IQ Test Score Improvement

This is a new section of the forum to discuss ideas about how to train to improve test scores.

My first question is about IQ test scores.

I don’t think that IQ test scores are important in life, but I think it would be interesting to see if there are ways to train or study in order to improve the test scores.

What do you think? Is it possible?

1 Like

Yeah there’s something called Dual N Back which is a test being administered by researchers in various universities at the moment and it has demonstrated increases in WORKING MEMORY ( which is linked to fluid intelligence and hence linked to IQ). This is the FAQ that was most informative to me and gives you everything you need:

My personal account: I started in September 2011, and then stopped in about April 2012 due to Medical school exams. I got up to 8-back ( you’ll understand this if you read the FAQ) and I find that after 3/4 months my level hasn’t really dropped, suggesting a change in brain structure?? I will admit that I did not take an IQ test before to establish a control in my experiment on myself, but my personal and subjective feelings are that I understand difficult concepts a lot more easily and I feel a lot calmer.

The link between working memory and IQ is still hotly debated but I undertook the exercise telling myself that at least my working memory would improve, which would have at least some beneficial effects on my studies.

Did you use any techniques or strategies to get to 8-back? Are you using Brain Workshop or another program?

Yeah I’m using Brain Workshop. The strategy varied depending on the number of objects to be remembered. I didn’t need strategy to remember up to groups of three, so that was the basis of my strategy. So:

4 back: remember a group of 3, followed by 1 object ( or vice versa)
5 back: remember a group of 3, followed by 2 objects ( or vice versa)
6- back: 2 groups of 3
7-back: a group of 4, a group of 3
8-back, 2Xgroup of 3, group of 2

The program works in a way that in each session your warm up is to work your way from the start to your highest possible n-back level. So if my maximum was 7-back, I would spend a few sessions working up to there, and I found that having surpassed up to 6- back meant that I no longer needed “grouping strategies” as I call them.

In terms of rehearsing them out loud or in my head, I spent the first few weeks saying them out loud. Subsequently I read the article below and I rehearse them in my head. I find that results are really variable and the exercise is often not enjoyable without rehearsing ( i.e. repeating what you’ve been shown whether vocally or mentally). - this explains the whole community’s opinion on strategy and I currently endorse the mental rehearsing strategy.

I think IQ tests are a bunch of ■■■■■■■■, to put it mildly. It’s basically problem solving and problem solving can be trained and improved in much the same manner as we’re improving our memory with mnemonics.

It’s not about your processing power it’s about how you use it, there are brain “hacks” to mentally accomplish just about anything you can think of, it’s all a matter of figuring out the thought methodology.

I’m a 2nd year medical student in London. Having compared my friends at university who do less knowledge based subjects like Mech Eng (versus Medicine which is highly knowledge based), I find that it is very hard to teach problem solving skills.

This is not to say problem solving skills are completely untrainable! Everything can be improved by hard work, but when you approach university level difficult and enter the realms of post graduate and PhD level, these skills reap more rewards if they are innate. You are assuming that every problem follows a similar pattern and hence can be solved by the same algorithm or steps. In higher education the difficulty approaches a level where problems often do not follow a pattern and so you need to develop new ways of solving problems on your own.

Before you cite some example of people who have learnt it, I would like to say that exams are ultimately not infallible and you can predict a lot of questions, so studying past question styles can get you great success without intrinsic skill. Is this what you are basing the “success” of these “brain hacks”? I’m pretty sure that no successful PhD who has been published in a well known journal for his/her research doesn’t have some level of problem solving skills that are innate.

UCL, GKT, St. Georges, Imperial or Barts?

I’m also very interested in your experiments with working memory and the n-back. Have you noticed any cognitive benefits other than the improvement in n-back, and do you think your time spent training n-back was worth it?

Sorry I am a little confused as to the process I am meant to be adapting with Dual N-Back. I am only around the level 3 mark because of such a horrid memory I assume. I say the letters in my head but just with instinct or not much attention with the position of the blocks.
What technique should I be using and how?

Hi yeah I’m at UCL atm. I have to say that I personally find IQ tests and career tests very limiting. I was told that I should be a social worker according to my career tests (similar to IQ but slightly different) which is something that really put me off IQ tests and such. I feel that the whole “IQ can never be changed” attitude really pigeonholes you into a limited mindset, when really any person can achieve most things!

The reason I mention this is because upon beginning DNB, you are supposed to take an IQ test before to establish your base level, and then continue to take IQ tests during your training to establish a change. Given my experience with such tests, I decided to skip that. Scientifically this isn’t very good but I was simply curious about DNB and whether in fact you can improve working memory, because although the link between your working memory and IQ isn’t established, improving working memory alone can have benefits ( proven I believe).

