Playing a Memory Game to Improve Intelligence and Increase Your IQ Score?

Remember the letters and improve your IQ. (Photo by Lainey's Repertoire)

It is widely believed that a person's intelligence is genetically fixed from birth and unchangeable by environment. If you're intelligent, good for you; if you're not the sharpest pencil in the box, tough luck.

This idea has now been put in question by Jaeggi et al. who studied the effects of playing a memory game on IQ scores. It's known that people can increase their IQ scores by taking lots of IQ tests, but all that means is that people get better at taking IQ tests – their intelligence in general does not improve. That is, the learning that happens when taking tests repeateadly is not transferable to other tasks.

Intelligence and working memory

The ability to reason and solve new problems independently of previously acquired knowledge is known as fluid intelligence, which is what IQ tests attempt to measure. And they do measure it quite well, as long as they remain novel to the person taking them. With enough practice, however, the test fails to accurately measure a person's fluid intelligence anymore.

The authors of the study propose that by improving working memory, a person can also increase fluid intelligence. The reason is that working memory and intelligence share a common capacity constraint. Basically, the more things you're able to hold in your working memory, the more interrelationships among elements you can keep in mind in a reasoning task. And the ability to solve reasoning tasks is, of course, what IQ tests measure and what intelligence is thought to be.

The dual n-back game

To prove their point, the authors developed a game that improves working memory and then divided the subjects into two groups. Both groups took two IQ tests; one the first day, and a second one some days later. During these between days, the first group practiced their working memory by playing the game.

Sure enough, the IQ scores of the group that played the game increased significantly more than the control group's. Even the control group scored better the second time around, which proves the point about learning how to take the test better by practicing.

The game itself, called dual n-back, is of that "simple to learn, hard to master" type. You can download it for free here. The letter n in the name refers to the number of items you have to keep in your working memory. The number increases as you perform better.

In the game, you hear a letter every 3 seconds, and you have to match it with previous letters you've heard. For example, if n=2, and you hear the letters "A", "B", "F", "B", you press the letter match button to indicate that the last B matches with the previous B, which you heard 2 letters back (hence, n-back). If n=3, you should not press the button, since "B" does not match with "A". You get a point for every match you catch and lose a point for every incorrect match and missed match.

If this sounds difficult, it gets worse. The dual part of the name comes from the fact that in addition to having to remember the letters, you also have to remember a visual signal that is presented with the letters. Every three seconds, a blue box appears in one of 8 positions on the screen, and again, you have the match them. The letters and boxes don't correlate in any way, meaning that you have to keep both things in mind all the time.

The study

The first few times you try the game, it seems impossible. But as you keep playing, you start getting better at it. Here is a graph depicting how the test subjects did in the game:

The subjects in the game group were further divided into four groups, each with a different number of training days before the second IQ test. As you can see, the number of items they could keep in the working memory increased with each training session in all groups. The 19 day group got all the way up to n>5.

Note that the mean n-back level of all groups is a respectable 3 even after the first day. And these subjects were normal people, not geniuses. Since the amount of games per training session was 20, this means that if you're averagely intelligent, you should be able to play the game with the "dual 3-back" setting after 20 tries.

The inhuman experiment

Who doesn't want to be smarter? At least to me, improving general intelligence just seems too good to pass on, so in the spirit of this blog, I'm going to make an experiment of this.

To match the study as closely as possible, I tested myself on two free IQ tests available on the web before playing the game. I've now played it for a couple of days, and even though it seemed extremely difficult at first, I've gradually gotten better at it. Still, scoring properly even with the 3-back setting still seems quite hard.

After 20 days, I'm going to take the same IQ tests again and see if there's an improvement. If you download the game, drop a comment and let me know how you're doing!

For more information on brains and intelligence, see these posts:

Increasing Intelligence by Playing a Memory Game – Experiment Update
Caloric Restriction Improves Memory in the Elderly
Moderate and Severe Calorie Restriction Alter Behavior Differently in Rats
How the Accumulation of Minerals Might Cause Aging