7 ms·

How about: human behavior is a complex system exhibiting chaotic behavior and is mathematically proven to be unpredictable? And don't even get me started on th

How about: human behavior is a complex system exhibiting chaotic behavior and is mathematically proven to be unpredictable? And don't even get me started on the absurd over-reach of the conclusions of almost all social science papers. They'll study Westerners and proclaim to have discovered something about humans - ignoring the monumental variable of culture. They also project their own values on research constantly. How many times have you seen research reported as "health benefits of sex"? And how many times have you seen the same research reported as "health dangers of abstinence"? They're the exact same thing, but a matter of cultural perspective. If you view sexuality as a normative part of human life, then observing that those who have fewer orgasms develop prostate cancer is a health danger of abstinence. But if you view sexuality as an aberration, something that's not a normal part of everyday life, then you report it as a benefit of increased sexual activity - establishing that you think the current level is 'normal' rather than 'so low as to present actual biological danger'(which is the truth).

HelloMcFly 12y ago
That it is seemingly chaotic doesn't make it chaotic, and I'm interested in the mathematical proof that behavior can't be predicted. Don't confuse limited ability to predict with impossibility. We can make many confident statements about the likelihood of certain behaviors/reactions/response, in the aggregate, under given real-world circumstances. You are correct in that much research, especially research older than 10 years ago, tended to ignore culture. I know my field of study (I/O Psychology) references cultural frequently, either through overt study when possible (desirable given the publication potential) or at least remarking on the limitations of any findings. I don't read much outside I/O research though, so maybe it hasn't progressed much on the whole and the global nature of I/O work for businesses rewards more recognition of culture. It's hard though - accessing globally-relevant datasets for studies of human behavior is often impossible. That doesn't make all studies meaningless, or unscientific. And yes, you're correct about value-laden framing of results. That doesn't invalidate the science of the results, it just makes it more necessary to think critically when interpreting them rather than relying on abstracts and headlines alone.
- lutusp 12y ago
  > That it is seemingly chaotic doesn't make it chaotic, Very true. A chaotic system has specific properties that we would be very pressed to identify in human behavior. > ... and I'm interested in the mathematical proof that behavior can't be predicted. Think abut what you're saying. Proof that behavior cannot be predicted would require proof of a negative. Proof of a negative, or (the same thing) an evidentiary burden requiring evidence of absence, is a logical error. So this is a non-starter. Let's say I believe in Bigfoot, and I think others are obliged to dissuade me from by belief. The problem is that this also requires proof of a negative, and Bigfoot could be hiding under a rock somewhere on a distant planet. Bigfoot cannot be proven not to exist, so I'm secure in my evidence-free belief. This is why scientists adopt the null hypothesis precept, the idea that an idea is assumed to be false until evidence supports it -- it's a simple way to avoid logical pitfalls like believing in Bigfoot. > We can make many confident statements about the likelihood of certain behaviors/reactions/response, in the aggregate, under given real-world circumstances. Yes, true, but until those observations lead to a falsifiable theory of human behavior, it's not science. Until then, we can only describe, we can't explain. Testable explanations are the threshold of science. > And yes, you're correct about value-laden framing of results. That doesn't invalidate the science of the results ... No, of course not -- thee reason is that such results aren't science. An observation without an explanation isn't science. Here's an example -- if I sit under a clear nighttime sky, then report that I saw many tiny points of light, I've made an observation of very little scientific usefulness, and one that can't meaningfully be falsified by repeated observation. But if I claim those points of light are actually distant thermonuclear furnaces like our sun, my assertion can be tested and possibly falsified. By moving beyond observation, by offering a tentative explanation, I've crossed the threshold of science.
  - HelloMcFly 12y ago
    > Think abut what you're saying. Proof that behavior cannot be predicted would require proof of a negative. Proof of a negative, or (the same thing) an evidentiary burden requiring evidence of absence, is a logical error. So this is a non-starter. Yes, I know. That was the point of the comment in response to the parent's statement: "[human behavior] is mathematically proven to be unpredictable." But thank you for the lecture. > Yes, true, but until those observations lead to a falsifiable theory of human behavior, it's not science. There are plenty of theories. They are typically considered theories on motivation and behavior rather than behavior itself. It's a notable difference, but a sensible distinction. Many of these theories are falsifiable, and in fact have been falsified (e.g., expectations-based models, for example). But they are typically piecemeal, as the complexity of human motivation and behavior currently requires. If the key requirement to be considered "science" is for The One Great Unified Theory, then I agree that it will never be considered a science in my lifetime. You seem to think psychology is all observation with no attempt at explanation, much less construction of logical theories supporting those explanations. That is not the case in my little fiefdom (though I have no doubt it is true in some disciplines, or acceptable in some journals). Much work has gone into proposing, testing, and refining many theories on motivation and behavior (e.g., reinforcement theory, goal-setting theory, social-cognitive theory or workplace behavior, job design theory, motivated action theory, etc.). It's not perfect, and it's got a long way to go, but it's also come a long way. Yet it's hardly just observing the stars and making claims, and it's certainly shown utility in helping design systems to influence behavior. But there, I've said my piece. I've usually got the discipline to stay out this pet topic for the HN crowd.

