3 ms·

I finished it and scored 138, which ostensibly is a high score assuming the bell curve at the end is correct... and I'm just left wondering what was it I was be

by yifanl 4y ago

I finished it and scored 138, which ostensibly is a high score assuming the bell curve at the end is correct... and I'm just left wondering what was it I was being tested on. Pattern recognition, I guess? It's a bit interesting since all the data work I've worked with involves labelling and categorization which is the same general idea, and I wonder if that' just coincidence. I've always thought that data science has been about labeling data because that's one of the most immediate business desires, or is that what the cognitive sciences actually consider to be a good abstraction of intelligence, if such a thing even exists.

outworlder 4y ago
I'd argue that pattern recognition is basically the brain's most important job. Acquiring new patterns is another (but you'll notice how easy it is when the new patterns mostly fit existing patterns). That's not to say that the tests actually mean anything. They probably don't.
- pessimizer 4y ago
  > I'd argue that pattern recognition is basically the brain's most important job. True, but I'd argue that there's little to no reason to think that this is innate, and that it can be trained. Doing these pattern recognition puzzles over and over again would make you very good at them.
  - outworlder 4y ago
    I think there's very little of what we call 'intelligence' that's innate. Not sure if there's any consensus, but most papers I've read seem to indicate that the difference is either tiny or non-existent, for similarly healthy brains.
- lordnacho 4y ago
  Not recognizing patterns that aren't there is another related and important job, which is in conflict with what you mention. I'm not sure this is really being tested. If you look at patterns like 1,2,4,8, etc (to choose something that I can write in text) there's actually a lot of next numbers that could fit a pattern, but to get the points you have to pick a certain one.
  - outworlder 4y ago
    Excellent point. Ignoring information seems to be a major function too :)

by yifanl 4y ago

outworlder 4y ago

I'd argue that pattern recognition is basically the brain's most important job. Acquiring new patterns is another (but you'll notice how easy it is when the new patterns mostly fit existing patterns). That's not to say that the tests actually mean anything. They probably don't.

pessimizer 4y ago
> I'd argue that pattern recognition is basically the brain's most important job. True, but I'd argue that there's little to no reason to think that this is innate, and that it can be trained. Doing these pattern recognition puzzles over and over again would make you very good at them.
- outworlder 4y ago
  I think there's very little of what we call 'intelligence' that's innate. Not sure if there's any consensus, but most papers I've read seem to indicate that the difference is either tiny or non-existent, for similarly healthy brains.
lordnacho 4y ago
Not recognizing patterns that aren't there is another related and important job, which is in conflict with what you mention. I'm not sure this is really being tested. If you look at patterns like 1,2,4,8, etc (to choose something that I can write in text) there's actually a lot of next numbers that could fit a pattern, but to get the points you have to pick a certain one.
- outworlder 4y ago
  Excellent point. Ignoring information seems to be a major function too :)