Fear without Jumpscares! — Key Insights into Horror Game Design
A look through the lens of emotional psychology to understand what makes a horror game good.
It’s been more than a decade since Amnesia: The Dark Descent took the world by storm.
At the time, it wouldn’t be an overstatement to say that Amnesia, singlehandedly, brought up the careers of many new-coming and established youtubers alike.
And since then, living in the shadow of such a breakaway success, we’ve had, on one hand, our share of heavy hitters such as Alien: Isolation, SOMA, Resident Evil, and The Forest; and on the other hand, we’ve had smaller-budgeted pieces such as Five Nights at Freddy’s, Hello Neighbour, Slender, and Poppy Playtime.
And while there’s no denying that there’s plenty of market for both categories, I’ve been meaning to explore what separates the lucky amateurs from the resoundingly successful pros.
Today, I’ll argue that the core distinction lies in their ability to manage their players’ emotional state, as determined by the VAD model of emotions (which I’ll explain further below).
To achieve this, I’ll slowly deconstruct two horror games (our case studies), and from there we’ll analyse what kinds of emotions each one is able to evoke, and how they manage to do so.
In this manner, it is my hope that this shall allow us, as developers, to elevate the quality of our work and learn from giants.
So let’s get started!
Oh wait, wait, wait… Hold on!
Before we dive headfirst into these two games I’ve prepared, please allow me to present to you a quick introduction to the tools we’ll use to deconstruct them.
Valence-Arousal-Dominance Model (VAD)
Psychologist are a smart bunch of people.
Many attempts have been made to understand human emotions analytically, and the Valence-Arousal-Dominance model (VAD. Also known as Pleasure-Arousal-Dominance, or PAD) is one of such attempts.
Developed by psychologists Albert Mehrabian and James A. Russell, the VAD model classifies emotions according to 3 distinct dimensions described by the 3 initials in the model’s name:
- Valence, or Pleasure, which stands for how pleasant or positive is the emotion for the person experiencing it.
- Arousal, which stands for how inclined to action, or energised, the person experiencing it becomes.
- And Dominance, which stands for how domineering or empowered, as opposed to how submissive or disempowered, the person experiencing it feels.
This classification allows us to sort emotions according to their characteristics, which facilitates drawing links between in-game experiences and their expected emotional outcome.
Side Note: There’s a resource made available freely for educational purposes by the NRC (National Research Council of Canada), which lists an extensive array of words and phrases along with their respective VAD scores (going from 0 to 1). And I’ll be referencing these scores for the remainder of this article.
So, if you get curious about the exact VAD scores, feel free to check it out.
Plutchik’s Wheel of Emotions
Robert Plutchik was a psychologist and professor emeritus at the Albert Einstein College of Medicine, as well as an adjunct professor at the University of South Florida. And one of his gifts to the world as a psychologist was this beautiful, colourful, and deceptively simple model of emotion.
His approach divided the human emotional experience into 8 distinct dimensions of emotion, that can be grouped into 4 axii:
- Joy — Sadness
- Anger — Fear
- Anticipation — Surprise
- Trust — Disgust
Which, when combined with each other, give rise to the whole array of emotions that we’re so familiar with, such as Love, Optimism, Awe, Contempt, and Remorse.
But do not be deceived by the simplicity of the image shown above. Plutchik’s model allows for all dimensions to interact, not only with their neighbouring dimensions, but also with non-adjacent ones too (even their opposites); such that they can give rise to emotions not shown on the image, but highly relevant to our discussion today, such as: Anxiety, Despair, Guilt, Shame, Pity, and Hope.
I won’t go into much more detail lest I risk losing you on the way.
So, for now, trust me whenever I tell you a set of primary emotions combine into a certain higher order one, just for the sake of keeping the discussion brief and moving.
Side Note: there is also another resource made available by the NRC, under the same terms as before, which offers a wide array of words and their emotional associations, broken down according to Plutchik’s model.
If this piques your interest, feel free to check it out.
