The Struggle of Predicting Players

So I recently decided to play through Signalis. The game was first brought to my attention by fellow blogger Frostilyte, who at the time (and perhaps still as of yet) had not played it, but their interest sparked my own. Part of my curiosity was the concept of an atmospheric horror experience, rather than a standard jump-scare horror experience. I’ve mentioned before my intense dislike for jump-scares.

Part of my interest was also driven by what we might call a sense of missing out. Some of the most prominent games in the survival horror genre – Resident Evil and Silent Hill – were games that I never played. While I had the consoles and therefore theoretical access to those games, they were never really on my radar until years and years after the fact.

And going into these older games comes with a lot of problems. Some part of the problem is just playing a game designed in an era where we were still learning design principles and experimenting with three-dimensional spaces. A while back I played through the original Resident Evil for the first time (or more appropriately the remake of it) and put down some thoughts about playing games from older eras. And one thing that certainly affected my experience was knowing what to expect. Not in the sense of knowing what every scare was supposed to be, but just in the sense of knowing the basic threads – zombies, a big mansion with weird puzzles, limited resources, and so on.

So playing through Resident Evil wasn’t exactly a fresh and new experience, even though it was my first time through the game. That’s why I was interested in Signalis – it would guarantee a fresh and new experience. I would finally get a chance to play one of these games before everything was revealed. Admittedly, I might not have needed to worry – Signalis might not reach the cultural heights that Resident Evil and Silent Hill had. But getting to play so close to the actual release date feels like taking part in some kind of ritual.

But in going through the game, there were a couple things I wanted to write about. Not really gripes, per se, but issues I ran into. I figure I’ll write a couple of short essays on those issues to help explain those problems in a bit more detail.

I was trying to think about how to approach writing these out, and the conclusion I reached is that I can’t really describe the problem adequately without revealing important information. So I will have to spoil some really interesting aspects of Signalis to make this whole thing make sense. If you’re a fan of the survival horror genre or even just interested in trying it out, I strongly recommend stopping here and playing the game first before reading further.

Psych!

Signalis borrows heavily from the sort of psychological horror that especially permeated Silent Hill’s writing. A lot of disturbing imagery, messing with the player’s sense of direction, and even the basic question of what is supposed to be real within the game’s world. These elements can often be the backbone of basic horror writing, but when meshed together well they make for a fascinating and gripping experience.

Part of the whole task of designing a horror game is making the player feel on edge, which means messing with the player’s expectations. If the player knows what’s going to happen or doesn’t need to worry about what will happen, then the player doesn’t have a reason to be scared. The player can be surprised, but not really frightened. And one way to mess with the player is to psych them out – to show them something that they expect, and then undermine that expectation.

So here’s where we get to the spoilery bit.

A good chunk of the way through the game, you’re given a fake ending. It seems like the character dies and fails in their mission, and you’re shown credits, and then taken back to a menu screen.

To start, this ending might feel disappointing to some, but for an indie game with a story and theming that deals a lot with failure, it also feels appropriate. Appropriate enough that you could just accept it as the real ending. After all, when you see the credits, that usually signals that you’re done with the game.

Whether you feel disappointed or not, you probably take this all as a signal that the game is done. You’re at the menu, and the menu looks different, but sometimes games change their menu screens after particular events. No big deal. So you hit the option to quit, and the game then starts back into the introductory sequence.

AHA! Expectation subverted. It’s an incredibly neat trick that puts you on edge.

And I almost missed it.

See, my first reaction was “I wonder what happens by starting a new game.” Because the option for playing the game is either “Begin” for starting a new game, or “Continue” for loading a save. Since the option said “Begin” my expectation was that I would actually be shown a brand new scene – my expectations would still be subverted, but the game would basically guide me to continuing.

Instead when I saw the starting cutscene and woke up in the level from the beginning of the game, my understanding was that it really was just starting a new game. I had no reason to explore around, because I’d just beat the game – I might replay it, but I didn’t need to do so now. So I quit out of the game.

