Doesn’t it seem like interactive games would be excellent teaching tools? I know I’m a little biased — I used to work in the educational games industry — but I don’t think you have to be a C++ programmer to realize the educational potential of this technology. I helped develop computer games that teach kids how to add, subtract, multiply, type, create grammatical sentences, speak French, Spanish . . . you name it. Of course some games were a lot more fun than others, but it seemed to me there was no way this technology wouldn’t transform education.
But that didn’t happen. Sales in educational software plummeted from $498 million in 2000 to $152 million in 2004. Most of the educational brands that hogged Wal-Mart end-caps in the late nineties don’t even exist anymore; I hoped to find out why at the Workshop on Games for Learning, Development & Change at the Annenberg School. Scholars presented fresh research about “serious games” and, while hope is alive, the news was often very discouraging.
One of the biggest problems is that it’s really hard to prove the skills learned in a game will transfer to No Child Left Behind aptitude tests, and even more importantly, to the real world. Why not just do an experiment? Well, unfortunately, testing interactive media is even harder than testing “passive” media such as TV. It took a huge investment to figure out what worked in Sesame Street — it’ll take even more time and money to measure the effects of more complex media, where the number of variables rises exponentially. And establishing control groups? Forget about it.
Most game researchers and aficionados understand that “flow” — that pleasurable feeling of absorption and concentration that makes games so addictive (and potentially educational) — is extremely hard to measure. And, even if you decide you won’t bother measuring it, you’re bound to disrupt it with all the boring quizzes and evaluation tools scientists use to figure out whether something was learned or not.
Art Graesser from the University of Memphis gave a particularly sobering talk. In one of his experiments he discovered that there was a negative correlation between liking the learning experience and “deep learning.” Also, if students complained about being confused, they generally learned more than their compatriots who had a better time. “No pain, no gain” is not the mantra that serious games folks want to hear. The main reason educators turned to games was to make learning enjoyable.
So is there a bright side here? Absolutely. Just because this is hard, doesn’t mean it’s not going to happen. And researchers have not given up the ghost. Many felt we should take the stealth approach. Create games that offer opportunities for incidental learning: make the messages implicit rather than explicit. It might be that the process of game playing itself — regardless of content — is the most fruitful way to expand cognitive capabilities.
Also, instead of trying to create games that allow scientists to isolate discrete variables, put more resources into developing smart tools to sort through the vast datasets that users create as they play games that haven’t been cooked up by a social scientist (i.e., games that are actually fun). Gathering the data is not the problem. Understanding it is. Alternately, try embedding the evaluation tools into the game from its inception so that you can tell, just from the way a player is playing, whether they’ve grasped a concept or not. Either way, getting out of the way of the gameplay is clearly the way to go.