(PS: No, I’m not associated with the NRC, haha. I just happened to come across this particular researcher’s work)
The Emotions of Horror
Instinctually, if anyone asked you what emotions are commonly associated with the horror genre, I have no doubt in my mind that most would answer “fear” as the one first and foremost of the lot.
And you’d most certainly be correct!
But fear is nothing more than the most basic component of horror, from which most other emotional experiences of the genre derive.
In actuality, the horror genre makes use of these 5 primary emotions the most:
- Fear ( V: Extremely Low — A: Very High — D: Very Low )
- Anticipation ( V: High — A: Neutral — D: High )
- Surprise ( V: Extremely High — A: Extremely High — D: Neutral )
- Disgust ( V: Extremely Low — A: Very High — D: Low )
- Sadness ( V: Extremely Low — A: Low — D: Very Low)
And they all pretty much have one single thread in common:
Disempowerment
Just take the classic jumpscare, the quintessential crutch of modern horror cinema and typical horror indie games.
When you analyse it using these tools, you come to realise that its effectiveness lies in fact that it abuses the emotion of anticipation, laced with a dose of fear in order to create anxiety in the player (anticipation + fear = anxiety), in preparation for the shock or surprise, that comes right after.
In other words, it takes the player from a state of neutral or rising arousal, and low valence, to a sudden and abrupt state of extreme arousal that quickly begins to subside right after the shock.
In this manner, the anticipation is used to build tension, and the surprise is used to release it.
But without proper disempowerment, a jumpscare is nothing more than a simple spook. Which is precisely the point were so many newbie developers fail.
Without disempowerment, a jumpscare conveys no fear. It only becomes an unpleasant surprise.
And so, in this sense, horror is nothing but the art of managing your audience’s emotional state. Swinging them between states of neutral and low valence (low-V), neutral and high arousal (high-A), while making sure to maintain a consistent state of disempowerment (low-D). Which, when put together, shall result in fear.
Our Case-Studies
Now that we have a framework we can use to analyse the flavours that create an experience, we can begin to deconstruct scenes into simpler parts that we can check for the emotions triggered.
But, before we begin, I feel obligated to disclose that… I am by no means a connoisseur of the horror genre. Rather, for the longest time I’ve avoided the genre altogether simply because I despise the lazy crutch that is the jumpscare, and how it has become such a staple of the horror genre. So be aware that the following analyses are gonna be highly-opinionated pieces.
That being said, now you know my bias, and have my tools. So feel free to make your own conclusions.
Five Nights at Freddy’s
Five Nights at Freddy’s (or FNAF, as is commonly abbreviated) has a fairly simple premise, and since essentially all of the action takes place within quite a reduced space, describing one scene mostly describes the rest of the game.
Though it needs to be said… The fact that you cannot move at all from your in-game station is worth highlighting.
It is quite a meaningful game design decision. But we shall get into why that’s the case in a bit, once I explain the game (just in case somehow you’ve never heard of it).
The game takes place within a security office in a restaurant famous for its animatronics. Your goal is to survive, using the tools at your disposal (security cameras, lights and doors) to keep the animatronics at bay until the end of your night shift.
As I said before, the choice to bind the player character to a single spot is meaningful from a game design standpoint, because it removes from the player the ability to flee.
The fight-or-flight instinct is quite central to every horror experience that hopes to evoke fear. And the subject of disempowerment is right at the centre of this issue.
Removing the choice of “flight” means that the player is essentially forced to face their fear. And when you don’t have the necessary tools to “fight” your foe, you’re forced to face your own disempowerment; and this produces anxiety. Because the player knows that the animatronics are out to get them (which creates anticipation), but does not know for sure when or how they’re going to get them (which heightens disempowerment and displeasure, and creates fear), nor can they flee as an option.
Being unable to flee in a disempowered context both amplifies fear and increases anticipation.
And since anxiety is the precondition necessary for using the classic crutch of horror gaming, then this leads us straight into FNAF’s central mechanic: The Jumpscare.
