Search
  • Pål Schakonat

Our brain tricks us into making worse games! Our brain is tricky, the consciousness wants to be the center of attention and want to put itself in the spotlight. When designing games your conscious mind takes up most of the attention while your subconscious mind is left in the dark. The problem here is that when playing games the subconscious mind is doing most of the work for the player. As a designer (of any kind) I believe it is your job to be aware of the inner workings of the player and your own, the subconscious mind, just as much as the player´s conscious thoughts. To make this easier to understand I will try to visualize how the subconscious and conscious mind work and the relationship between the two. I will shine a light on the dark side of our brain.

Consciousness and subconsciousness is a way to divide functionality

The concept of consciousness and subconsciousness is a way of dividing the functionality in the brain. The conscious is a generic term for all the functionality that we feel we have control over, like comparison, critical thinking and allocate attention. The subconscious is a generic term for all the functionality that we don’t feel we have agency over, like reflexes, balance and recognition of objects. Another thing separating the two groups is that the subconsciousness work much faster than the consciousness.

You underestimate your subconsciousness in favour of your consciousness

The problem is that when you think of what defines you, you only think about the actions that your conscious brain performs. When you perform a subconscious action like, say, flicking your arm to wave away a fly, you never think “Ah, that’s me all right, flicking away a fly”. But when you choose clothes for a party, you could definitely see those actions representing yourself as a person. We see thoughts that we believe defines us to be more important than something as banal as flicking away a fly. This is a common pitfall when we play and design games. Coming up with a design feature for our conscious mind is a lot easier than creating one for our subconscious mind. For example, it is easier to implement a shop system than it is fine-tuned design for enemy behaviour. Why? Well, because a shop system is a game system for the conscious brain – it allows the player to make conscious decisions in the game world, like choosing their next weapon and spending currency. You give the player agency. Designing enemy behaviour on the other hand is less intuitive, because you can only base your design decisions on your own conscious perception of the gameplay. Many have tried to visualise how to design game for the subconscious experience, often used words such as “game feel”, “This feels right” or “juice it!” I hope my contribution to this topic will help to easier understands these concepts.





The fast subconsciousness

Take a look at this experiment (Kahneman, Daniel, 2011): Two pictures were presented to people laying down in a brain scanner. Each picture was shown for less than 0.02 seconds and was immediately masked by visual noise –  a random display of dark and bright squares. When the researchers looked at the scans the brain activity blossomed up when the participants “saw” the eyes. The interesting part was that when they asked the participants if they have seen the eyes, every participant said no, despite the obvious brain activity. The participants never realized they had seen the pictures of the eyes because it was only registered in the subconscious part of their brains. This tells us two things: We can register the information from the environment around us even if it’s visible for only a few milliseconds, and there is a lot of brain activity out of reach for our conscious thoughts. One report I came across during my research said that we unconsciously process approximately 11 million bits per second of information while only consciously processing 50 bits per second (Wiliam, Dylan, 2006). That’s a relationship of 22 000 to 1. When we design player experiences we tend to priorities those 50 bits per second, while ignoring the other twenty-two thousand more bits. Think of the exceptionally fine-tuned experiences we would be able to build if we took all of them into account.

The slow consciousness


If we buy that we perceive things as fast as 0.02 seconds and only extract the most important information to the consciousness – how slow does that make our consciousness? One experiment can give us a rule of thumb for the speed of our consciousness (Soon, Chun Siong, 2008). The experiment was explained as followed:

“The subjects were asked to relax while fixating on the center of the screen where a stream of letters was presented. At some point, when they felt the urge to do so, they were to freely decide between one of two buttons, operated by the left and right index fingers, and press it immediately”.

Pressing one of the buttons is truly an action of our consciousness, there is nothing to react to. The participants choose when they want to press one of the buttons. The interesting part is that they found out that they could predict when the subject would do their action 1-2 seconds before the action were performed by looking at their brainwaves.




How Consciousness and Subconsciousness perceive Mario´s jump

A more game related example of how our conscious mind only works with a compressed projection of our perceived reality is Mario’s jump. When Mario jumps, it is perceived as a normal, evenly spaced jump. In actuality, Mario’s jump is tweaked to match the timing that an action needs to feel right. It’s uneven. It takes 18 frames to reach the top of the arc and only 10 frames to land. If the jump took an equal number of frames to reach the top as it took to land it wouldn’t look too different, but the game feel would change. If you compared them the players would prefer the uneven one, despite not really seeing what the difference between them was. This is because our subconscious brain perceives it much faster than our conscious brain, and therefore notices a change without really seeing it. In this case it’s the slight timing difference that changes the feel. By increasing the fall speed, you are minimising the time the player needs to wait before they can perform the next action. This shows how small details like changing a few frames matters. The problem here is that it’s hard to say why you would like the jump with a faster downfall, because consciously it will be experienced the same as the evenly spaced version. There is only a hunch that the fast downfall jump is better.