At that point I decided to see if there were different endings to the game – because sometimes these games feature different endings based on how you play. And the first article I pulled up noted the fake ending. Had I not been curious, had there not been that expectation built into the genre, I would have completely missed out on the rest of the game.

Predicting Player Behavior

Now one could take this experience and say “the designers of the game messed up.”

I think that’s not quite the right conclusion.

Instead, I wanted to use this experience as a way of briefly exploring the inherent difficulty of crafting a game for players that have the ability to interact with the game as they choose. The problem of trying to build a narrative and surprises for players who could just miss them entirely.

A useful example on the opposite end of this is when a game directs the camera to some kind of vista or set piece. Players can often chafe at having control being taken away from them, arguing that if they want to see the thing, they can look at it themselves. I am sympathetic to the argument, but I think it misses out on two points. Firstly, that all of these things have had a lot of work put into them by people for the sake of the player, and they would like to have those things be appreciated: it’s a basic human desire that we all have. Secondly, that sometimes taking away control for a brief period is meant to help the player avoid missing something that could be important – it is a momentary frustration to avoid a bigger frustration later.

The obvious rejoinder would be to just design the game so that the player can’t miss these things. But at this point you are getting into the weeds of trying to perfectly predict how players will interact with the game. In other words, you are doing something impossible. Some subset of players won’t look at the vistas – even if they would find those vistas beautiful and impressive – because they’re just focused on the immediate tasks in front of them. Some players will miss important pieces of dialogue or events or puzzle pieces because they were paying attention to something else at the time.

In other words, there is literally only one way to make sure that a player can’t miss something important, and that is by taking away control from the player. There are plenty of techniques to help players see things organically, but no matter what some subset of players will just not see those things.

The problem seems especially pronounced in the horror genre, specifically because players derive a sense of safety from not having direct control. If you can’t move your character or the camera around, then you don’t have to worry about what’s going to happen next – you can’t mess up. There is a particular sense of anxiety that comes from a jump-scare that occurs because of something you did, as opposed to one that occurs because the camera whips around to show you something scary.

And so when Signalis had given players the freedom at that menu screen, they had done so predominately with the expectation that players would select the “Quit” option. Thrusting the player back into the game would immediately send the signal that something was up and the player should keep playing.

And when I hit “Begin” instead, I broke the developer’s expectation. Or maybe I did it when I decided I was being put back at the literal beginning, as it had been a couple days since I saw the introductory cutscene and couldn’t spot any immediate differences. Or maybe it was when I decided to just quit right then and there. Whatever it was, I missed something because of my choices. If the developer wanted to absolutely make sure that I didn’t miss something, the only solution would have been to take away my choice.

Perhaps we could say that the cutscene could have been more markedly different. Perhaps we could say that rather than “Begin,” the menu could have said “Continue.” Perhaps the game could have literally prevented you from quitting until you got to the next save point. Some of these solutions would work, some probably would break what the developers were trying to accomplish.

But arguing here over what could have been done misses the larger point: the fact that we as players are given choices, and we can make the wrong choices. We in some capacity need to be able to own those bad decisions. Sometimes we can lay blame on developers for poor design, but if we are never willing to take any blame ourselves then we engage in this discussion dishonestly.

Concluding Remarks

At the time of writing, I have not actually finished Signalis. I will probably finish it today or tomorrow, which will be several weeks before this actually gets published. I am eager to continue with the game and get to the end.

In looking up the endings and finding out about the fake ending, I now know about the alternative endings to the game, which makes me want to play the game even more. I’ll likely wind up streaming a playthrough to get the “best” ending just because I find the game so interesting, and it gives me an excuse to keep playing and get that ending (which I would most certainly never get on my first playthrough).

But this particular experience stuck with me in a specific way. The fact that I had almost missed out on maybe a third of the game got me thinking. At first I was unhappy, feeling like it was the developer’s fault. But in interrogating that feeling I am much less certain, and I see in my own behavior how I might have avoided the mistake.

And in a sense, I wonder just how much different my whole experience with the game could have been if I’d just hit “Quit” instead.