Studies have shown that exercise video games, or exergames, can benefit elderly people who sometimes have trouble getting enough daily physical activity. But few studies have dealt with the question of how to get older people to engage with video games in the first place.

"I can give you a stationary bike, but if you don't bike it, I doubt that you will get any benefits from it," said Annerieke Heuvelink, a researcher with Dutch research firm TNO, at the Games For Health conference this week in Boston. "And [recent] studies were all done in very structured settings. People either came into a lab, doing something three times a week, or they would go to a certain senior place regularly, and they all did it because they were told to do it."

Working with the Mayo Clinic at their Healthy Living Lab in Rochester, Minnesota, Heuvelink recently led a pilot of 19 seniors where they were given training and unstructured access to exergames. Although the sample size was very small, the study did yield some insights about what factors are correlated with engagement in exercise games for older people.

All the participants were shown how to play the game in two initial sessions during a three-week orientation period. Researchers also helped half of the participants to build a personalized avatar. The others were given default avatars. Then the games were made available to study participants for 12 weeks in a special room at the Charter House independent living facility. The study offered five different Kinect titles including sports and adventure games. Some didn't play at all after the initial orientation, while others played weekly and even continued to play after the study was over. Ultimately about half of the participants played on a regular basis.

The study found that the top predictor of participation was frequency of already playing computer games -- even if the computer games in question were, in Heuvelink's words, "basically Freecell and Freecell." The next biggest predictor was engagement with games in general, not just video games but also board games.

"My hunch is it's more about people that think their time is worth playing games. They have enough time and they want to spend it playing games because they enjoy it," she said.

The other three statistically significant predictors were how outgoing or social the person was, how good the person believed themselves to be at the games, and their score on a short physical performance battery -- that is, people who were already fit were more likely to play the games. There was no statistically significant difference due to age, gender, BMI or cognitive function (although conditions like Alzheimer's and dementia were exclusionary criteria for the study).

Anecdotally, Heuvelink found that many older people resonated with particular games from their past. One man who boxed on his college team became attached to Kinect boxing, even though it was one of the more strenuous games offered in the pilot. Heuvelink stressed that older people in the study chose to play these games, even though they weren't designed for them. She hopes that if a company tackles designing an exergame just for seniors, developers will draw from the data in this small pilot study.