5 ms·

The author makes some really good points about the quackery in EP. But at no point are the merits of some of these theories even mentioned. Take for example th

The author makes some really good points about the quackery in EP. But at no point are the merits of some of these theories even mentioned. Take for example the proposition that women are attracted to older men. If it was found that all current human cultures and other mammals exhibited this behavior you could begin to think that this behaviour is grounded in biology. > Evolutionary psychology does draw on empirical data and laboratory studies, and those data are falsifiable. But the adaptationist explanations that evolutionary psychologists offer are not. If an explanation fit all current observations and predicted future experiments why is it not correct? I'm having trouble expressing what I want to say, and some cursory googling for sources was met with paywalls but I wasn't a fan of some of the author's arguments. Even if EP somehow explained that some very nasty behaviours were actually evolutionary adaptations that would not make the behaviours correct, and it would not make the explanation incorrect. Human behaviour is complex and influenced by factors both societal and biological. While current pop sci has swung the needle way too far into the "biological" side of the argument I feel like the author has swung a bit too far into the "societal" side.

api 12y ago
> If an explanation fit all current observations and predicted future experiments why is it not correct? http://www.tylervigen.com http://www.tylervigen.com
- PavleMiha 12y ago
  I don't think correlation/causation confusions apply here. It's a question of whether these evolutionary explanations for actual observed behaviours are falsifiable. What is correlated but not causated (is that even a word?) in this situation?
kijin 12y ago
> If an explanation fit all current observations and predicted future experiments why is it not correct? In order for anything to count as a "future experiment" with respect to something that takes as long as evolution does, the predicted phenomena would have to occur at least thousands of years from now, and you would have to control for all the socio-economic-cultural-political differences between now and then. Which would begin to sound like an impossible task, even you embarked on a millennia-long research project, because humanity is quickly approaching (and may already have gone beyond) the point where technology overtakes evolution. Evolution of the regular type at least has billions of years' worth of fossil records to compensate for its lack of predictions for the future. The only thing EP has going for it, on the other hand, is a bunch of surveys, small-N experiments, and beloved-by-Cosmopolitan stereotypes about the contemporary courting behavior of affluent, predominantly white college students. That's like analyzing a tub of ice cream and then making bold claims about the fate of the Antarctic ice cap.
- JoeAltmaier 12y ago
  With fast-mutating creatures your thousands of generations might only be months...
  - kijin 12y ago
    Unfortunately, fast-mutating creatures tend not to have much of a psychology.
- tim333 12y ago
  >The only thing EP has going for it, on the other hand, is a bunch of surveys, small-N experiments, and beloved-by-Cosmopolitan stereotypes about the contemporary courting behavior of affluent, predominantly white college students. There is other evidence also. EP for example would suggest males would tend to massacre their neighbours and grab their females as this would be a genetic benefit. To study that experimentally you could look at the historic record of conquests and the genetic results, study uncontacted human societies, the stuff ISIS is doing now, chimp studies such as those recently in the news (http://www.nytimes.com/2014/09/18/science/lethal-violence-in-chimps-occurs-naturally-study-suggests.html?_r=0 http://www.nytimes.com/2014/09/18/science/lethal-violence-in...) and such like. You could also run MRIs on people brains to try to understand the mechanisms and so on. It's not college surveys and just so stories. Indeed the fact that ISIS massacring the Yazidi's and rounding up women fits the evolutionary chimp style raiding hypothesis better than the loving God hypothesis is significant and might discourage the guys from doing if they realised that. Then maybe the massacres could stop through scientific enlightenment rather than us reluctantly having to go bomb them. Again it's more than Cosmo type stuff.
  - kijin 12y ago
    It's very difficult to avoid sampling bias and confirmation bias when you're looking for historical and contemporary events that seem to fit your curve. What about all the men throughout history who didn't massacre their neighbors? How do you control for massive confounding factors such as religious brainwashing, economic circumstances, poor education, and individual idiosyncrasies?

api 12y ago

> If an explanation fit all current observations and predicted future experiments why is it not correct? http://www.tylervigen.com http://www.tylervigen.com

PavleMiha 12y ago
I don't think correlation/causation confusions apply here. It's a question of whether these evolutionary explanations for actual observed behaviours are falsifiable. What is correlated but not causated (is that even a word?) in this situation?

kijin 12y ago

> If an explanation fit all current observations and predicted future experiments why is it not correct? In order for anything to count as a "future experiment" with respect to something that takes as long as evolution does, the predicted phenomena would have to occur at least thousands of years from now, and you would have to control for all the socio-economic-cultural-political differences between now and then. Which would begin to sound like an impossible task, even you embarked on a millennia-long research project, because humanity is quickly approaching (and may already have gone beyond) the point where technology overtakes evolution. Evolution of the regular type at least has billions of years' worth of fossil records to compensate for its lack of predictions for the future. The only thing EP has going for it, on the other hand, is a bunch of surveys, small-N experiments, and beloved-by-Cosmopolitan stereotypes about the contemporary courting behavior of affluent, predominantly white college students. That's like analyzing a tub of ice cream and then making bold claims about the fate of the Antarctic ice cap.

JoeAltmaier 12y ago
With fast-mutating creatures your thousands of generations might only be months...
- kijin 12y ago
  Unfortunately, fast-mutating creatures tend not to have much of a psychology.
tim333 12y ago
>The only thing EP has going for it, on the other hand, is a bunch of surveys, small-N experiments, and beloved-by-Cosmopolitan stereotypes about the contemporary courting behavior of affluent, predominantly white college students. There is other evidence also. EP for example would suggest males would tend to massacre their neighbours and grab their females as this would be a genetic benefit. To study that experimentally you could look at the historic record of conquests and the genetic results, study uncontacted human societies, the stuff ISIS is doing now, chimp studies such as those recently in the news (http://www.nytimes.com/2014/09/18/science/lethal-violence-in-chimps-occurs-naturally-study-suggests.html?_r=0 http://www.nytimes.com/2014/09/18/science/lethal-violence-in...) and such like. You could also run MRIs on people brains to try to understand the mechanisms and so on. It's not college surveys and just so stories. Indeed the fact that ISIS massacring the Yazidi's and rounding up women fits the evolutionary chimp style raiding hypothesis better than the loving God hypothesis is significant and might discourage the guys from doing if they realised that. Then maybe the massacres could stop through scientific enlightenment rather than us reluctantly having to go bomb them. Again it's more than Cosmo type stuff.
- kijin 12y ago
  It's very difficult to avoid sampling bias and confirmation bias when you're looking for historical and contemporary events that seem to fit your curve. What about all the men throughout history who didn't massacre their neighbors? How do you control for massive confounding factors such as religious brainwashing, economic circumstances, poor education, and individual idiosyncrasies?