Increasing your score on an IQ test is one thing; getting the benefits expected from more "intelligence" may be another.
In such matters, keep in mind Goodhart's Law: "When a measure becomes a target, it ceases to be a good measure."
In the domain of IQ, we might observe a very high correlation between higher-IQ and certain desirable results, like a longer/healthier life, lower rates of victimization, higher incomes, and so forth. And we might have a strong case that the test is, at least at the outset, measuring some sort of "general intelligence", and that there is a causal link from "general intelligence" to the other correlated beneficial outcomes.
But, then you start aiming to increase "general intelligence", and your only measure is the IQ test, so your target for evaluating all interventions is the IQ score. Now you are quite likely finding interventions that only raise the IQ score, and perhaps skip any "general intelligence" benefits. Even if your IQ score goes up, you might not get the other benefits. Meanwhile, others still researching the correlation/causation questions may see the original correlations start to weaken, or the case that IQ measures "general intelligence" weaken.
And in fact the originally-valid ideas are weakening! Motivated action around that target has started to thwart whatever signal it once offered.
The problem is universal - in regulation, in education, in investment, in semantics. It's why a lot of high-performing organizations keep their true 'measures' secret... to slow their dilution through motivated optimizations.
The other thing with optimising as a strategy is while it likes small changes, it really hates structural ones as they make it very hard to compare what you have now with what went before, so anything that is hard to measure is resisted. But usefulness and ease of measurement are not always particularly aligned.
In such matters, keep in mind Goodhart's Law: "When a measure becomes a target, it ceases to be a good measure."
In the domain of IQ, we might observe a very high correlation between higher-IQ and certain desirable results, like a longer/healthier life, lower rates of victimization, higher incomes, and so forth. And we might have a strong case that the test is, at least at the outset, measuring some sort of "general intelligence", and that there is a causal link from "general intelligence" to the other correlated beneficial outcomes.
But, then you start aiming to increase "general intelligence", and your only measure is the IQ test, so your target for evaluating all interventions is the IQ score. Now you are quite likely finding interventions that only raise the IQ score, and perhaps skip any "general intelligence" benefits. Even if your IQ score goes up, you might not get the other benefits. Meanwhile, others still researching the correlation/causation questions may see the original correlations start to weaken, or the case that IQ measures "general intelligence" weaken.
And in fact the originally-valid ideas are weakening! Motivated action around that target has started to thwart whatever signal it once offered.
The problem is universal - in regulation, in education, in investment, in semantics. It's why a lot of high-performing organizations keep their true 'measures' secret... to slow their dilution through motivated optimizations.