No account? Create an account
Math v/s Emotion ... but not really.
05:45pm 15/09/2004

Here's a little theory I've been working on. Basically, it's my best stab so far at the expressing logic and decision making mathematically, and how emotion has historically gummed up the works.
Basically, when people try to express the decision making process mathematically, it doesn't work out well. I think I've found the flaw that keeps getting in the way. Basically, they're not treating emotions properly in the representation. Emotions have been shown to be of great importance in decision making, even for those decisions not generally considered emotional in nature. (There's an article in Discover magazine attesting to this) Usually, people try to symbolize emotions as a variable. I think that they shouldn't be expressed that way. Here's how I think it ought to go for the decision making process.

For example, deciding whether or not to cheat on an exam:

The known quantities/factors/circumstances would be expressed as the integers/constants of the equation. (numbers like 1, 57, 42 or constants like lambda, pi, you get the idea.)

1. Bob knows that he will fail the test on his own.
2. Bob knows that if he copies answers from Jill, who sits next to him in class, he'll pass, because she the smartest student in the class. (Whether or not this is actually the case is immaterial. He "knows" it to be true, which is all that matters in regards to his deciding).
3. Bob does not want to get busted.
4. Bob wants to pass the class.

The unknown factors, probabilities, and uncertainties would be expressed as variables.

1. The probability that Jill would notice he's copying.
2. The probability that the professor would.
3. The probability that anyone else would.
4. The probability that the similarity of answers would be noticed.
5. Basically, the odds and means of his getting busted.

C. The interactions of the two above categories would be the actions of the equation, aka the actually addition, subtraction, multiplication and division that goes on between integers and/or variables.

1. Probability of being observed cheating times the probability of the observer making this known to or being the professor would make one section of the equation.

2. Bob's desire to pass opposing his desire to risk certain failure would be another.

This is the point at which things break down in the metaphor, because most people would try to factor in Bob's emotions and feelings as variables because of their analog, precision-resistant nature. That's the mistake. Emotions are not "things" or "factors" themselves, actually. They're contextual in nature. When you feel good, you're more likely to do good, and vice-versa, to grossly over-simplify it.
I think that the emotions would be more accurately represented as the order of operations, known to math students as "PEMDAS" (Please Excuse My Deficient Algebra Skills).

The parenthesis and the brackets used in math are what determines the order in which one handles the various sections. The emotions and feelings of Bob determine his priorities in deciding whether or not to cheat. The important thing here is that his feelings aren't operators in the equation so much as they set up the order of operations, the mental milieu in which the decision is made. As any math student knows, screwing up the order of operations will knock the answer into a cocked hat. Similarly, Bob's emotional state has the ability to do the same to the viability of his decision. Note- This is all independent of morals or ethics, except as they apply internally to Bob.

His fear of getting busted would parenthesize the probability of her noticing and the probability of her reporting it. His fear of failure would put parenthesis around his desire to cheat and the choice of who to cheat from.

Basically, emotions and feelings aren't factors themselves in our decisions as much as we think. They are the context in which we weight the factors, the priorities we assign the individual factors and the groups of factors. They are the parenthesis, not the variables or integers. That's why we've never been able to reliably quantify them. Not because they are illogical, but because they are a fundamental part of logic itself. Because they are an aspect/foundation/milieu/context of the decision making process (logic, understanding, comprehension) itself, trying to make sense of them is like trying to study the microscopic aspects of the only microscope in the world.

[EDIT] -C'mon, people. Someone has to have some opinion or something on this. Even if you think it's the dumbest thing since helmets on skydivers, lemme know. I'm trying to figure out the world here, and I could use all the help I can get.
 mood: accomplishedmusic: Concrete Blonde - Mexican Moon

(no subject)
 _corvidae_

04:14pm 15/09/2004 (UTC)

I think your going to have to add in factors for morals and ethics. Possibly as a divisor of the emotional factor. The largest problem would be putting values to all of the variables. You'd need a significant history of the people involved and their past choices.

(no subject)
 kires

05:29pm 15/09/2004 (UTC)

morals and ethics fall into the same category as emotions and feelings. in the decision making process, they have the same effect if it's the morals and ethics of the person involved, if we're talking about the morals and ethics of others, as understood by the person, then they're one of the other two, usually variables. misunderstanding them and the errors resulting from not know either yourself or others has the same magnitude of effect on the result for both mathematics and decisions.

