Many people like to be challenged. Not everyone and not all the time. But faced with an ‘appropriate challenge,’ people will rise to the occasion more often than not.
The amazing success of Angry Birds, Candy Crush, and the recently ‘self-exiled’ Flappy Birds proves that a sequence of increasingly difficult challenges, even with very modest graphics, can captivate huge numbers of people for hours and hours, and hours – struggling to ‘complete a challenge.’
Recent neuroscience research may provide part of the answer for these amazing feats of human focus and concentration. Part of the allure/addiction may come from the burst of a brain chemical, dopamine, that flows into the pleasure centers of our brain when we finally reach our goal. Research by Allan Reiss, a psychiatric researcher at Stanford, has shown, through brain imaging studies, that video game play activates the central pleasure circuit in the brain. That ‘ta-da’ moment when you finally complete a game level literally produces a ‘rush’ that people aim to achieve again … and again.
This influence of games over human behavior can be exploited to make millions of dollars, as King, Zynga, and Roxio have demonstrated. Or it can be used to help people respond to devastating illnesses like Parkinson’s disease.
We produced a series of physical therapy games designed to help people with Parkinson’s disease improve their balance and gait. We collaborated with some great folks at the School of Nursing at UCSF (Glenna Dowling, Marsha Melnick, and Judy Mastick) who had previously demonstrated that a specific set of exercises and movements could improve gait and balance for people with this disease.
Our goal was to design the games so that they would encourage the kinds of movements that Glenna’s and Marsha’s research had shown were effective in improving gait and balance. And like a good physical therapist, we wanted the games to adapt to each person’s ability as well as provide new challenges as the person improved at a particular skill.
For each game, we developed a ‘difficulty progression’ of game levels that would match the clinical goals of a therapist. Level 1 would be very easy – we wanted to avoid intimidating any players. Level 2 would be a little bit harder, Level 3 a bit more difficult, and so on. This would allow each player to find the level that had the ‘appropriate’ amount of difficulty. Not too easy to be boring; not too hard to be frustrating. Enough of a challenge so they had to ‘try’ to complete the level. It required some work.
We also designed the progression so that as the player/patient was attempting higher levels, they were also improving some aspect of their physical condition. For example, one game involved standing up and sitting down in a chair. This may not sound like much of an exercise but remember, these games were designed for people with Parkinson’s disease and getting their weight properly situated over their knees while attempting to stand can be a challenge.
In the game, we translated the player’s up/down motion into the up/down motion of the lever on an old-fashioned hand railcar. As the player repeatedly stood up and sat down, they propelled a rail car through a virtual scene.
We included a replayable tutorial for each of the games so that players could learn how to play the game without any pressure.
The first level, as mentioned before, was very easy. The player only needed to stand up and sit down 3 times. Level 2 required 5 repetitions. At level 3, we switched to a fixed duration of time with a minimum number of reps. We increased both the time and minimum reps for levels 4 and 5 as shown in the chart below.
“Bob, these patients have Parkinson’s,” exclaimed Judy Mastick, the lead nurse responsible for recruiting patients for the study as she reviewed the difficulty progression. “There is no way they are going to reach level 5.”
“Well, maybe you’re right Judy,” I responded. “But we’re not forcing them to attempt level 5. They can pick the level they want.”
“Ok, but I bet you they won’t get to level 5,” replied Judy.
“I’ll take that bet, loser buys lunch?” I returned, confidently.
We created the 3D graphics, programmed the games using the Unity engine, constructed a custom ‘sensor suit’ that was like a set of Wiimotes on steroids, and then distributed 26 laptops and sensor suits to the patients enrolled in the study. They would use the games for one hour, three times a week, for 12 weeks. We had connected the laptops to a server-side database so we could record and review how the patient/players were doing every session. Some of the patients didn’t like the ‘sensor suit’ and that affected how much they stuck with the program. Many others, afflicted by their disease, would miss a session or two.
But most of the patients continued on. We tracked the players progress as the study continued, our database collecting valuable data about the subjects’ game-playing behaviors. We began to see a familiar pattern in the highest game levels they attempted per session and whether they won or lost. Here’s a typical pattern of game levels for one game by one of the subjects.
Players would either get bored with their current level or curious about the next one and as a result ‘attempt’ a higher, harder level. And typically, they would lose. They would go back to their previous successful level for a session or two and then, undaunted, try the more challenging level again. This time or the next, they would usually win the ‘harder’ level. After settling in with their ‘new’ highest level, they would start to get itchy again. They wanted a bigger challenge. They would reach for the next rung on the ladder and even if they lost on their first attempt, they would come back and try again.
The key, I think, was that the difference in difficulty between the levels was small enough that there was a ‘reasonable’ chance that the player would succeed on the next level. No one is likely to take on an impossible challenge voluntarily. We believe, we hope, that we can win, otherwise, why would we try? So too with these patients. Despite their disease, despite the unfamiliar ‘sensor suit’, these patients took on new, harder challenges, over and over again because each level was just a little bit harder than the previous level and the patients wanted to be challenged.
After 12 weeks the results were in. More than half of the subjects in the study reached level 5. I won the bet with Judy (although I have yet to collect it). And because the higher levels involved more effort, the patients increased their amount of exercise, the number of ‘game movements’ they performed, by average of 280% over the course of the 12 weeks.
Build them a staircase and they’ll climb the highest walls.
We are currently seeking funding to convert this research prototype into a marketable Kinect™ -based game. Anyone interested in helping us with this goal, please contact us. We would love to get this product into the hands of the people who need it most: the people living and coping with Parkinson’s disease.