Now we start to get a feel of how our subconscious operates. The subconscious analyses the game frame by frame, while it compromises everything that happens during around one second and sends that information to our consciousness. Mario´s jump will be observed every frame subconsciously and consciously it will get the compressed to a regular jump. This also tells us that we not only think our conscious actions are more important, they will also only be based on compromised information. To trust your “gut feel” can a lot of the times be better because it is based on all the information the subconscious picks up. To repeat the numbers, the subconscious can perceive things as fast as 0.02 seconds and the consciousness perceive things within 1-2 seconds.

One last thing, The Reaction time

There is one last variable to get a good understanding of the subconscious game design concept – our reaction time. The best way to test this is to get interactable! Click the link to test your reaction time:

https://www.mathsisfun.com/games/reaction-time.html

You probably performed around 0.275 second. What happens inside your brain during these milliseconds is a long chain of operations: You will perceive the environment, identify a particular object of interest (the shift in color in this case), determine what muscles are needed to press the mouse button and then perform the action. This is our subconscious reaction time. By knowing that our conscious perception is around 1 seconds we can design a game were the player is able to perform multiple subconscious reactions before being aware of it. We can also rely on the reaction time when we design our games. As an example, I like to put my cooldowns (and almost all game feel related variables) at 0.2 seconds. This creates an impression of not being able to spam the action but still allows the player to perform a new action towards a new goal.

To sum up:

The player consciously perceives things at around 1 second, react within 0.2 seconds and subconsciously perceives things down to 0.02. Instead of relying on the abstract concept of gut feel we can use these concrete numbers when designing for the players subconscious. A good practice when you design your game is to think that the game is running in slow-motion and you zoomed in on the specific event. When designing a jump, the three frames of anticipation, slight shift in the character´s stance, the effects when the feet leaves the ground and the hair becoming weightless as the character changes their trajectory will all be noticed and make a huge difference despite them only being flashes of movement. Same goes for every other design decision in your game. Every detail matters.



Talk about subconscious game design

“We should have a one frame freeze when the enemy is hit, the player will notice unconsciously”

“We should add a three-frame outline of the explosion to highlight the damage area, the player will notice unconsciously”

“How does that laser beam look in slow motion? It should look good in slow motion”



After match:


I recommend seeing the GDC talk from Martin Jonasson and Petri Purho – juice it or lose it with subconsciousness in mind and that the details they add will be perceive less important for the conscious but the subconscious will be very satisfied.

https://www.gdcvault.com/play/1016487/Juice-It-or-Lose

Wiliam, Dylan. “The half‐second delay: what follows?.” Pedagogy, Culture & Society 14.01 (2006): 71-81.

Kahneman, Daniel. Thinking, fast and slow. Macmillan, 2011.

Soon, Chun Siong, et al. “Unconscious determinants of free decisions in the human brain.” Nature neuroscience 11.5 (2008): 543-545

Wong, Aaron L., Adrian M. Haith, and John W. Krakauer. “Motor planning.” The Neuroscientist 21.4 (2015): 385-398.



Thanks to all my proof readers and feedback (Sebastian Larsson <3). A special thanks for Elsa Varland for a complete rewrite of the article.

  • Pål Schakonat

Game design is too complicated. My experience of game design theories is that they add too many unnecessary components for them to be used in practice. They become too advanced to grasp or you end up sinking too much time analysing the game instead of making the game. I studied game design at Gotland University, and although I learned to understand games better, I never use game design theory as a tool when I make games. Instead I have built up my own guidelines that I use. A red thread through all my theories is that they are based on non-game design literature.

I call this theory of Goal Oriented Design. It is a cut from the rules Csikszentmihalyi suggest setting the human in a Flow state (Csikszentmihalyi, 1990), then extend it through more research. It´s built upon self-acclaimed goals and feedback systems. It makes you ask yourself “What is the current goal of the player?”, and check if the game is supporting that goal at that given moment. But first, let´s see how we humans handle movement. You will get why this is important soon, so bear with me.

How we handle movment with Optimal Feedback Control

Movement is complicated. We have more than 600 skeletal muscles in our body, and every muscle can contract themselves on different levels. That combined with all the things we can do with the muscles, the possibilities become almost infinite. To figure out what to do is a very tiresome process for our brain. Rather than constantly calculating what our muscles should do out of these endless possibilities, our brain spreads out its computations by the use of goals (Wong, Aaron L., Adrian M. Haith, and John W. Krakauer. 2015).  


The latest theory in motor control is called Optimal feedback control (Todorov, Emanuel, and Michael I. Jordan, 2002). It´s based on that we set up goals. Example of goals can be: grab a coffee cup, move the character away from an attack or press that arrow key. Setting up a goal takes around 0.2 seconds, this is the time before we start to move our muscles, in other words, the reaction time. During this short period, we analyse our environment and figure out what is the most important thing to do. In other words, we are figuring out a goal. Like taking a gasp of air before we can change our actions.