(no subject)
 taiste

04:39pm 15/09/2004 (UTC)

I would love to respond, but I do not speak math. The upshot, I'm guessing, is that you're saying people generally think emotions are pieces of the decision making puzzle, but in fact emotions are the environment of the puzzle-solver?

If that is what you're saying, I would tend to agree. I think emotions can affect the outcome, but are not part of the decision at hand. They more affect the way the decision/consequences/possible outcomes are perceived.

(no subject)
 kires

01:41am 24/09/2004 (UTC)

yep. exactly.

(no subject)
 phinnia

05:18pm 15/09/2004 (UTC)

I never thought of it that way ... hm. But it makes a lot of sense.
What's that Discover article, by the way? I'd be interested in seeing how that fits into this, just as a background kind of thing.
(And I'm here via Shadesong. It's her fault. :-D)

(no subject)
 thryn

07:18pm 15/09/2004 (UTC)

I think it makes sense, but I'm not quite mathy enough to apply it to other aspects of life, nor do I knoww that I agree with your take on the influence of emotion.

And *poke*

You know you want to make a whirlwind trip to Hotlanta.

(no subject)

07:24pm 15/09/2004 (UTC)

Dude, I just-now poked him about it in a comment in my post that I made pointing to this post.

C'mon, . You know you wanna. You have a pretty girl here asking you. She has leather pats.

You have me, too, but I shrank out of the pats.

I swear this all makes sense in context.

(no subject)
 thryn

07:49pm 15/09/2004 (UTC)

*laughs*

Yes, yes, in context, it does indeed.

(no subject)

06:15am 16/09/2004 (UTC)

Lo, I am patsless!

(no subject)
 colinzen

08:15pm 15/09/2004 (UTC)

I don'e get it. Do you have cliff notes or something?

(no subject)
 spc476

11:09pm 15/09/2004 (UTC)

I think it's less like a mathematical equation (E=mc2) and more of a neural network (pun not intended). A neural network (at least in Computer Science) is a parallel set of inputs pumped through a meshwork of values that results in a parallel series of outputs (or a single output, but generally, less outputs than inputs) (figure 1). It's still mathematical in nature, but less a single equation and more of a multiple filtering type effect—the input values (say, a value between 0 and 1) percolate through each layer, being multiplied by other inputs and certain weight values to an output value (say, another value between 0 and 1) to represent an outcome (in this case, whether Bob decides to cheat).

Weight values can either be constant or changed during an evaluation process and often times it's difficult to reason out why certain values end up being used. The canonical example of this is a neural network designed by the DoD to recognize tanks. They fed pictures of friendly tanks into a neural network and “trained” it to recognize the friendly tanks. Then a series of pictures of enemy tanks. When the neural network could “recognize” the test pictures, they fed it a new batch of pictures where it did about as good as random guessing. It took a while before the researchers realized that the neural network was trained to spot clouds, not tanks (the friendly tanks were photographed on a clear day, the enemy tanks on a cloudy day).

 picword: hmmm

(no subject)
 pbristow

05:48am 16/09/2004 (UTC)

Yep, that's certainly the working model I use in understanding myself, and it hasn't proved itself wrong yet.

My detailed answer to this has taken 2 hours to write and won't fit in a comment, so I'll post it my own journal shortly, OK? See you there... I'm late for work right now!
 picword: geeky

(no subject)
 pbristow

12:06pm 16/09/2004 (UTC)

OK, I've wrotten - writted - er, *written* it, and it's here:

http://www.livejournal.com/users/pbristow/112436.html

Comment welcome; Fights, flames and nit-picking less welcome. =:o}
 picword: XI-sing

(no subject)
 pbristow

12:07pm 16/09/2004 (UTC)

OK, I've wrotten - writted - er, *written* it, and it's here:

http://www.livejournal.com/users/pbristow/112436.html

Comment welcome; Fights, flames and nit-picking less welcome. =:o}

(no subject)
 dream_labyrinth

01:12am 16/09/2004 (UTC)

