Academia struggled with this problem for a number of years. When I started in 1999 it was typical to toss 50% or more of data from kindergarteners. Then around

Academia struggled with this problem for a number of years. When I started in 1999 it was typical to toss 50% or more of data from kindergarteners. Then around 2002-2005 I saw a few independent groups take this tact. Some gave stories to kids (e.g., about a spaceship or giant camera). Another group used head-mounted motion tracking to train children to lay still while watching movies. I don't know from this story when Doug had his epiphany. But by 2005 it was becoming increasingly common to skin the MRI and even whole rooms. It became clear that you could get better results with a friendlier room and machine. But the bigger difference was an explicit training session. Kids are seldom told to lay absolutely still and for no apparent reason. The challenge was understanding that as the problem to be overcome. The design helped wrap the experience and the memory.