HelloMcFly 12y ago

That it is seemingly chaotic doesn't make it chaotic, and I'm interested in the mathematical proof that behavior can't be predicted. Don't confuse limited ability to predict with impossibility. We can make many confident statements about the likelihood of certain behaviors/reactions/response, in the aggregate, under given real-world circumstances. You are correct in that much research, especially research older than 10 years ago, tended to ignore culture. I know my field of study (I/O Psychology) references cultural frequently, either through overt study when possible (desirable given the publication potential) or at least remarking on the limitations of any findings. I don't read much outside I/O research though, so maybe it hasn't progressed much on the whole and the global nature of I/O work for businesses rewards more recognition of culture. It's hard though - accessing globally-relevant datasets for studies of human behavior is often impossible. That doesn't make all studies meaningless, or unscientific. And yes, you're correct about value-laden framing of results. That doesn't invalidate the science of the results, it just makes it more necessary to think critically when interpreting them rather than relying on abstracts and headlines alone.

lutusp 12y ago
> That it is seemingly chaotic doesn't make it chaotic, Very true. A chaotic system has specific properties that we would be very pressed to identify in human behavior. > ... and I'm interested in the mathematical proof that behavior can't be predicted. Think abut what you're saying. Proof that behavior cannot be predicted would require proof of a negative. Proof of a negative, or (the same thing) an evidentiary burden requiring evidence of absence, is a logical error. So this is a non-starter. Let's say I believe in Bigfoot, and I think others are obliged to dissuade me from by belief. The problem is that this also requires proof of a negative, and Bigfoot could be hiding under a rock somewhere on a distant planet. Bigfoot cannot be proven not to exist, so I'm secure in my evidence-free belief. This is why scientists adopt the null hypothesis precept, the idea that an idea is assumed to be false until evidence supports it -- it's a simple way to avoid logical pitfalls like believing in Bigfoot. > We can make many confident statements about the likelihood of certain behaviors/reactions/response, in the aggregate, under given real-world circumstances. Yes, true, but until those observations lead to a falsifiable theory of human behavior, it's not science. Until then, we can only describe, we can't explain. Testable explanations are the threshold of science. > And yes, you're correct about value-laden framing of results. That doesn't invalidate the science of the results ... No, of course not -- thee reason is that such results aren't science. An observation without an explanation isn't science. Here's an example -- if I sit under a clear nighttime sky, then report that I saw many tiny points of light, I've made an observation of very little scientific usefulness, and one that can't meaningfully be falsified by repeated observation. But if I claim those points of light are actually distant thermonuclear furnaces like our sun, my assertion can be tested and possibly falsified. By moving beyond observation, by offering a tentative explanation, I've crossed the threshold of science.
- HelloMcFly 12y ago
  > Think abut what you're saying. Proof that behavior cannot be predicted would require proof of a negative. Proof of a negative, or (the same thing) an evidentiary burden requiring evidence of absence, is a logical error. So this is a non-starter. Yes, I know. That was the point of the comment in response to the parent's statement: "[human behavior] is mathematically proven to be unpredictable." But thank you for the lecture. > Yes, true, but until those observations lead to a falsifiable theory of human behavior, it's not science. There are plenty of theories. They are typically considered theories on motivation and behavior rather than behavior itself. It's a notable difference, but a sensible distinction. Many of these theories are falsifiable, and in fact have been falsified (e.g., expectations-based models, for example). But they are typically piecemeal, as the complexity of human motivation and behavior currently requires. If the key requirement to be considered "science" is for The One Great Unified Theory, then I agree that it will never be considered a science in my lifetime. You seem to think psychology is all observation with no attempt at explanation, much less construction of logical theories supporting those explanations. That is not the case in my little fiefdom (though I have no doubt it is true in some disciplines, or acceptable in some journals). Much work has gone into proposing, testing, and refining many theories on motivation and behavior (e.g., reinforcement theory, goal-setting theory, social-cognitive theory or workplace behavior, job design theory, motivated action theory, etc.). It's not perfect, and it's got a long way to go, but it's also come a long way. Yet it's hardly just observing the stars and making claims, and it's certainly shown utility in helping design systems to influence behavior. But there, I've said my piece. I've usually got the discipline to stay out this pet topic for the HN crowd.