Sensor Data and Interactivity: Real Time vs Discrete

I need to organize some thoughts for an upcoming article. There is a lot of interest in using Internet of Things sensor data in my research group, and my work in persuasive computing is being linked to it rather heavily. There are multiple ways to categorize sensor data interactivity. I have lately been mostly considering a simple categorization: real time interactive and discrete interactivity.

In the former scenario (real time interaction), the end user application interacts with sensor data in a continuous fashion. This has the potential to create a feedback loop where a change in sensor data prompts action in the application, which in turn affects future sensor readings. For a very simple example, if a GPS navigator shows that the user is going the wrong way, the user can change their course in response, and therefore future GPS readings will be affected. This approach is technologically challenging, because it demands continuous connection between the application platform and the sensors. For instance, as of now, many physical activity meters have limited compatibility with mobile devices. If compatibility exists, it may be limited to only a certain mobile OS. One key agenda in Internet of Things research is enabling wider interoperability between sensors and user devices, but we are not there yet.

If we are able to guarantee real time interaction between sensors and applications, new opportunities open up for designing persuasive games. These opportunities arise because they give us the ability to include new kinds of activities as game actions. In the past, this has been achieved through specialized controllers and more recently with sensors built into devices (accelerometers in mobile phones, Microsoft Kinect etc.) Once sensors in everyday environments become more and more available, they also become more prominent as something to build a game on. All sorts of new physical activity games are rather obvious developments, but for creative designers all kinds of sensors can become game interfaces - whether for a persuasive purpose or simply harmless fun. Contemporary alternate reality games are just a glimpse of all sorts of crazy stuff that can be done in the future with IoT technology.

However, like I stated, we are not there yet. Still it is quite usual for persuasive applications or games to use sensor data in a more discrete fashion. Pedometer step counts are downloaded into an application periodically - and perhaps more importantly - after the fact. While performing the activity itself (walking), the user will not directly see how the application reacts. Therefore the user in turn cannot react to what the application is telling them. All interaction is therefore delayed. Furthermore, by the time the data is downloaded into an application, it has become just a number. Although it still represents activity, the activity itself took place in the past. From the perspective of using the application, we could substitute the data with a random number within the same range and observe no difference. Generating data is not part of the interaction between user and the application.

This is basically the scenario that I suggested the three axes for in my previous post. To reiterate: it highly resembles free-to-play games because something is brought into the game from the outside (i.e. real money in f2p games). No matter how much we try to dance around the issue, the fact remains that the game rewards the user based on the sensor data it receives from them. It is hard to obfuscate this fact due to the discrete nature of interaction. To alleviate the problem, in-game benefits granted by sensor data input should be carefully considered. For instance, the "rewards" should be thematically appropriate and function as gameplay or customization elements instead of direct measure of success. If the game is used to persuade players to become more fit, their in-game "fitness" should reflect improvements in their real life fitness. Powering up an avatar is one rather obvious example. This allows players to see virtual improvements in their virtual selves before real life improvements become perceptible.

In theory there is an upside to this more limited use of sensor data: because the input is just a number, a game developed for one purpose can rather effortlessly be fielded for another purpose. In reality however persuasive power is likely reduced if there is no thematic connection between activity and application. This concludes my random musings for now.

The Three Dimensions Model of Persuasive Game Design

Hi! It's been a very long while since I last updated this blog. Not at a whole lot has been happening to write about because the research has been kinda stuck. I have also been writing entries to that other blog. I've been doing a lot of reading though, and there should have been some more updates about that stuff. Maybe later. This time around I just want to put something on paper.

I have been following the gamification discussion mainly through Twitter, and it pains me that it not much has changed in the past two years. Scores, leaderboards, badges and whatnot are still considered perfectly valid. Although there's no entirely unanimous research against them, the studies are kind of piling up. Although it can be argued that studies supporting rewards are also piling up, the mere fact that the issue is highly controversial should make us pause. We're often working in domains where mistakes can be quite harmful and if a method is potentially damaging, its use should be seriously reconsidered. There's that, and of course there's also the fact that it's not really game design, as has been pointed out by actual game designers in the debate. Awesome people around the world are making "gamified" applications that go way beyond the simplistic approach, and for that I commend them.

Somewhat recently I ran into some advice that had been given to companies that provide well-being programs to other companies' employees. Sure enough, the standard issue gamification bullshit was all there was. Obviously I don't want to take this route because a) I am concerned about the potential risks of using rewards and b) doing so would degrade me as a researchers and designer. However it does turn out that coming up with game concepts while avoiding the obvious routes is freaking hard. To help me in this task, I came up with a three-dimensional model to which all design should adhere to. The basic idea of the model is to help every player being able to feel like they are in the game. For instance if the game is cooperative, everyone should be able to contribute. I generally prefer cooperative games in this domain anyway, because competition has a higher risk of dropping players from the game.

