h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

You might want to add a comment in the section about problems with IQ tests talking about how you can learn to do them.

Theoretically your intelligence is a fixed attribute and however may tests you take you should get a similar result. However if you practice you can learn to answer the questions in an IQ test correctly. This is mainly because they all use the same sort of questions to test for the same set of attributes.

As a child I can remember improving my score by as many as 20 points in a day because I decided that the tests were fun.

Roz.

good point, although the attitude that I.Q. doesn't change is also rather questionable.

I didn't mean that IQ didn't ever change but rather that on any given day it shouldn't change (and a difference of 20 points is not good)

Roz.

I will think about it, and see what I can add about it.

IQ can vary dramatically from day to day (and even within a day).

A serious illness can bring my effective IQ down to that of a highish level university graduate on average over a time period. For example, I still got a distinction in a post-grad diploma type exam while pain-killers were helping for part of it but was unable to operate a pen properly or concentrate on correlating simple tables by the end. Despite the good result, *I* could tell the difference between my normal and local functionality even during the better part of the exam.
Having two serious illnesses simultaneously can bring my IQ down to below human normal, eg not really able to cope with money in the chemist/pharmacist shop. Fortunately this is a rare thing as I don't like it. I believe it is not as bad to never have had something as to lose it for some time and be able to tell the difference. As a general example of this, those people who acquire brain damage in such a way as to not notice anything is wrong are usually regarded as luckier than those who are aware of what is happening to them.

I'm not sure whether this variability (NB in the opposite direction to IQ apparently improving with practice) is anything you would fit into your project though.

funny you should mention that, because I suffer from a funny blood sugar problem which can drop my psychometrics results by over 30% after eating.

It is currently being played up by this blasted heat.

What you are suggesting would probably be a section on increasing iq and another on decreasing it.

I will have a think about it.

I regard most IQ increases as fake - ie the person hadn't been concentrating properly before or the test hadn't been fair so that practice made a difference. These are all things I'd noted before but it is interesting to see the establishment position finally catch up. I usually find they are 20 to 30 years behind on anything so this is probably about average.

Oops, digression. What I meant to say was that the only variability in IQ from its _true_ level is downwards. Anything else is the result of measuring on a bad day or in a bad way. But since the variability is large and frequent that is almost inevitable. Dehydration is another factor which decreases physical performance and concentration and therefore is bound to affect IQ. It is just possible that there are some (illegal? or military?) drugs which speed up reactions and perceptions temporarily and might make a positive difference to IQ. There is something on the net about the drugs used to keep US soldiers alert in Iraq but I don't recall any IQ data.

If there are not any drugs which temporarily increase iq, then I am sure the drug companies are working on them.

While I agree that generally, iq doesn't change, there are aspects which could reasonably change, and any test which had a significant sensitivity to these could give you better results over time.

for example there is the fluid intelligence, which is related to how good you are at problem solving (all other things being equal). This shouldn't really change, but if the tests are sensitive to particular types of problem solving being learned, then you could score artificially low in early tests and artificially high in later ones.

In fact, most of the problems of learning to do better at iq tests are to do with sensitivity of those tests to chrystaline intelligence (which covers the sort of things you tend to put into expert systems and databases).

change with age isn't really relevent either, as although you do loose some brain cells over time, the increase in connectedness between the ones you have left more than makes up for this loss in anyone with an active mind.

Don't forget that I'm saying IQ does change - downwards - when ill.

Many IQ tests (possibly more previously than now) seem to confuse knowledge with intelligence. I think I probably prize most highly what you are calling fluid intelligence. I do have a lot of what I'd taken to calling co-processors and you seem to call specific intelligences though. It irritates me that more use is not made of expert systems and databases in areas where crystalline intelligence is clearly the key, eg medicine (assuming we are talking about the same thing here). I tend to regard that as rather less than intelligent - both the behaviour of not doing so and the type of supposed intelligence itself. I should probably go off and find your definition to check though.

I also think there isn't much change of IQ with age (barring serious degradation), though the level of knowledge on which intelligence can act can increase with age. This is more of a problem with any IQ test which assumes some allegedly common knowledge that a young child (or alien) wouldn't have.

This is probably the wrong place, but apart from the very brief section on genetics I hadn't seen much on the factors leading to high (or low) IQ. In particular oxygen deprivation (or the prevention of its normal occurrance during birth) has been proved to make a significant difference. Put simply, all normal people are brain damaged to some extent before they start - except for those born under exceptional circumstances (eg very quick/easy birth, some caesarians, deliberate experimental intervention with oxygen supply).

Also there is the matter of brain size. While this has mostly been shown to be unimportant in the direction some narrow-minded people expected, ie bigger is not necessarily better and the foldedness and connectedness is what matters. There remains something I had observed as possible and which would be supported by that 70% underlying and mechanically measured speed factor. Small should actually be fractionally better (as with microchips and integrated circuits).

You are right, I hadn't got anything about the causes of low intelligence. It should probably go of the intelligence and learning difficulties page.

Brain size shouldn't matter much, because the transmission speed is so slow, and the brain so compact that the difference in transmission speed won't make much difference.

Regarding the use of expert systems, it is mainly to do with the way funding works. Where there is plenty of funding the problems seem to be so simple that everyone knows the answer, so there is no use for the resulting system.

Where experts are in short supply, they are usually too busy to put in the number of hours needed to create a good expert system, so those ones don't get built.

Where there are committed experts, and enough time, there tends not to be that much money.

It is quite possible that the specific intelligences identified by gardner and others are due to parts of the brain acting as dedicated co-processors. Equally it could be that one part of the brain acts like a floating point subprocessor and is used in lots of similar tasks. we won't really know which is true until we get more results from the detailed research which is only just begining to be done.

I can confirm (because of my strange mixture of abilities and disabilities) that these co-processors are quite separate and dedicated in some cases - hence why I started calling them co-processors in the first place. For others I don't have enough data on my own to locate them or even assert that they have such a unique location, but could still make educated guesses.

Anyhow I suspect this would all be out of the scope of your article(s).

not necessarily, the scope keeps changing.

however it might have the problem of only having anecdotal evidence, and thus being regarded by the editors as an opinion piece.

My testimony alone might be anecdotal (though some of it is quite objectively provable) but, since I see from your articles that the rest of the world is catching up, there may well be some more research out there already.

MENSA reckon on a 10 point drift either way. I took a series of IQ tests as a teenager under exam conditions and consistently came out with 155 +/-10. More recently I applied to join mensa, got 155 on the first test and 143 on the second, therefore making me unable to join, which grated ever so slightly! Surely an IQ test that takes longer to complete is likely to isolate this 10 point drift, as it will make the end result much more accurate?

This was exactly the approach taken by the 4 sigma group.

they had 2 tests you could take, both of which had questions which you answered and then returned the tests.

the answers were then clustered (in their system) so that they new which questions could be answered by which ranges of intelligence.

unfortunately, they would only take the top 1 percent of people worldwide, and now seem defunct.

a pitty really.

Part of it could be that the top 1% (or less) don't want to waste money and time on joining in such things. Americans may be slightly more gullible (only judging by TV rather than real statistics) but people round here just can't be bothered with the pointlessness of it all - MENSA being regarded as a joke.