demonstrated with the following simple question:

If I flipped 2 coins, both landed, and at least 1 landed heads, then whats the chance that both landed heads?

Most people answer 1/2 or 1/4 chance, but that's the wrong fraction.

The flaw in Human psychology is best demonstrated when people eliminate 1 of the coins from their thoughts and think about the other coin separately. At least 1 coin landed heads so they think they can choose that coin (whichever it is, or either coin if both landed heads) and say its a constant heads. Its not a constant. I did not tell you which coin landed heads, and it does result in a different answer. There are 4 ways 2 coins can land, and "at least 1 landed heads" only excludes "tails tails" leaving 3 possibilities which happen equally often. Then I asked what is the chance of 1 of those 3 possibilities.

That is why most people do not understand which events are more or less probable in combination with other events. They eliminate 1 of the events when its not correct statistics.

What you call "plural thinking" is very similar to http://en.wikipedia.org/wiki/Bayesian_network  which is a kind of statistics software that always gets such questions right.

My plan is to include bayesian networks and other artificial intelligence (AI) algorithms in a peer-to-peer network of musical instruments, mouse movements, realtime generated sounds, and lots of other interfaces, and organize all those people and computers together in subtle psychology ways to form 1 much smarter mind across the internet.

Its all open-source (GNU GPL, various versions), but I could make some exceptions to the licenses for connecting to your hardware, if necessary.
The whole system is called Human AI Net and is athttp://sourceforge.net/projects/humanainet
I've not connected the music parts to it yet. The newest version of them is in this software:http://spacecollective.org/BenRayfield/6090/Artificial-Intelligence-learns-music



To most people, the lack of ability to do "plural thinking" (or "bayesian" thinking) appears to be a small problem, something to ignore, but I think its the cause of most other problems on this planet. I think the global economy is becoming unstable faster each year because people do not understand the combination of any 2 events accurately. I think government complexity increases faster each year (and is now beyond the understanding of any group of people small enough that they can effectively communicate with eachother) because the "plural thinking" (bayesian) flaw in Human psychology results in handling every case separately instead of generalizing and simplifying. 2 laws becomes 4, and 4 laws become 8... and now its completely impossible to even count how many laws in USA there are, and USA's Congress doesn't even read them anymore (the Read The Bills Act was rejected, no time to read laws before passing them). This is all leading to World War 3. Nobody understands how these global organizations work or how to fix them, but you (Sonicport) somehow figured out that it was a good idea to use "plural thinking" to improve how brains think. You didn't know what it was useful for, but you found 1 of the many solutions. Using software to control the interaction of many musical instruments through the internet to teach people "plural thinking" (the simplest kind of bayesian thinking), we can fix the psychology problems that lead to other problems... that lead to World War 3.

I'd like to work with you to use software and musical instruments. We can combine our existing projects to get it done faster. I know you started on this research path just because it was interesting or fun or for curiosity, but because of the specific things you've chosen to research, I think you are one of the few people with the state of mind that can help me avoid the extinction of Humans. All we have to do is use our research to get people to understand why the answer to the following question is 1/3, and for them to understand it at an intuition level so they get it right even if its asked in many different ways:

If I flipped 2 coins, both landed, and at least 1 landed heads, then whats the chance that both landed heads?

If you teach people "plural thinking" (as you call it), then they will gain intuition that causes them to get that question right, and from that small change, the whole world starts changing at very very slow but exponentially increasing speed. Its similar to the http://en.wikipedia.org/wiki/Butterfly_effect  except not as hard because we know how people think. We know most of them can't answer that simple question right. A society built on such a fundamental misunderstanding of the combination of any 2 events must eventually fail. Lets fix it at the core of the problem.

I can help you with the software parts. You can do more of the hardware parts. This can also make us lots of money. I've been planning for a long time to compete with the Guitar Hero video games which had over a billion dollars in profit. As open-source, we would not get as much of the profits, but the market for such musical instruments is big enough it doesn't matter. We can work together to build musical instruments and software systems that are more fun than that game, and simultaneously solve the lack of ability to do "plural thinking". We can get rich and avoid an extinction level event. You want to help?