To further decrease the processing power used by our brain, the brain only gives us a rough sketch on the action we should do to complete its goal. If we have the goal to grab a cup of coffee, instead of calculating how and when we should bend our muscle to make a perfect cup grab; we use our previous knowledge on how we grab coffee cups and do similar actions.  Then we set up a feedback system to measure progress on how our action is performing. If we want to reach for a lovely cup of coffee, our body sends out it´s hand towards the cup as we done before, and we only correct our movement if we deviate from the path towards the cup. We get positive feedback if we have a good trajectory and bad feedback if we deviate. This also helps if there is something unexpected in the way. If we would make a perfect cup grab, were we calculate every muscle move at the same time and we did´t account for the strong wind in our calculations, our hand would miss the cup because of the force from the wind. But with an optimal feedback control we are able to recalculate the trajectory of our hand movement with the wind force in mind, but only if we need to, all the way to the coffee cup.

I think one of the best examples is crumble a flat paper to a ball. The goal is the paper ball and our rough sketch is squeezing the paper where it sticks out. Then we sense with the palms of our hand if the paper sticks out and we press a little bit harder on that spot, then repeat this until we complete our goal of making a flat paper in to a round ball. We control our movements with goals and feedback. That´s the foundation of Optimal feedback control. This help us understand the foundation on what Goal oriented design is. Self-acclaimed goals and a self-acclaimed feedback system. We as humans set up goals and ways to measure progress.


Setting up goals and feedback system´s in games

A good example to describe goal-oriented design in a game is Minecraft. Building in Minecraft, the player first their goal, let´s say a house. Then the player set up a feedback system of what matters to complete that goal. In this case it would be changing the space, so it looks like a house. Then you start to progress towards the goal. Every time you put out a block your brain registrate that as progress and we release positive reward chemicals.

I think death pits, that is in the middle of the map flow, in generic deathmatch games is a good example of bad design from a goal-oriented perspective. I want to note that every game is unique, and that it may be a good reason to have a death pit in a deathmatch mode map. The goal for you is to shoot the enemy, you will set up a feedback system that supports shooting your opponents. You will shoot and if you miss you will revaluate your movement to hopefully hit the enemy next time. In the middle of this comes a more urgent threat, a death pit! The game is now presenting two goals where one leads to instant death and demands focus on precision platforming and one is tweaking your aim and macro movement, little on how exactly you are placing yourself on a platform. Either you will shift focus towards precision platforming or continue to shoot the player and potentially fall to your death. If this was a game built on risk management this could be good. Then the main goal would be to find the best chance for success, but now the main goal for the player is shooting the other player. In this case they will either fall to their death or get shot down by the other player. A problem with the death pit is that most times they are sporadically placed on the map. This means that most of that the player will foremost focus on shooting. Then suddenly a death pit appears that is more important than the main goal. Even if the two goals are conflicting in a second our reaction time is 0.2 seconds and we can registrate feedback much faster than that (See my post about subconscious game design). So even if it appears to be a small thing for our consciousness it´s a huge thing for our subconsciousness.

To summarize: think of what goals your game is signalling to the player. It can be a game that is very open for what the player set as a goal, like Minecraft or The Sims. Both of them still acknowledge that the player will set up goals and give feedback if they progress towards that. The game can also be more precise in what the player needs to do. Take the example of a classic corridor shooter, where all the feedback should support the goal of killing off enemies. That´s why many games like doom for example focus so much on the satisfaction of destroying enemies. I strongly recommend seeing the push forward GDC talk on the latest DOOM game, which I think is very well executed design, using the principle of a strong goal-oriented design process. It´s easier to remember a concept if it is put in context. I too put the goal-oriented thinking in sentences to help your brain remember.  

Talk about Goal oriented design.


“I think it´s okay to add random elements to our rouge-like. The goal of the player is to estimate the best choice”


“The player seems to like the raft building mechanic; the raft could sink down a little bit if you build heavy objects on it to show progress.”


“We can´t have explosive barrel in this game, this game is about using close combat, it will be to fun to shoot the barrels from a distance.” 


Links:

GDC talk: Embracing Forward combat in DOOM

https://www.youtube.com/watch?v=2KQNpQD8Ayo


Csikszentmihalyi, Mihaly. "Flow: The psychology of optimal performance." (1990)

Wong, Aaron L., Adrian M. Haith, and John W. Krakauer. "Motor planning." The Neuroscientist 21.4 (2015): 385-398.

Todorov, Emanuel, and Michael I. Jordan. "Optimal feedback control as a theory of motor coordination." Nature neuroscience5.11 (2002): 1226.