As I had been explaining before, the jumpscare is essentially composed by 2 parts: the anticipation (or setup), and the shock.
It’s not rocket science, and that’s precisely why it is so accessible and overused by newbie developers. Its simplicity is what makes it so appealing to a beginner.
The shock is the simplest of the two parts, conceptually speaking. And most beginners rely on the tried and tested technique of abruptly showing the face of the monster right at the centre of the screen, accompanied by loud sounds.
But the shock can also be achieved in other manners, since fundamentally speaking a shock in the context of horror is just surprise in a tense situation.
Darkwood shows a more nuanced take on this concept at a moment that those who have played it may remember.
Darkwood has a series of events that can happen in the middle of the night. And it turns out that one of those random events is simply: a knock on your door.
The knock can basically happen at any time during the night. And since the developers knew that their game is inherently tense (the night especially so), and keeps you in a heightened state of anticipation for extended periods of time (again, especially at night), then the knock occurring at random does not break at any point the prerequisite of needing anticipation before delivering the shock.
I believe this moment qualifies as a jumpscare at a deep level, despite not following the conventions that most indie titles seem to do. But I mention this moment, because I want to convey to you that as long as you hit the same emotional beats, you can deliver an experience that feels fresh while still rehashing a proven and tested technique in a more experimental way.
Back to FNAF, it is time to discuss how the game builds tension in preparation for the shock.
FNAF is a silent game for the most part, in that it only uses some ambient sounds to build atmosphere. But a lot of the heavy-lifting is done through its presentation and graphics.
The security camera does a great work in building tension and anticipation, by simply showing the “movements” of the animatronics, while avoiding literally showing them moving. That way, the player comes to the conclusion that the animatronics move while they’re not looking, which heightens anticipation and discomfort.
The style and presentation of what *is* shown on camera does the rest of the work. By using its unnerving graphics and VFXs, FNAF aims to heighten fear.
Adding the two together, you reach the anxiety state in which the following shock works best.
And so, that’s basically all you need to know about FNAF’s fear mechanics. It’s not much, but FNAF is a simple game.
Amnesia: The Dark Descent
Actually… Nope. Scratch that.
You’ll have to forgive me for not doing the obvious thing and bashing FNAF to death with some killer Amnesia Game Design from 2010. This article has gotten so long, it actually gave me time to stumble upon a very interesting indie title during the whole writing process that’ll probably serve our purposes much better.
Enter our surprise guest:
It Steals
A highly experimental horror title from 2020 by game developer Zeekers.
There’s a reason why I went out of my way to disregard Amnesia: The Dark Descent in favour of this one.
While nobody can argue that Amnesia is anything other than a masterpiece of its time, it’s been more than 10 years already since then. It’s time we move on.
I mean, of course there’s still merit to understanding how Amnesia did things in the past so we can learn from it for the future. But I really have a special place in my heart for projects that aim to push the boundaries of convention, in order to try something new.
It Steals is kind of a bundle of experimental projects unified by a single common setting and mechanic. In it, you’re dropped into a maze and made to collect some kind of blue light orbs, very much like Pacman, but in 3D and without as much wacka-wacka.
From there, the game has several game-modes, each featuring a different kind of monster with a distinctive behaviour for you to learn in order to succeed.
What really sets this game apart, is its masterful use of its wide array of tension-building tools.
Classic Mode is quite vanilla, featuring a monster that stalks you right out of sight, and vanishes once you flash it with your torch/flashlight. And the fact that the monster’s red eyes linger for a bit after it vanishes makes it all the more terrifying as it imbues you with the sense that you only barely managed to save your own skin. Its visuals really manage to ramp up the discomfort in you.
The game for the most part sticks to a fundamental tenant of horror:
“The longer you get to stare at a monster, the less scary it tends to become.”
It tries to keep the monster just out of sight, while also doing its damndest to make its threat be felt. And to that end, the sound design picks up a lot of the slack here.
As a horror game designer, you mustn’t ever allow your player to become domineering, or you’ll lose their fear.
