*This article was originally written by Japanese indie game creator
Daraneko, and translated by PLAYISM.
Now, we'll continue on.
The previous installment can be found here.
[h2]◆In Practice #1: What a game designer must consider[/h2]
Well, the super-long "Facts" section is finally over. Now let's use these facts to think about game design. First of all, as I mentioned earlier, there are two main things that a game designer must consider.
――――――――――――――――――――――――――――――――――――――
◆How to increase “game playing” behaviorThe game should have elements that "increase said behavior" using "positive reinforcement", "negative reinforcement", and "reversion". If this part of the game design is done half-assedly, there will be no reason for players to want to go on playing that game.
――――――――――――――――――――――――――――――――――――――
――――――――――――――――――――――――――――――――――――――
◆Eliminating elements that reduce “game playing” behavior as much as possibleThoroughly eliminate from the game any elements that can be considered "positive disinforcement", "negative disinforcement", or "extinction", and remove any elements that "reduce said behavior". Note that containing too many of these elements will make the game basically "unplayable" for a lot of players.
――――――――――――――――――――――――――――――――――――――
Simple. Now we need to consider what we should do in order to achieve these goals.
[h2]◆In Practice #2: How to create a “reinforcer”[/h2]
How do you make the “reinforcer” required quickly? The quickest way is to set up a goal and a means to achieve it. It's a real joy to find yourself moving closer to a goal you've set for yourself.
The so-called "token economy method" is also used in clinical settings in which behavior analysis is applied (for example, to improve truancy, etc.). It's like a point card, where if you accumulate a certain number of points (tokens), something good will happen to you, so you feel motivated to take the required actions to accumulate points.
These "tokens" aren't useful on their own, but if you accumulate enough of them, you will be rewarded. Like stuff like EXP and likability. Be sure to use these things well.
For example, when a goal is created, such as "get a likability rank of A in order to see an event," the "likability rank" becomes a token ("the reinforcer"). As shown in the figure below, the composition is somewhat hierarchical.
Now, when it's time to actually create the goal, make sure to make the "goal" a very specific goal. For example, a goal such as "increase the character's level to 10" is easier to aim for than just "increase the character's level".
Also, it is important to be specific and clear with regard to the "means to achieve the goal". If you want to get to level 10, but you don't know how to gain EXP, then your goal is meaningless. There is no way to achieve it.
In the case of Fire Emblem, which I mentioned earlier as an example, there are a lot of things hanging in the air when it comes to class changes.
――――――――――――――――――――――――――――――――――――――
◆Class change ・ Above a certain level ⇒ Reinforcer: EXP
・ Class-based skill level ⇒ Reinforcer: Skill EXP
・ Item used for class change ⇒ Reinforcer: Items, money
――――――――――――――――――――――――――――――――――――――
This is a good example of how various elements in a game can function as a “reinforcer” by having multiple small goals dangling from the larger goal of a class change. It's not just a matter of having a lot of them hanging around (because it gets complicated and confusing); it's easy to associate "class change" with "skill level", so it's good that the game isn't confusing even with a lot of elements to consider.
[h2]◆In Practice #3: Combining continuous and partial reinforcement[/h2]
Of course, you should use both continuous and partial reinforcement according to specific characteristics present, but it is also possible to use a combination of both. You can use continuous reinforcement to handle some parts and partial reinforcement for others.
For example, in Dragon Quest, you basically get EXP by defeating enemies (continuous reinforcement), but if you can defeat the occasional metal slime, you can get a lot of EXP (partial reinforcement).
In Monster Hunter, you basically get materials by defeating enemies (continuous reinforcement), but sometimes you can get very rare materials like plates and such (partial reinforcement).
So yeah. I tried to think of a good way to end this part, but I couldn’t come up with anything so, uh, I’m just gonna go ahead and move on to the next part. Here we go!
[h2]◆In Practice #4: How to avoid “reduction in behavior”[/h2]
I've already talked about extinction in "Fact #6 – Special Edition". In the last half of the game, it's easy to run out of “reinforcers” and face extinction, so you have to be careful.
So, what should we do to handle the disinforcement aspect?
――――――――――――――――――――――――――――――――――――――
◆Positive disinforcement (disinforcement through provision of aversive conditions) ⇒Acting causes “bad things to happen”, therefore behavior decreases
――――――――――――――――――――――――――――――――――――――
As for positive disinforcement, I guess I'd say that we should eliminate elements that cause unnecessary stress. For example, "loading screens that are too long", "slow movement for no reason", "encounter rates that are too high", "confusing UI", and so on. Even if each of them doesn't have that big of an impact alone, they all add up to make a bad game in the end. Fortunately this isn't as common in recent games, though.
In competitive games, it's better not to make "losing" into a big example of an “aversive condition”. If you keep losing, you will get sick of playing and quit. This is especially true for board games. It is quite stressful to play a game when you don't think you can win. On the other hand, if you can make the player think, "I lost, but at least I had fun," then they won't be so turned off.
If you want players to try to win, it's probably healthier and less likely to turn players off if you use reinforcement by increasing the rewards for winning, rather than increasing the penalties for losing and using disinforcement.
