"In every job that must be done, there is an element of fun. You find the fun and snap! [snaps her fingers] The job's a game!"Mary Poppins, Mary Poppins, 1964

I attended a session at the Helsinki Oppi festival, hosted by Lauri Jarvilheto (Helsinki Academy of Philosophy), and Jari Multisilta (Helsinki University) about the power of play and "ways in which it can drive more effective learning". It was big on claims and short on evidence, alas, as these things often are. I'd love to find some evidence to substantiate the claims that the Fun Learning camp makes – I'm like an atheist, glumly investigating every miracle hoping to find God – but I'm yet to see anything convincing.

 Can learning be fun? Of course. Is learning sometimes fun? Undoubtedly. Should it be fun? That's a whole different question. Simply saying yes damns every act of learning that isn't enjoyable, and you would have to be completely bonkers to think that everything you learn should be fun as well. Almost everything worth achieving requires sweat, grit and the ability to stick with something when it's hard – also qualities I'd like to see in my students in general. I don't want them bored, but I have no problem if something they do is boring, if it's necessary. I want them to plug away at problems until they break through: that's the pleasure, and it's worth its weight in diamonds.

 Lauri told us about Pavel, his brother, lazy at school but animated to distraction once he found a subject he loved. And we can applaud that, without extrapolating that children should only learn what they love. The point about educating children is that we teach them things that are valuable, not just exciting. Their cultural legacy; science; the inheritance of the world. Lauri claimed that we need to learn to love learning, or we'll see a digital divide – some will know quantum mechanics or three languages; they had the opportunity to start learning early because of digital technologies. I await this generation of savants, teaching themselves Chinese, with patience. As Sugata Mitra's Hole in the Wall project demonstrated, children usually need to be guided through tasks that aren't immediately pleasing. That's why they need adults.

 He also claimed that "In ten years in Finland we can have a 'Thank God it's Friday' culture to a 'Thank God it's Monday' culture if we can help our kids have fun while they're learning". (NvdR: volgens PISA-2012 voelen vooral de Finse lleerlingen zich echt ongelukkkig in klas.)  But the problem is that this entails children driving their own learning, and crucially only learning when they want to, what they want to. You think most children will choose to learn algebra or grammar or anything abstruse or complex? You think they'll choose learning over actual play? Learning physics through Angry Birds over just playing Angry Birds? This project, while gorgeous in its ambitions, has pitfalls you could drive a school bus through.

Jarvilheto introduced us to the latest play-learning platform: Angry Birds Playground. In an experiment conducted by Rovio, the kids who used these materials were – they claimed – "enthusiastic, motivated and concentrated".

 I'm sure these people are engaged in the most rigorous of science, but the area that it addresses is devilled with darkest, emptiest aspects of bad educational research: small intervention groups, interested parties, cognitive bias, short term studies, conclusions that don't necessarily follow from the data, an aversion to testing a theory to destruction, etc. This matters, because huge and enormously expensive wheels are turning in education ministries around the world. Children's lives are chained to this wheel. Poor children can't afford to fix the mistakes of state education, as middle-class children can, through tutoring and familial support.

 I need to emphasise: the people I saw today cared passionately about improving things for children, and I would happily let them babysit my own child. But I'd be very cautious about giving them the keys to the school bus.

 "Playing games can have a measurable, positive impact on how well students learn," so the programme said of this session. So I spent half an hour looking for research that backed this claim up, from either Jarvilehto, Multisilta, the University of Helsinki or even Rovio themselves, who were partners to this session. I'm sure it was just my poor search skills, but I couldn't find anything that related to this session, or that could substantiate the claims that children learned better when learning was in some way gamified. And this matters. Because in my previous attempts to substantiate the claims of the fun lobby, I've come up with similar, plum-shaped results. Which isn't to say that gamifying doesn't work, just that there seems to be precious little research that seems reliable. Which kind of kicks the whole thing into the 'unproven' territory.

 Here's some of what I found on the Rovio website:

 “The concept allows children to experience learning in a fun way. It has been scientifically studied and proven in cooperation with the University of Helsinki, Cicero Learning Network – making education both engaging and inspiring”, said Sanna Lukander VP Learning and Book Publishing at Rovio.

 “What if learning was fun? That was the question we asked ourselves when we started to develop this exciting new concept. Having seen the enthusiasm when children and parents spend time with Angry Birds, we wanted to create fun new learning activities for them,” said Peter Vesterbacka, Mighty Eagle and CMO at Rovio.

“The most optimal circumstances for learning are set when having fun, being motivated, being appreciated for who we are and having permission to be autonomous and experimental. Learning is rewarding and effective when you feel safe to experiment, and this is totally in tune with the Angry Birds Playground fun learning philosophy”, said professor Jari Multisilta, director of Cicero Learning Network at the University of Helsinki.

It really behooves anyone presenting at a conference, or promoting a product online, to link clearly and easily to published research that substantiates their claim, especially if they preface it with "research proves this". Otherwise, people might think that maybe it doesn't. Or worse, they might think that it does, when possibly it doesn't. Extraordinary claims require extraordinary evidence. The idea that children best learn in gamified conditions is extraordinary, given that for most of humanity's history, that's not how conditions were. Where is the mountain of evidence? Where is the data and the analysis that will melt my bronze heart and lead me to love the Angry Birds classroom?

 I'm getting tired of these kinds of claims from people who, coincidentally, sell things.

 This matters because there are thousands of enthusiastic teachers being seduced by these claims, without, it seems, very much to justify them. I've raised the problems of gamification many times before: children only work at tasks they enjoy, the game elements take up huge amounts of time that could be spent actually learning, much that looks like learning is just play... In matters gamified, I am an atheist: show me and I will believe.

 "My name is Thorfin, I am an educational games entrepreneur..." said the next questioner. And I left. It felt like I was turning my back on a tsunami, hoping it would go away. This won't go away. Game-driven education is this year's Brain Gym, and with the kind of momentum it has, it isn't going anywhere soon.

 PS If anyone from any of these bodies wants to respond by presenting at researchED London, please get in touch.mmm