Is it really fair to say there has been “no strong evolutionary or cultural force sufficient to focus us on the very narrow behavior patterns that would actually be effective”?
Clearly we’ve evolved the ability (trainable hardware) to do the kind of planning, abstract reasoning, and analysis that would help us find these optimal courses of action. Furthermore, we’ve evolved the tendency to do a fair amount of this (compared to other life forms) automatically.
This isn’t just a hardcoded ability to execute plans that bring food, shelter, and sex. If you decide you want a new pair of shoes, it’s trivial for you to mentally construct and carry out the relatively complex (again, comparing to other species) plan required for you to get them. You’ll even carry out some optimizations without too much effort (“wait, there’s a closer shoe store east of here”).
While it’s trivially true that we haven’t evolved to automatically seek the optimum path in all things (which there might be a good reason for, e.g. time-constraints on assessing and choosing paths), I think it’s fair to say evolution has given us a running start.
And the selective pressures are pretty clear: something like the Machiavellian Intelligence hypothesis is almost certainly at work, selecting those genes that are best at carrying out plans, since in the ancestral environment most plans were directly concerned with survival and breeding. Granted, no shortage of optimization was hacked on towards attaining those goals specifically (which might explain why it’s so hard to focus on your bigger goals when you’re hungry), but the fact is humans didn’t evolve to mindlessly seek out food and mates. We’re capable of pursuing other interests.
So I’d say there have been strong selective forces to help us choose effective courses of action. Not absolutely optimal courses of action, but when you consider the massive size of action-space (as you pointed out, rocks fail calculus tests by default), it’s apparent that evolution didn’t totally leave us out in the cold.
This is true and a valuable correction, however I would argue that our planning ability was evolved for very different goals in a very different environment, and while it works pretty well at “figuring out if your friend is backstabbing you” or “figuring out how to get calories”, when it comes to long-term goals in the modern environment (“how do I manipulate this laptop so as to make me millions of dollars over the next 5 years?”) it performs miserably, and all Anna’s points then apply.
Paul Graham recently made a related point, the world is getting more and more addictive, and in order to be productive we must develop more effective screening and anti-time-suck methods: http://www.paulgraham.com/addiction.html
On the plus side, there are a few great people working on this problem—Merlin Mann comes to mind.
Is it really fair to say there has been “no strong evolutionary or cultural force sufficient to focus us on the very narrow behavior patterns that would actually be effective”?
Clearly we’ve evolved the ability (trainable hardware) to do the kind of planning, abstract reasoning, and analysis that would help us find these optimal courses of action. Furthermore, we’ve evolved the tendency to do a fair amount of this (compared to other life forms) automatically.
This isn’t just a hardcoded ability to execute plans that bring food, shelter, and sex. If you decide you want a new pair of shoes, it’s trivial for you to mentally construct and carry out the relatively complex (again, comparing to other species) plan required for you to get them. You’ll even carry out some optimizations without too much effort (“wait, there’s a closer shoe store east of here”).
While it’s trivially true that we haven’t evolved to automatically seek the optimum path in all things (which there might be a good reason for, e.g. time-constraints on assessing and choosing paths), I think it’s fair to say evolution has given us a running start.
And the selective pressures are pretty clear: something like the Machiavellian Intelligence hypothesis is almost certainly at work, selecting those genes that are best at carrying out plans, since in the ancestral environment most plans were directly concerned with survival and breeding. Granted, no shortage of optimization was hacked on towards attaining those goals specifically (which might explain why it’s so hard to focus on your bigger goals when you’re hungry), but the fact is humans didn’t evolve to mindlessly seek out food and mates. We’re capable of pursuing other interests.
So I’d say there have been strong selective forces to help us choose effective courses of action. Not absolutely optimal courses of action, but when you consider the massive size of action-space (as you pointed out, rocks fail calculus tests by default), it’s apparent that evolution didn’t totally leave us out in the cold.
This is true and a valuable correction, however I would argue that our planning ability was evolved for very different goals in a very different environment, and while it works pretty well at “figuring out if your friend is backstabbing you” or “figuring out how to get calories”, when it comes to long-term goals in the modern environment (“how do I manipulate this laptop so as to make me millions of dollars over the next 5 years?”) it performs miserably, and all Anna’s points then apply.
Paul Graham recently made a related point, the world is getting more and more addictive, and in order to be productive we must develop more effective screening and anti-time-suck methods: http://www.paulgraham.com/addiction.html
On the plus side, there are a few great people working on this problem—Merlin Mann comes to mind.