Generally speaking there should be three different axes available for players to affect the game's outcome. The first two axes are directly related to whichever behavior is being reinforced by the game. I use the terms long term progress and short term progress. The former shows how far the player has come since starting to play the game or how much progress has been cumulated from the very beginning; the latter follows more recent trends in behavior and should generally be normalized (e.g. to make amateurs and professionals able to compete, we compare who has the bigger relational increase in the behavior). These two axes form the two core dimensions of persuasive applications. They generally provide rich feedback with the purpose of making the user more aware of their own progress and thus more motivated. The third axis on the other hand is what I think is really required to make a persuasive application into a game: decision making.

Using this model, each player can affect the game in three different ways: improving their overall behavior over a long time span (progress); improving their rate of improvement (improvement); making meaningful decisions in the game (strategy). The model bears similarity to some free-to-play games. We can consider the behavioral axes as something that is bought with currency because it comes from outside the game (e.g. physical exercise measurements) but gives the player some advantage. The last axis reflects how well the player does with what they get from the first two axes. Although there is a similarity, in general the two behavioral axes should not be treated the same as in-game payment systems in f2ps. A game that aims for behavior change should allow these axes to have more impact on the game. Unlike f2p games where "pay to win" is frowned upon, a persuasive game should be e.g. "exercise to win".

It is also noteworthy that the two behavioral axes are interrelated in most scenarios. The higher a player is in the overall progress axis, the harder it is for them to keep improving their performance. It means that players who are "ahead" will be stronger on this axis while players who are "behind" will be stronger in the second axis. Over time, that strength in the second axis will gradually move over to this axis. Notice that these axes are different, which means they should have different effects in the game, making the "ahead" players stronger in one way and the "behind" players useful in another way. When done like this, the poor performers will feel motivated to improve their performance because it yields quick returns all the while the good performers won't feel cheated or punished for "doing too well".

Finally, the third axis does not necessarily serve any behavioral change function on its own. It can do so, if the gameplay is related to the behavioral change goals (e.g. requires knowledge of good habits). Its primary purpose is to make the entire thing more interesting - it's what actually makes it a game. Being good or bad on the third axis is not related to behavioral change goals of the game but simply the player's own interest in the game. It's the glue that actually makes the entire system work. The third axis needs to have significance because otherwise players are likely to lose interest in the game (i.e. it's really just a system for tracking progress - nothing wrong with that, but don't call it a game!). Regardless of your target audience, this is very likely the hardest axis to work with because it requires brilliant game design. However, your target audience is very likely to make it even harder.

So there you have it for now. This model is my design philosophy for behavioral change games and so far I can only tell that designing games that are just games is nowhere near as hard.

Two Approaches to Increasing Physical Activity

It's been a long while since I wrote something. For quite some time there were no topics to write about. Lately I have been doing a lot of research into persuasive technologies and pretty much anything that surrounds the issue. While there is still a lot more to look into, I'm starting to form a clear enough picture.The topic I'm about to delve into should have a familiar ring to it. However, this time around the perspective is more informed.

It starts with an article by Richard M. Ryan and Edward L. Deci, Intrinsic and Extrinsic Motivation: Classic Definitions and New Directions. We're hearing a lot of talk about intrinsic motivation from gamification... uh, actually, let's not go into terms. Just check this blog post by Chelsea Howe to get a reference point. In their taxonomy of human motivation Ryan and Deci place intrinsic motivation on the far right, opposed to amotivation in the far left. The implication is that intrinsically motivated actions are the most self-determined and the most likely to be carried out. This we already knew. What's interesting about this taxonomy is in the middle. Instead of considering extrinsic motivation as one big undesirable lump, it is shown to have different degrees of internalization. The leftmost type of extrinsic motivation is the antagonistic, reward-driven motivation called external regulation. It is the type of controlling relationship between the actor and the motivator that is criticized by e.g. Alfie Kohn in Punished by Rewards. 

Kohn's criticism also applies to the second type of extrinsic motivation called introjection. Actions in this stage of extrinsic motivation are driven by social approval. They are important for the person's ego. Although it is less externally controlled than pure external regulation it can still involve a controlling relationship between the source of approval and the actor. Praise is an ego-stroking reward that fits into this category. Or losing weight to avoid being mocked. In these two stages are dependent on external pressure. If the pressure goes away, motivation to do the activity goes away with it. People who have once been motivated through external rewards may fall below their base level once the rewards stop coming. In encouraging physical activity, it should be clear to us that this kind of motivation cannot be healthy, even if it works as long as rewards last. Behaviorists are not wrong, operant conditioning does indeed work - but important matters like physical activity should not be approached with a "whatever works" attitude.

However, there are two more types of extrinsic motivation where the source is to an increasing degree internal. What they have in common is that the activity's goal truly matters to the actor. Although they might not enjoy doing the activity in itself, they understand the importance of doing the activity because it progresses them towards the goal. Even better, they identify with the goal and it becomes a part of their ideal self ("I want to be fit to stay healthy" versus "being fit is a part of who I want to be"). The activity becomes valued by the actor and they feel they are doing it for themselves. In developing persuasive technology, this is where we should by playing ball. Of course it only works if the actor actually values the goal - if they do not, then things will be a lot more complicated. However, I consider that situation to be beyond what we can do with just technology.

