3 ms·

Just keep in mind that it’s correlation you’re talking about. You can have a high IQ and none of those capacities. You can have a high IQ and be severally autis

Just keep in mind that it’s correlation you’re talking about. You can have a high IQ and none of those capacities. You can have a high IQ and be severally autistic. Etc. You can also test normal but be very high functioning. So there is no necessary or causative connection between IQ and effective capacity. Robots like ChatGPT May test substantially in some IQ tests and have no intelligence or agency at all.

csa 4y ago
IQ is not a measure of low functioning or high functioning or “effective capacity”. I certainly have never claimed that it is. I think many people put the onus of predicting success on measures of IQ, and most/all good IQ tests don’t claim to do this. There would be a face validity issue, and there are much better tests that can predict success (however success is measured) than IQ tests. I think that this is grossly unfair to IQ tests. Will AI be able to do well in IQ tests? Maybe, but I don’t think that it will be relevant. My stance has been and is that IQ tests measure the capacity to learn, extend, and expand on new topics (in humans). Nothing more, nothing less. Some other points: - IQ tests are not a measure of effectiveness. Many, many more variables go into that. - IQ tests are biased, and pretty much have to be. That said, good IQ test designers try to minimize the bias for their target audience. I think that this has been more or less effectively done in the US for folks in the middle class or higher. - Related, any give score on an IQ test should be seen as an IQ floor rather than an IQ ceiling. There are many reasons why someone may not score their maximum score (or even close to it) when taking an IQ test. Anxiety and fatigue are simple examples. - IQ is correlated to income, but is this is a very broad correlation. Specifically, average income goes up with IQ up to about 135 or so (can’t remember the exact number) and then starts decreasing (possibly due to being in lower paid fields like academics, possibly due to social dysfunction). But these are all aggregated numbers. The individual results vary wildly within any narrow IQ band. - As you alluded to, effectiveness and/or success is predicted by many other things other than IQ, and in many cases IQ may only be a weak predictor. IQ tests don’t claim to predict success. Things like social skills, socio-economic status, access to resources, physical health, and mental health can be much better predictors of effectiveness/success depending on the task. - IQ tests are quite good at finding people at the extremes (very low and very high), but actually does a fairly poor job of differentiating folks at those ranges. On a personal level, I think the main use of IQ scores would be to identify these folks early and get them the resources they need. To close, here is a decent article that discusses one view on how IQ and “success” are only loosely related: https://www.verywellmind.com/are-people-with-high-iqs-more-successful-2795280 https://www.verywellmind.com/are-people-with-high-iqs-more-s...

csa 4y ago

IQ is not a measure of low functioning or high functioning or “effective capacity”. I certainly have never claimed that it is. I think many people put the onus of predicting success on measures of IQ, and most/all good IQ tests don’t claim to do this. There would be a face validity issue, and there are much better tests that can predict success (however success is measured) than IQ tests. I think that this is grossly unfair to IQ tests. Will AI be able to do well in IQ tests? Maybe, but I don’t think that it will be relevant. My stance has been and is that IQ tests measure the capacity to learn, extend, and expand on new topics (in humans). Nothing more, nothing less. Some other points: - IQ tests are not a measure of effectiveness. Many, many more variables go into that. - IQ tests are biased, and pretty much have to be. That said, good IQ test designers try to minimize the bias for their target audience. I think that this has been more or less effectively done in the US for folks in the middle class or higher. - Related, any give score on an IQ test should be seen as an IQ floor rather than an IQ ceiling. There are many reasons why someone may not score their maximum score (or even close to it) when taking an IQ test. Anxiety and fatigue are simple examples. - IQ is correlated to income, but is this is a very broad correlation. Specifically, average income goes up with IQ up to about 135 or so (can’t remember the exact number) and then starts decreasing (possibly due to being in lower paid fields like academics, possibly due to social dysfunction). But these are all aggregated numbers. The individual results vary wildly within any narrow IQ band. - As you alluded to, effectiveness and/or success is predicted by many other things other than IQ, and in many cases IQ may only be a weak predictor. IQ tests don’t claim to predict success. Things like social skills, socio-economic status, access to resources, physical health, and mental health can be much better predictors of effectiveness/success depending on the task. - IQ tests are quite good at finding people at the extremes (very low and very high), but actually does a fairly poor job of differentiating folks at those ranges. On a personal level, I think the main use of IQ scores would be to identify these folks early and get them the resources they need. To close, here is a decent article that discusses one view on how IQ and “success” are only loosely related: https://www.verywellmind.com/are-people-with-high-iqs-more-successful-2795280 https://www.verywellmind.com/are-people-with-high-iqs-more-s...