Computers, creativity and learning

Analysing exergame use over time

My PhD student Andrew Macvean and I  just got a CHI 2013 paper accepted about the fitness game work we did last summer. This is the pay off for standing around in the freezing rain watching teenagers stumble through muddy puddles. (The pay off for us, that is; the kids' reward was the joy of increased fitness.) Basically we are arguing that a) there is no point in doing one off pilot evaluations of exergames because you can't get fit for playing for a single 20 minutes b) exergames should be designed for more than light exercise and c) you can use self efficacy theory to predict how individual users will set and respond to goals. I think we put it more eloquently in the paper though.

Get the paper: Andrew Macvean and Judy Robertson.2013.Understanding Exergame Users’ Physical Activity,Motivation and Behavior Over Time.To appear in In Proceedings of the 2013 ACM annual conference on Human Factors in Computing Systems (CHI '13). ACM, New York, NY, USA Download CRC18thJanJRTwo

And here's the (now) obligatory CHI video. It turns out that we are better at writing papers than we are at making videos. Who'd have thought it?

  

Tired happy kids

Today was Andy the Model PhD student's usability study in school so I went along to help. I must say I was dreading the logistics of it: 20 kids, 10 iPhones, 2 busy teachers, gym kit eating pets and the Scottish climate to juggle! In fact it worked out very well in spite of a few initial problems about the kids not being out of class in time to get ready.

We had the rewarding spectacle of a class of excited 13 year olds zooming around the football pitch running away from invisible ghosts. It's a location aware version of Pacman, so the user plays paman by physically running around the field. A ghost appears on the phone screen and chases pacman, and the user must run to escape it. The ghosts get faster as the player gets through each level. The kids really seemed to enjoy it! They did not have the same problems that me and some other adult pilot testers had with working out which way their pacman was moving and how the real world was represented on screen. It just goes to show why it is worth user testing with members of the target demographic group instead of your dim and ancient supervisor: it can save you some programming work. I was also surprised to see that the kids were really exercising hard! They were out of breath and red faced with effort before the end of the 6 minute initial test. Even although they noticed that it is sometimes possible to evade the ghost by walking, it didn't stop them rushing around at top speed.

As a proof on concept, this is pretty gratifying. Andy's game can make teenagers take exercise while smiling. It's a pity this lot weren't fat to start with, but you can't have everything. :-) Next question: can the game keep them motivated after the initial novelty wears off?

When everything is embarrassing: designing for teenage girls

[Cross posted from CACM blog]

Do you remember what it was like to be 15? I was reminded today of the toe curling embarrassment which teenage girls seem to face on a regular basis. I went with my PhD student, Andrew MacVean, to a user centred design session with six fifteen year-olds. Andrew is designing a location based fitness game for iPhone, and he wanted to observe members of the target user group playing fitness games on the Wii Fit to give him some feedback about effective game mechanics.

I often find user centred/ participatory design work useful just to get a feeling of what it is like to be a user in the target group. Mostly this has involved working with cute little kids (10 year olds are my favourite) who give copious amounts of useful feedback and little gems of wisdom which you can write fondly about in papers. Teenagers are not so straightforward, it would seem. But I still think it is useful to work with them because it gives you a sense of what it really important to them, and it makes you aware of stumbling blocks to using your technology which you may not have anticipated.

It is important to us that teenage girls should enjoy using the fitness game, because in Scotland girls between 12 and 16 are at particular risk from obesity. We would like to encourage them to exercise more in an enjoyable way, which is one half of the battle in fighting weight gain (the other half being caloric intake). However, it is an incredible design challenge because body image is such a sensitive issue among teenage girls. Actually, based on today's session, everything is a sensitive issue if you are a teenage girl. Standing up in front of 5 other friends to use a Wii Fit? Too embarrassing! Everyone would be watching! The horror! You'd think we had asked them to tango naked around the assembly hall. (We didn't. The ethics board wouldn't approve it). We finally managed to convince two girls to play a water skiiing game together. And then suddenly I realised what they had been making a fuss about. The girls were subject to altogether more negative commentary and patronising remarks about their performance in the game than the boys were. In fact, the girls did not comment on the boys' performance at all. I am not sure whether this is because the boys went first so the girls chose not to comment to avoid retaliation when their turn came, or whether the girls did not feel qualified to comment. At any rate, I don't think I would have wanted to play the game with such a critical audience. And these boys were quite sweet and gentle in comparison to their peers, their teacher assured me.

There also seemed to be an issue relating to image management. The girls claimed not to like games, apart from Tetris which they would play on their phones on the bus. They didn't want to be associated with gamers or gaming. Yet, when asked further, they were actually familiar with quite a range of games and admitted to enjoying them. During the session today they spontaneously exclaimed about bits they liked and were disappointed when the bell rang to signal the end of the lesson. So while they enjoyed playing games, they were not willing to admit it.

So designing with and for teenagers is hard. It seems to me that observation is as important as interview in terms of methodology because they may be reluctant to answer honestly in front of peers. Their behaviour may reveal their opinions more honestly, but in a way you could argue that is irrelevant because if they are reluctant to be seen to play with a piece of technology, then they will not use it willingly for long enough to enjoy it, or in this case increase their fitness. This project will be a challenge. But who says research is meant to be easy?

Standing in a windy playing field

I all windswept and exhausted after standing for 40 minutes watching a school PE class. The windsweptness comes from being outside, the exhaustion from Mr Bitey's nocturnal teething habits. I went with Andy the Model PhD student to a local school to interview a PE teacher about the location based fitness game for teenagers that he is designing. As luck would have it, the next lesson she was teaching involved a fitness test so we stayed to observe it. This was ideal as I really wanted a sense of what the target use group do during a normal fitness session. The answer is "suprisingly little".

On the one hand you have the government's guidelines about physical activity for teenagers where they should be exercising for 2 hours a week. And then you have the reality of how that works in school. (I know that it isn't only up to schools to take responsibility for fitness, but they are part of the jigsaw). Of a 50 minute lesson, the teacher told us that 5 mins might be spent in warm up, 5 in cool down and 10 minutes for vigorous activity. But she said that many of the kids would find it difficult to jog without stopping for 5 minutes at a time. It was interesting to observe a lesson, where even during the warm up time, the kids tended to stop exercising when the teacher turned her back to get something set up. They did a speed test which was 100% effort for 35m x3, with a recovery of 3 or 4 minutes inbetween. There was also (necessary) time for instruction, and time for changing clothes. The lack of time spent in activity during the lesson has various repercussions for our project!

It means that the mini games in our project can be much shorter than we thought. 5 minutes is a long time when you are on a playing field, I can testify. It also means that we might well find it difficult to get pre-test post-test fitness gains even after several weeks on the grounds that they won't be exercising enough. We may have to get data from the teacher on what rate of change one would normally expect.

I was interested to learn that the children set their own goals for improvement based on a variety of test results (such as the speed test we saw today). They compare their performances to normative data published online, identify strengths and weaknesses and set targets. This is actually pretty close to what we intend to do, but in game form. As we observed, the kids were already comparing their scores to each other and trying to beat each other, which is interesting for the motivational theories Andy the Model PhD student is looking at.

Another major design concern which cropped up is the safety aspect of the kids looking at screens while on the playing field. Running while looking at screen, falling, breaking neck? Not good.We'll probably have to come up with some audio game design at least for parts of it.

The next stage is to run some focus group sessions on early design ideas for the game with some teenagers. That should be interesting too. The teacher told us what they need to do for exercise; hopefully they will tell us how best to incorporate it in a game.