The changes that I saw in short: I began to grasp difficult concepts more easily and the speed of understanding improved. Furthermore, I did much better in my medical school tests. This is all very subjective of course, because tests are predictable and medicine is still fairly knowledge based. For me, it was definitely worth it purely on a curiosity basis, but the one downside is that once you get to D8B or so, it takes up about 40-50 mins of your day. This is definitely a downside for me because of the limited time that I have.

If you want proof before embarking on it, read the following site below. It has many more opinions than my own, including the researcher who pioneered this test ( Jaeggi). A lot of them are positive, including the research paper published in 08 which reported successful results.

If you want more information, join the google group for Dual N backers ( google it).

I think you’re considering things too broadly. I’m not saying that every problem follows a similar pattern, that’s not what I meant at all. I’m talking about learning how to analyse problems, how to find patterns, how to THINK. If there’s a problem that nobody has ever seen before even someone with an IQ of 200 won’t figure it out straight away, relying on past experience/data definitely helps but that’s not what problem solving is.

And these days with the internet getting information is extremely easy. Even mnemonics is problem solving, the best mnemonists figure out better and faster ways of encoding data into imagery and location, having a good innate memory doesn’t help because no matter how good your memory is you still won’t memorise 50 numbers in 5 minutes without mnemonics, there’s no such thing as a photographic memory and there’s no such thing as a person that’s innately good at problem solving, if that was true we wouldn’t have any problems. =P

1 Like

Two things I’d like to address here. Firstly, I don’t think I mentioned memory and mnemonics training as in the same category. I completely agree with you that memory can be trained.

Secondly, your clarification tells me that you want to teach people how to analyse problems and how to find patterns. Look I agree that this can be taught, but what is the success rate? To me those instructions are obvious and most of us think of them already. I agree that for those to whom this thought process doesn’t occur naturally, then you are perfectly right.

  1. Read the question properly
  2. Highlight key words and re-read to understand what a problem is asking
  3. Correlate what the question is asking to what knowledge you need to answer it
  4. Retrieve this knowledge
  5. Use the knowledge etc.

e.g.: Why does the Earth rotate around the Sun? ( not a novel question but if we were to think it through as a novel problem):

  1. What causes a rotation for all objects ( centripetal force)
  2. What is causing a rotation in this case…etc.

Also “innate” doesn’t necessarily mean a superhuman power to always be able to solve every problem. Innate simply refers to genes that allow your brain to have slightly greater capabilities, which MAY lead to better problem solving skills. A person with an IQ of 200 may not be able to solve every problem, but he certainly has a higher probability of success than a person with an IQ of 100.

I don’t agree with this at all. That completely depends on how the problem is tackled and the 200 IQ guy can make a mistake while the 100 IQ guy might get it on the first shot depending on his background/past experience. If anything it’s your education and environment that matters the most not your genes.

Your type of thinking leads to the problem that people keep saying “oh but I’m not as smart as that guy so I can’t do this” or “well he has talent” etc…

Even though plenty of professionals keep saying it’s nothing but practice and good strategy, people keep falling into the trap of blaming lack of talent or god’s will or whatnot. If you’re not completely brain damaged, you can figure it out. And every time you figure a problem out your brain gets better at figuring out problems, just like with each new bit of information that we memorise, memorising other information gets even easier.

Everything snowballs, you just need to start somewhere or tackle the problems you can’t solve in a different way.

In science if a problem seems unsolvable it’s usually because you don’t have enough data and/or need to revise the problems within the problem.

And I think you’re wrong about mechanical engineering, it all comes down to maths and higher level maths depends on your ability to deal with the lower level maths and logic solving. Also don’t forget, if you’re having issues with some test questions that sometimes they are just stupidly written, which is where english comprehension comes in to unravel the mess that lecturers often write.

Look I’m going to end this discussion after this post because this discussion is more about Dual N back rather on whether people can be trained or not.

You’re taking this whole discussion way out of context I’m afraid. I never said that innate skill is more important than hard work and strategy. My point was that there is a limit to what you can teach in life, just like there is a limit to having a large IQ.

You keep interpreting everything in such an extreme way! I’m not in any way trying to say that people are destined to fail or succeed based on their IQ tests!

Genetics, education, and hard work all count towards success. Perhaps hard work and education can have a more profound impact. However, if you compared two people with identical level of education and hard work ( and other factors that are equal), but with different IQs, it may be that the person with a greater IQ might be more likely to do better. In life, this may not happen of course due to either luck, varying levels of hard work, and other events ( stress, family issues).

No-one is a bigger proponent of hard work than me - this is evident by the fact that I’m trying to learn 100 different techniques to try and improve how I study ( which is why I’m here on this forum!). But you cannot simply shut your eyes and avoid reality. Genetics does play a part - it is not the “be all and end all” and I never suggested it was. I cannot control genetics, but I can control my hard work and my education, and I’m doing that.