So there are two fields where we can operate. The most common approach taken in HCI research is to aid users that are in the better half of extrinsic motivation - they have already established a goal of increasing their physical activity. In this domain, persuasive applications are helpers. It is their job to improve the actor's feelings of competence. There are a lot of ways to go about this. Simple technologies like pedometers alone fulfill one specific need of competence: progress feedback. Technology can quantify activity, allowing the actor to see how well they are doing. It goes up from there. Technologies can help to pick suitable short term goals. They can nudge. They can help connect with like-minded people who are going through the same things. The wealth of persuasive technologies that can be employed once the actor is at least somewhat motivated towards the big goal is huge.

The other field is where exergames operate. It's kind of like a shortcut. Because physical activity is a very broad term, there are a lot of specific activities within it. Exergames in particular aim to create a new activity that the actor might find fun to do. It is possible to jump from amotivation straight into intrinsic motivation if a suitable activity can be found. Dance Dance Revolution has been mentioned to death. Other popular examples include Geocaching. It's not just new activities though. Simply finding an enjoyable sport can do the trick. Technology can step in and help a person to find an activity they might actually like. This is best achieved by making it as effortless as possible to try out new activities. Exergames, especially mobile games, can have a very low barrier of entry (e.g. $1 for a mobile game that allows you to start immediately, or even free-to-play).

I believe that both of these approaches are correct. The first approach is suitable for people who are prepared to get into exercising but need a little push to start and a lot of pull to keep going. The second approach can work wonders with people who do not care about exercise and don't have any fitness goals. It is worth notice that these people can also get started with an exergame and realize that they actually have the competence to become fit if they keep going, which in turn allows them to assume a healthy exercise routine. Or they might want to start doing other exercise to become better at their game or sport. At this point we can combine methods from the first approach to help their training.

To summarize, there are two clearly distinct fields to play in regarding motivating physical activity. Both have their uses, and both have their unique challenges. Research has been done in both fields, but we're still waiting for a physical activity revolution to happen. In this field it seems rather typical that research prototypes do not turn into products, and commercial products are not necessarily up to date with their psychological research. There is still a lot more space for both research and new products in this domain. Oh, and physical activity is just one thing, there's more to come.

How the World Affects Games

Back to the topic of the world and games, it's now time to explore it the other way around. Previously it was concluded that games do indeed have a transforming effect outside the game itself. The term of the magic circle, introduced in Homo Ludens (Johan Huizinga), and explored in detail from the perspective of games in Rules of Play (Salen & Zimmerman), is important in the analysis. As already stated by Salen & Zimmerman, the magic circle blurs in certain gaming activities such as live action roleplaying. This blurring has two directions, games affecting the world, and the world affecting games.

What we mean by the blurring of the magic circle is that: 1) the game is no longer entirely determined by its rules alone but rather affected by external influences and 2) the game's influence is no longer limited to its participants and the area of the magic circle. When we were discussing how the meaning of places changes for geocachers, we were talking about the second point. Ditto for discussion on sub cultures emerging around games. This time we're going to talk about point one. It's quite convenient for me to start with something familiar: geocaching.

If any game is highly affected by the environment where it's played, geocaching fits that definition. Caches are hidden into the environment. The play experience of hiding a cache is defined almost entirely by the physical environment: what are the possible places to hide, are there any interesting terrain challenges available, etc. Similarly, the play experience of seeking a cache reflects these environmental influences. This is largely expandable to any game that takes place in the real world but especially true for games where the arena of play is not involved in the game's design. A live action roleplaying game that is bound to a certain area can be designed with full knowledge of the playing area. Geocaching as an overall game cannot.

If we get back to the magic circle, in the LARP example the magic circle is still effectively closed. The area is bound, the game is played within a certain time frame and people inside the playing area are all participants in the game. LARPs that are not bound to a certain area or time frame are different, and indeed they come closer to what we call alternate reality games. The concept of the magic circle becomes more blurred in these cases. In a sense, a geocacher is always inside the magic circle. However, it is clear that they are not constantly actively participating in the game. When I'm seeking a cache, I have the intention to play, and therefore I am in the game. When I do not have the intention to play, my actions nevertheless affect my performance in the game.

This might need some clarification. Suppose I am playing two games: one on my Playstation Portable, and then geocaching using my navigator. If I choose to travel to Helsinki on a weekend, I can take both of these games with me. However, when I power up my PSP and resume playing a game, the state of the game has not changed due to my being in Helsinki. Geocaching on the other hand has definitely changed: I now have access to caches hidden in Helsinki, but not to caches hidden in Oulu, even though that's where I "signed off" from the game. I have traveled inside the game's world without active participation. From theoretical perspective, when I turn off the PSP, the magic circle effectively goes away entirely; this is not so with geocaching. The circle persists even though I choose not to be actively inside it.

So what's the significance? This: since the magic circle is ubiquitous in alternative reality games (and similar), what follows is that any action can affect the game world. Location-aware gaming is just the tip of the iceberg. Link in data from physiological sensors, or even brainwaves. Suddenly it becomes possible to make a game of everything if we so desire. We can skip points, leaderboards and badges - we can do much better than that. The tech is not so far in the future either and we can start with what is available now.