This topic will probably be one of the more ambitious topics for a number of reasons. First of all emotions are not just feeling excited about playing that new game you bought today or feeling sad because your favorite character in game of thrones just got killed. It’s very closely related to longer lasting moods as well. Secondly, psychologists aren’t completely sure on how to explain human emotions. Like many topics in psychology there are a number of different theories that explain what happens when we experience an emotion. Many of these theories are well supported by scientific studies. I won’t be going into these theories because I don’t think they are relevant to this article. Here is a link to crash course psychology just in case you’d like to know about emotions in general.
So what is an emotion? And more importantly why should you take them into account when you design and develop games? Emotions are a bit ambiguous, even psychologists can’t agree on a unified definition. One of the definitions I found: ‘an emotion is an internal response to an event’. Something within your body changes when you experience an emotion. This can be an increase or decrease in your heart rate for example. Other psychologists say an emotion is more like a feeling or mood. From these definitions it feels as if emotions aren’t very tangible and difficult to study. However, specific emotions and moods can be very useful when designing games and are easy to study. Taking emotions into account when designing games can definitely help you to enhance the player’s experience. Although the topic of emotions is an ambitious and broad topic, it also means there are countless ways you can apply it in your game design.
Just like there are multiple theories of emotions, there are several models to classify them as well. I will keep to one: the picture at the right is Russell’s model of affect (Russell, 1980). This is a two dimensional model in which emotions are classified based on how active (level of arousal) and pleasant (positive or negative) an emotion is. Many action games use the model to some extent. When you feel your heart pounding in your chest, your arousal is up. You might feel stressed and tense as you approach the enemy camp. On Russell’s model this would be high arousal and a sort of negative emotion.
Why should you apply all this to your game? Here are a number of reasons:
Even negative emotion such frustration can improve your game. It can motivate your player applied well. Remember when you fought an end-boss in a game but lost? What did you do? Did you quit the game or did you go back to the last save and try again? Most games have a difficulty curve of some form to keep players challenged and when the curve is just right, you will occasionally loose and have to try again. This trial-and-error will come with a bit of frustration but quickly changes to excitement and motivates you to try again. Frustration in these situations only become a problem when the difficulty curve is too steep and the player gets stuck somewhere in your game. It that case they might even quit all together which is not very good for your retention. Of course there should also be a moment of joy when the player finally overcomes an obstacle to make all the effort feel rewarding.
Be careful with too much frustration and confusion. It’s never good when your players become frustrated because they can’t figure out how the controls work, how to read the UI of your game or don’t know what to do. Obviously you need to address this kind of frustration and figure out how to minimize it. Unfortunately, it’s not always possible to get rid of the bad kind of frustration in your game for all players. Not all players are the same and for some the difficulty curve might be a little on the steep side. Some players will always be a bit frustrated about your UI. In those cases you can benefit from the Halo effect (Nisbett & Wilson, 1977): certain salient characteristics bias the perception of other less salient characteristics. It’s not about getting rid of frustration all together. Make the desired emotions stand out more and the player will focus on them more.
You can apply the knowledge about emotions in your game design regardless of the genre. However, I’d like to show you some examples for narrative and puzzle games. Puzzle games are all about frustration, confusion and joy. The halo effect is at work here: the joy of the eureka moment, when the player completes a puzzle is much more salient than the frustration and confusion from the trail-and-error process. Puzzle games are a great example of the good kind of frustration I talked about before. A great example of a puzzle game that uses the good kind of confusion and frustration is Anti-chamber. The player is told very little when they start the game. The game is to figure out the game (game-ception!). It can be great example if you want to make a puzzle game without a tutorial that takes the player by the hand each step of the way.
Narrative games probably are the best type of games to evoke emotions in players. When done right, your player will have a memorable experience of an emotional journey. As I talked about before emotions help form memories. There is nothing better than remembering the joy you felt when you helped your character do something amazing. Narrative games can allow players to really empathize with characters when something truly sad happens. My favorite example for such a game is Thomas was alone. The emotional narration makes it such a memorable journey. The designers did a great job expressing a full range of passive emotions such as sadness, happiness and serenity. Everything within the design of the game supports these emotions: the choice of the abstract art style, music and the way it is narrated. I’ve never felt so much empathy towards any video game character as I did for Thomas and his friends (and they are just colored squares!).
Some tips and examples
How could you implement all this knowledge into your game or narrative design? It seems like a lot of stuff to take into account but it all depends on your game. A good place to start is to identify the overall feeling or mood you want the player to get when they play your game. Ask yourself: how should the player feel after each session? What about when they finish your game? Maybe your game has some key-events where you want the player to feel a certain way. Of course your game design document describes how players should interact with your game but why not add a section on how they should feel at certain moments?
Playtesting is where you find out if players experience the intended emotions. Set your playtests up in such a way that you can either see or film the play-tester’s face directly. To decode all the different emotions you can use the coding system for facial emotions (FACS) developed by Ekman and Friesen (1978). You can also use software to decode even the subtlest emotions. There is a huge range of apps, software and even APIs and SDKs to use such as EmoVu (http://emovu.com/e/).
When you don’t have the money for these tools, time to get familiar with FACS or you want to be more thorough with your playtests, you can use PANAS (Watson, Clark, Tellegen, 1988). PANAS is a questionnaire where your play-testers answer questions on how much they experience a certain emotion. The picture at the right is a good example of what a PANAS questionnaire can look like. With PANAS you can find out which emotions the player experienced during the game or during key-events in your game. It will be a bit time-consuming to set up but once you’ve created one you can use it for all future games. There is a link to a PANAS worksheet in the references below to help you get started.
Some useful links and references