My observations on engineering are based on having about 20 friends who are mechanical engineers and their own comments. There are also my two uncles who have been mechanical engineers who shared information from their 20 and 30 year careers respectively. They all agree that mechanical engineering is a more problem based subject versus medicine which is relatively more knowledge based. You can be a perfectly competent engineer with moderate talent and extreme hard work, but those at the top of their field are generally very talented AND hard working.

Out of interest, what do you study?

I agree with that. I don’t think the tests are completely useless, but maybe they should have a different name and not be tied to the word “intelligence”. IQ tests are a test of how well one takes IQ tests. :slight_smile:

My main interest is: if a person took an IQ test today, and then trained, could they raise their score by 20 or 30 points the next time they took the test?

I agree with that. I think that having a certain mindset, along with hard work, has much more to do with “intelligence” or “achievement” than a test score.

1 Like

[quote=Josh Cohen]

Well just as an example I’m pretty sure most IQ tests have a memory component and we all know mnemonics whoop memory test arse, so I don’t see why you couldn’t improve on the other areas of the test in similar ways.

So just by doing mnemonics, which we’re all doing here anyway, you probably already increased your score by a bit.

I’m a similar critic of IQ tests. I definitely think someone could raise their IQ by 20 to 30 points by training. As an admissions test for medicine in the UK, we have a test called the UKCAT, which works similarly to an IQ test. In the UKCAT you get a score of between 300 and 900, about 620 is the average score. I did the UKCAT without any study and got a pretty good score, but the interesting part, I think, is that I know of a guy, sat the UKCAT first time round and got a score of 517.5 - an awful score. A year later, he sat it again and got 812.5 - an absolutely phenomenal score.

Now the magnitude of points in UKCAT and IQ are very different. It’s much easier to jump 50 points in the UKCAT than it is to jump 20 points in IQ. But I am more than sure that someone could get this kind of a jump with a bit of training.

One of my favourite examples of why IQ doesn’t mean squat is this woman. She has a conflated view of her own intelligence because of her IQ score, but I think she’s an idiot.

Writes book criticising proof of Fermat’s Last Theorem
Has never studied maths
Displays ignorance of first year college mathematics

1 Like

Well yes there’s no doubt that any test is not completely infallible. Since the IQ test does have specific problem types, the “re-test effect” can lead to higher scores with some practice. I’m less interested in the tests and more interested in developing the concept called ‘g’ or “general intelligence”. It is debatable to what extent this exists, but in whatever form it may be, I am interested in improving it.

Hype, I did the UKCAT too and it is a horrible predictor, in my opinion, of intelligence. I observed that there was really no correlation with UKCAT test results and the general academic success of my friends.

There’s also an interesting philosophy on IQ tests that the true potential of an individual or his ‘g’ is only truly revealed when he/she does an activity that they’re interested. Hence the logic follows that sometimes, bad IQ scores reflect poor interest in the task.

Exactly why I think the two tests are similar.

I agree with this. There are certain mental abilities which are innate, but are hard or impossible to measure with tests such as IQ. IQ is certainly correlated with “intelligence”, but it is not infallible. There are those people who are exceptionally strong in one area while being weak in others (Kim Peek comes to mind). Lateral thinking is the most important skill when it comes to problem solving IMO, and it is trainable to a certain extent, but it is much like athletic ability in that some are just more naturally gifted than others and only a small percentage of people can reach elite status.

The benefit for those who are not naturally gifted is that they have more motivation to put the work in. Michael Jordan couldn’t make the varsity team in high school, so he put in the work and became the most dominant player in the history of the NBA. A lot of natural talents try to get by on their innate skills and never learn that discipline necessary to optimize their abilities.

I think that “intelligence” is one of those words that is difficult to have a conversation about, because it doesn’t really have a single definition. Solving puzzles on a test doesn’t necessarily translate to solving problems in real life (like some of the examples above illustrate).

Here is one definition:

To my mind, a human intellectual competence must entail a set of skills of problem solving — enabling the individual to resolve genuine problems or difficulties that he or she encounters and, when appropriate, to create an effective product — and must also entail the potential for finding or creating problems — and thereby laying the groundwork for the acquisition of new knowledge.

Here is another definition:

The ability to deal with cognitive complexity.

I can think of people who would score very high on an IQ test but who cannot apply basic logic to other aspects of life that require cognitive complexity.

Here’s another quote:

It is difficult to define what constitutes intelligence; instead, it may be the case that IQ represents a type of intelligence.

…to base a concept of intelligence on IQ test scores alone is to ignore many important aspects of mental ability.

I’d agree with the last sentence. Even terms like “cognitive ability” and “learning ability” are a bit ambiguous. It may be possible to get a sense of what they mean when you see them in a person, but I don’t think they can be easily measured by assigning people a number that is never supposed to change.

Even though I really dislike IQ tests as a measure of “intelligence”, I’m curious about ideas on how the tests could be “hacked”. :slight_smile: