Improving fluid intelligence with training on working memory

"Improving fluid intelligence with training on working memory" is the name of the paper by Susanne M. Jaeggi, Martin Buschkuehl, John Jonides, and Walter J. Perrig. It details how, in this study, subjects increased their intelligence (as measured by gF) by playing the dual n-back game. Should this research be confirmed, it would be unique. At present, no other reliable way to increase intelligence exists. See here. In the words of Wired magazine, "Your IQ is basically hardwired." That's why it's so surprising that this study uncovered something as simple as a game that can influence IQ (or gF, a proximate measure of IQ).

In the study, groups of people took an intelligence (gF) test and then practiced the dual n-back game for 8 to 19 days. At the end of that time, they took another intelligence test and scored substantially better than people who didn't play the n-back game. This in itself is certainly an interesting result, but in addition to this, it was also shown (though training groups were small) that the people who played the game longer (19 as opposed to 8 days) did even better on the gF tests. This indicates that the longer a person "trains" on the game, the greater the benefit.


While this research is very exciting, it's also very preliminary. There is no way to know whether or not this research will eventually pan-out. To be sure, the methodology seems fine and the journal is pretigious, but this finding has yet to be replicated anywhere. Here are some potential problems:

Effect Duration

It's unclear how long the presumed intelligence benefit will last. The authors haven't had time to look at this, so it's unclear whether the effect persists for any amount of time after training. That is, even if the effect is confirmed in other studies, it may be necessary to continually train in order to maintain the benefits of training.

Nature of Intelligence

Robert Plomin argues that gF isn't immutable as the paper authors suggest. This is a reminder that the very nature of intelligence is still under debate. There are a minority of scientists (and presumably Plomin is among them) that contend that schooling has a substantial effect on intelligence. While there aren't high-quality studies that clearly support this notion (despite decades of research), it is still a debated point. Even then, Plomin doesn't argue against the study, its findings, or the benefits of training. Rather, he's debating the point of whether or not this is anything new.

Nature of Intelligence Tests

Michael Merzenich (a former neuroscientist at the University of California) argues that the dual n-back task might just make people good at taking intelligence tests and not neccessarily improve their actual intelligence. While this is certainly possible, it will take much more research to figure out if this is actually true. However, it is known that gF tests are fairly predictive of future success, and it's unclear how playing the dual n-back game would change that predictive power.