And so, in this game you’ll hear the monster’s footsteps approaching heavily and rapidly, giving you just barely enough time to turn around and flash it in response, only barely managing to catch a glimpse as it vanishes again out of existence.
It is truly heart-attack-inducing. An excellent example of building intense anxiety that doesn’t let up. But since each game is really short in duration, this high intensity fits perfectly.
Then, we have Shutter Mode, which is an even more intense variant of Classic Mode.
Pretty much everything stays the same, except for the fact that now you have a camera flash instead of a torch/flashlight. Meaning that you only get to see a flash of the monster as it approaches, but the monster no longer disappears afterwards, as it instead gets stunned by the flash.
This means that the tempo gets even higher, and so does shift upwards the range of emotional arousal.
It forces you to keep moving with little rest, and the intense pressure you feel ramps up. And as the intense emotional arousal increases, so does the sensation and intensity of the fear.
Such overwhelming intensity would normally be ill-advised, as it can and will become extremely exhausting after an extended period of time.
But for quick and short games like this one, it is perfectly okay. Though it is still something to keep an eye out for.
Hide and Seek Mode, which is the last of the game-modes I’ll cover, does something quite interesting.
It’s premise is that the monster is hiding somewhere on the maze, and if you find it then it becomes its turn to find you (and kill you!).
This mode is TENSE.
While the other two modes we covered were high tempo, high intensity experiences (ergo, high arousal) at a nearly constant pace; this mode achieves an amazing balance between rest and intensity (i.e. a range between neutral and high arousal), which allows the player to bask in the highest anxiety yet.
The mode works in three distinct phases from the point of view of the monster:
- Hiding,
- Seeking,
- And Pursuit.
The game begins on its Hiding phase, in which the monster is hidden (but will teleport around when out of sight). In this phase it is completely harmless and will allow you to collect orbs unmolested; but if you accidentally spot it, then it enters the Seeking phase.
Upon entering its Seeking phase, the monster will begin to count down giving you a chance to flee and hide. Then, it will begin to seek you out. Stay too long on a single spot, and it will automatically find you. If it manages to find you, it will give you a couple seconds to flee as it enters its Pursuit phase.
If you somehow manage to outpace it as it chases you during its Pursuit phase, then it’ll return to the Hiding phase, and the cycle will continue.
This cycle is absolute genius.
The Hiding phase on this game-mode is the one that does the best job at utilising high degrees of anxiety and anticipation, in order to build an insane degree of tension.
You never know when you’ll turn a corner just to find the monster crouched, hoping for you to find it.
It gets icy cold once you spot it right out of the corner of your eye, and watch it stand up to chase after you.
In short, It Steals is a masterful neat little set of horror minigames that I can’t help but praise.
It uses light and shadow skilfully to disempower the player, its sound design to heighten arousal to extreme degrees, and its mechanics deliver a delightfully discomforting experience, that cannot be described as anything other than scary.
And there’s a lot to learn from it.
It has a good grasp of how to manage the emotional state of the player, and it does so in interesting and experimental ways that really do a lot for the genre. So I hope that if reached this point of the article, you’ll endeavour to push boundaries yourself.
Happy spooking!
Hey there! Before you go, I just wanted to tell you something.
I wrote this article in an effort to reflect about what makes a good horror game, and why for the longest time I’ve stayed away from the genre.
This came about as I finished polishing my entry into the Brackeys Game Jam 2022, which was themed “You’re not alone”.
I wanted to make an interesting experience in which you had to face the monster without outright realising it was the monster in the beginning. And as the realisation dawned on you, I hoped that the dread would also start climbing.
Horror’s a genre I had never attempted before, so I think I didn’t do that great a job on that entry. Though if my game still piques your interest, you can give it a try at:
You can drop me a comment over at Itch.io if you do give it a try, and you can let me know what you thought of the experience.
But regardless, if you liked this article, you can also follow me here on Medium so you never miss out on my future insights.
See you soon!