Pointed to this by .
Even though my mathematical knowledge is really just the things I can't avoid remembering from 12 years of school, I think I got the point.
And I agree.
Emotions do affect our decisions, but not always in the same way, and not in a way that can be expressed by numbers. (Like, when you're feeling good, the probability of cheating goes down by 10%.)
Also, I agree with you saying that morals and ethics are only relevant if Bob has made them his morals, if he feels that it is wrong to cheat not because he might be busted if anybody finds out but because it is wrong in an ethical sense. (Am I making sense here?) And that would mean morals and ethics can be treated like emotions for the equation.
Interesting theory, really. Adding to memories.
 picword: My two cents

(no subject)
 mantic_beholder

01:16am 16/09/2004 (UTC)

I don't really get it.... Given the equation p(desire)-p(noticed), adding parenthesis doesn't change anything. Adding parenthesis, or otherwise messing with operators, just in general strikes me as a far less precise tool.

Plus, there's a lot of contingencies - who you cheat from is based off of the chance of getting caught, but also off of how much you'll benefit.

I think the reason that an emotional equation doesn't work out is that, in general, emotions are vague things - I've noticed that I, quite often, lie to *myself* about my emotional state. To other people, I actually often appear happier when I'm seriously depressed. So anyone trying to factor it in based on my apparent emotional state is simply working off of flawed data :)

(no subject)
 kires

01:48am 24/09/2004 (UTC)

you're still looking at the emotions as parts of the equation, which is pretty much the opposite of what i'm suggesting here. I'm offering the notion that emotions are not a part or parts of the decision making process, but rather the environment in which the decisions are made. The parenthesis describe the order of operations, the priorities of the various bits of logic, and are not part of those logical bits themselves.

knowing yourself is how well you're able to assign priorities to the various factors, i.e how much wright you give the various factors in a decision. lying to yourself is simply placing the parenthesis where you want them to be, instead of seeing where they are. and that results in an answer that isn't right, and decisions that aren't what you "really" want to do.

(no subject)
 mantic_beholder

02:50pm 24/09/2004 (UTC)

I think I'd need actual numeric/mathy examples before I could really understand this. However, the description of it as "enviroment instead of factors" is an intriguing one. Sort of like how we're used to just thinking in cartesian coordinates, where parallel lines can't intersect, but if you just change the enviroment you can get effects like that? (Although I suppose that's more changing the logic, the underlying assumptions, than it is moving a few parenthesis)

(no subject)
 theano

01:41am 16/09/2004 (UTC)

Hello! I am a random mathematician wandering by.

You might be interested in reading a bit about Nigel Bennett's "drama theory", which is a derivative of game theory. (John Nash got his Nobel in economics for a game theory insight, in case you've seen/read "A Beautiful Mind".)
Try here: http://www.gametheory.net/News/Items/092.html
(That's from a 1998 article in "New Scientist", a British magazine vaguely similar to Scientific American.)
Or perhaps here: http://www.ima.org.uk/conflict/papers/Bennett.pdf for an example of a slightly more formal treatment.

Mathematics isn't, in this case, going to be terribly useful for explaining people's decisions; I do think it is possible to mathematically model decisions which involve emotion. Once you've got your models, then you might be able to gain new insights.

I think you're more or less on the right track with what you're thinking. Putting it in the context of high-school algebra is using the wrong tool, though, and perhaps confuses your argument here. You're right that emotions aren't something that should be quantified (an object that you're manipulating, say number, variable, etc.), but rather should be part of the deductive system (the rules for manipulating, your "parentheses"). I point out that there isn't just ONE system of logic out there, or just one deductive system.

I believe that in mathematics in the past, (in classical game theory), emotions were disregarded not because they are considered unimportant or irrelevant or because mathmaticians are socially incompetent, but rather because at the time it was too hard to factor them in. Like in high school physics class: first you solve problems of moving a block along a surface, then along an incline, then you add in friction. Or you talk about the flow of water, but you disregard friction, viscosity, and so forth 'cause you don't have the tools yet to take that stuff into consideration.

Ok, that was long-winded. Sorry.

And thanks for reminding me why I decided to study logic - I'm was in the throes of dissertation writing misery!

(no subject)
 cissa

07:57am 18/09/2004 (UTC)

I know I'm late coming to this discussion, but I love the approach! It makes a lot of sense to me, and explains much. Nice insight and explanation!

(no subject)
 kires

01:43am 24/09/2004 (UTC)

than-q.