I wrote Fatherhood partially in response to your (Greg Costikyan's) earlier essays on games-as-art. In particular, the "game's criticism" article strikes home to me. I wanted to make a game that withstood criticism by possessing more than a surface layer of messages. We've all seen or read over-analysis of games that read meaning that likely was never intended: support for communism read into Super Mario Brothers, for example. There are also games that make their agenda clear, Hidden Agenda as an example. Or games, that while the specific agenda might be unclear, are so simple that one is left no choice but to try and look for a higher reason to explain the existence of the game -- Passage falls into this category.
My goal with Fatherhood was to make a game with an agenda, but not let that agenda control the game. The Catch-22 I've put myself in is that I do not think it is necessarily right for me to explain my agenda -- that would seem to defeat the whole purpose of wrapping a game around it. But, then, what to do if the agenda is completely missed by those who play it? One interpretation is that the author is dead -- if the readers miss the agenda, the agenda does not exist. But, perhaps, it does exist, but is operating at a subconscious level not picked up readily?
All this text leads to this simple conclusion. I will explain what game mechanics I added that were motivated around achieving my underlying message, along with said message.
Goal of the Game: Transmit the feeling of "Fatherhood" to players. Since Fatherhood is hardly a simple emotion that could be captured in a seven day roguelike, I picked two simple aspects. One is discipline. The second is being a role model.
Discipline is a very difficult task. The implementation in Fatherhood is through three commands: praise, scold, and talk. My philosophy is inspired from two sources. A GDC presentation (I think in '99?) by a psychologist who trained inch-worms hammered home the point that positive reinforcement is superior to negative reinforcement. She also addressed the other question: why is this so counter-intuitive? The answer is that negative reinforcement produces immediate results while positive reinforcement doesn't. In Fatherhood, if you scold your children, they will immediately return to rock piling. If you praise them while rock piling, however, there is no immediate result -- they still will break off to play tag. However, repeated scolding will result in children that only briefly rock-pile before they revert to running around and playing tag -- they learn that this pattern gains them the most attention. Repeated praise will instead lock them into rock piling -- while they enjoy tag, they prefer the attention of their parent. The underlying game mechanic is intentionally simple. Children seek the activity that maximizes attention (or, if another child has received more attention than them, they ape that child's behavior). Both scolding and praise count as attention -- however, the attention is applied to the activity they were performing when the scold/praise was applied. Thus, scolding a tag playing child teaches them that tag will gain attention, while praising a rock piler teaches them that rock piling will gain attention.
The third form of communication, talking, is implemented in response to ... ack, I don't have the book on hand. In any case, the key insight was the comment that whenever you find yourself saying "How many times do I have to tell you?" you should have a red flag go up that you've already told them too many times. Children are quite capable of understanding the first time, but, if given the chance to feign ignorance, they'll happily lead you to talking for hours rationalizing and re-rationalizing your arguments. As such, the "talk" command is wired up to do nothing.
Now the second agenda: being a role model. There are many, obvious, positive aspects to being a role model. However, the tough part is the realization that they are always watching you. You only have to goof-off and portray an unsafe action once for them to pick up on it and emulate it themselves. This is governed by how the children behave around the natural dangers. By default, they do not travel very close to the water or fire - they will keep a fixed number of squares away. However, whenever they have line of sight to you, they will check how close *you* have gotten to the water and update their own safe distances to correspond. This means that if you walk next to the water the kids will also be willing to walk next to it, which can lead to drowning if the water decides to advance that tick.
I had hoped that the high-score mechanism would encourage people to try and win by saving more farmland/forest. In my
play testing, to do real aggressive stopping of the flood/fire you need the help of your children. Maybe I just need to make the goal of increasing the score more clear so people don't see it as necessarily complete when they merely stop the flood/fire.