――――――――――――――――――――――――――――――――――――――
◆Negative disinforcement (disinforcement through removal of a reinforcer) ⇒Acting prevents “good things from happening”, therefore behavior decreases
――――――――――――――――――――――――――――――――――――――
About this one... Um, to be honest, I don’t really have many ideas on this...
I think it's best to "not do anything that would discourage the player". For example, a companion who is leaving the game permanently is holding some super valuable equipment, and that equipment is lost forever when they leave. Stuff like this.
If anyone out there has any good ideas, please quietly let me know (so this is on you guys now).
------
I mentioned that it's good to evaluate "how strongly the 'reinforcer' is functioning" with regard to reinforcement, and the same is true for disinforcement. It might be a good idea to evaluate whether there are any elements that could be considered "disinforcement" when playing the game.
[h2]◆In Practice #5: Why people play extremely hard games[/h2]
There are games that are popular despite their high difficulty, you know? A famous example would be Dark Souls, and a more recent one would be
Sekiro.
Why does everyone like to challenge themselves so much? I mean, you might not even be able to win. Even if you just want to enjoy the story, you can't because of the high difficulty, and extinction or disinforcement are likely to become issues due to the all the stuff that could be considered an “aversive condition”. It feels like it would be a no-win situation for the developer.
So, is everyone just a masochist or what...? Well maybe, but this isn't really a "behavior analysis"-type viewpoint. These games are actually popular, and I, too, like to play and create games with high difficulty levels. In science, reality always takes precedence. If this stuff is popular in reality, then it must have behavior contingencies that make it so.
The fact that the behavior is maintained without reduction means that there is either some form of reinforcement after defeat, or there is a strong resistance to extinction. As it turns out, it's both.
-----
First, let's start with the reinforcement when you are defeated. If I randomly try to think of something common in boss battles in difficult games, there's one thing that pops into mind right away: the boss enemy's HP bar.
This plays a surprisingly important role. When you play a difficult game, you may not be able to do very well at first, but the more you play, the better you get. The boss's HP bar shows you how much you have improved in an easy to understand way. Even if you keep getting killed over and over, you can see that you have reduced your enemy's HP more than before, and that you are getting better, so it is easy to see how much you have grown, which constitutes a type of “reinforcer”. Everyone looks at that HP bar, right?
In other words, in a game where you have to try again and again, it is important to make the player feel that he or she is getting better than before and has made progress even when defeated. This sense of growth becomes a type of “reinforcer”, and it reinforces the behavior even if the player loses. So even if you lose many times, you can continue to fight without giving up.
By the way, I'm not saying that you absolutely have to display the boss enemy's HP bar. You just need to know how far you've progressed, so it's good if you can feel your progress by seeing how your enemy looks when they're damaged, or how their dialogue changes.
The battle against Sans in Undertale is very well done, in that there is no HP indicator, but the dialogue changes according to the number of turns. It's a boss battle of unbelievable difficulty, but there are also a lot of other mechanisms to make the player want to retry, which is great. It's a good learning experience.
-----
Next, let's talk about strong extinction resistance. If there is a strong resistance to extinction, then there is partial reinforcement. What can act as a kind of “reinforcer” is a sense of accomplishment. You can't win every time, but the sense of accomplishment you get when you try and win again and again works as a strong example of a “reinforcer”. Or something like that. Oh, and just to be clear, that "sense of accomplishment" is just one example of many possible types of “reinforcers”.
Yes. Giving the player a sense of accomplishment. Sounds really good, doesn't it? Hahaha.
In other words, what game designers need to do is figure out how to train players through partial reinforcement. They pick up on the line at which the player feels "this is hard, but I can do it," and train the player to experience the joy of victory through hard work. This is how game designers create tough players who will not be daunted by defeat.
OK, so, how do we train players? A technique that is often used in conjunction with the token economy method is called "shaping". A rough explanation is as follows.
[h3]◆Shaping[/h3]
A method of reinforcing a specific behavior by starting with a task of low difficulty and gradually increasing the difficulty.Think of it as players having a "reinforcement level" where they challenge difficult games over and over again. The higher the level, the more difficult games they have played in the past, and the more they know the taste of "accomplishment".
Now, if you suddenly hit a person with a low reinforcement level with an enemy of a high difficulty level, their heart will break. They'll get knocked out. So, you need to hit them with relatively low-difficulty enemies first to raise the player's reinforcement level. Here's a diagram.
The only people who can endure something like Boss E out of nowhere are those who like very high-difficulty games. It would be a pretty niche game.
So, start with Boss A and gradually increase the difficulty level, and give players a sense of accomplishment when they defeat Boss B, Boss C, and so on. If you release Boss E when the players are properly trained, even if they only started off at reinforcement level 1, they will keep playing the game because they have progressed up to the 5th reinforcement level by that point. They know the taste of accomplishment.
By the way, this idea is not limited to enemies of high difficulty. It's a good idea to start with easy reinforcement and let the player get a taste of the difficulty curve, and then gradually make it more difficult, as it's an effective way to increase some kinds of behavior.
[h2]Wrap-up[/h2]
That's it for now. I feel like this was about ten times harder to write than my usual articles. By the way, the actual number of words is almost five times as high as usual...
Let’s wrap all this stuff up, shall we?
------------------------------------------------
[h3]Facts[/h3]
- The cause of a behavior lies “after the behavior”
- A “reinforcer” is the player’s reward
- An “aversive condition” is a negative thing that act as a penalty
- Behavior changes when reinforcers or aversive conditions appear or disappear after certain behaviors
- Behavior increases after said behavior ⇒ “reinforcement”
- Behavior decreases after said behavior⇒ “disinforcement”
- Reinforcement and disinforcement are not forever ⇒ “extinction”, “reversion”
- Partial reinforcement, which provides a random "reinforcer", is more resistant to extinction than continuous reinforcement, which provides a "reinforcer" every time
- It's best to release reinforcers or aversive conditions as soon as possible, but no later than 60 seconds after the behavior in question
------------------------------------------------
[h3]In Practice[/h3]
- Think about how to best reinforce, and how to avoid disinforcement
- The quickest way to create a “reinforcer” is to have a "goal" and clarify "what it takes to achieve the goal"
- Dangling multiple subgoals against a larger goal makes it easier for various systems to function
- It is also possible to consider a combination of continuous and partial reinforcement
- Eliminate unnecessary stress components that could cause disinforcement
- ”Fun, even when you lose” is the ideal
- On higher difficulty levels, make sure there is some kind of a “reinforcer” available in defeat
- It's important to be able to see that you're making more progress than before when retrying
- It's better to give the player a taste of what it's like to clear something difficult before making them try even harder things
------------------------------------------------
[h3]Special Edition[/h3]
- if you rank the reinforcements for each feature of your game to see whether they are working properly, it will help to highlight what you are missing
- It's also good to simultaneously rank the various elements to also see if there is anything that could cause disinforcement
- Note that if a type of “reinforcer” that worked in the early to middle stages of the game stops working in the later stages, there will be a decrease in motivation, often seen when players make it to the end and give up
- It's also important to know how to keep the “reinforcers” alive and well until the game is over
That’s a lot of stuff... I wonder if anyone has even read this far? I’m starting to worry, actually.
What I've written in the "In Practice" section is just what I would do if I were to apply my knowledge of behavior analysis, so you can probably use it in other ways if you want to. Please let me know if you think you can actually apply this stuff to design a good game.
Alright then. I hope this stuff helps somehow. That’s it for the wrap-up. Now for the extra stuff.
------------------------------------------------
[h2]References[/h2]
The information presented here is only the basic fundamentals of behavior analysis, so if you are interested in behavior analysis, please do feel free study up on it yourself. As long as you have the knowledge, the rest is up to how you use it.
By the way, the books I read to write this article are as follows (*Note: These titles are in Japanese and do not appear to have English translations, unfortunately):
------------------------------------------------
Koudou bunseki-gaku nyuumon: Hito no koudou no omoigakenai riyuu (Shueisha Shinsho)
This is a good introduction to behavior analysis. It explains the basics in an easy-to-understand manner.------------------------------------------------
Merit no housoku koudou bunseki-gaku: Jissen-hen (Shueisha Shinsho)
This one is a little more advanced than the book above. It includes many references from actual clinical cases.------------------------------------------------
For example, for partial reinforcement, there is research on "reinforcement schedules", which is about the proper timing to use for releasing a “reinforcer”, and whether different timing changes the properties of said reinforcer. If you're interested, this is good stuff to check out.
[h2]One last little problem (messing around)[/h2]
Alright, we're done with the wrap-up, so allow me to pose one final question. What is the “reinforcer” motivating me to write about game design theory? Do you know?
oh yeah – you know... right?
.
.
.
That’s right! “Likes” and “Retweets” are a great way to show that you've written a good article and managed to help game designers. Of course, direct feedback also works as a “reinforcer” to the extent that it tends to get me super hyped up for hours on end.
By the way, you can get more impressions from "retweets" than from "likes", so "retweeting" is sometimes stronger as a type of “reinforcer”. And no – this is just an explanation; nothing personal, OK? Hahaha.
―――
So, did you find this article helpful?
If you're interested in checking out some games that take advantage of the "game design that makes you want to play even more" stuff described in this article, then I've got just the game for you.
The Use of Life by Daraneko Games.
That’s right! Seeing as how the game was made by me – the writer of this article – it goes without saying that I put a lot of this stuff into the game.
The demo version is now available on Steam, so by all means go ahead and give it a try if you’re interested.
It’s a “game book-style” RPG in which the player-character’s various “attachments” change in accordance with the choices you make, with the story’s ending also changing ultimately depending on your “use of life”.
The Early Access version has been released. If you've played and enjoyed the demo, please be sure to check out the Early Access version, too!
Personally, I love seeing my games being added to more and more Wishlists, so if some of you readers out there were to go so far as to add The Use of Life to your Wishlists, that would be a nice little piece of the “goodness” for me.
I just might get motivated to write up some more of these game design articles.
https://store.steampowered.com/app/1483370/_/
