top of page

Alice in "Wonderland"


  • Role:

Programmer/System Designer

  • Platform:


  • Description:

It's a 2D horror & escape game which allows        players to use their eye sight to freeze the movement of the monster in the maze while trying to find the outlet.

  • Project Duration:

1 week

  • Assigned theme:

Fast prototyping a fun game within one week. ​

  • Contribution:

1. Designed the core mechanisms, including gazing at the monster to freeze it and limited eyesight.

2. Impletemented the eye-tracking interaction with Tobbi API.

3. Programmed core mechanisms, including enemy AI (using A* algorithm), monsters and main character animator.

Alice in "Wonderland"

Alice in "Wonderland"

Play Video

My thoughts on Tobii Eye Tracker

Tobii Eye Tracker is an intriguing interaction. It gives users a sense of using their minds to control things in the game.
Since moving sight is always faster than moving a mouse, I think eye tracker is really powerful in games filled with actions like searching and aiming.


Initial idea

Initially, we envisioned our game to be a horror game. The horrible feeling will mainly come from the oppression of the monster in our game.

When it comes to designing an interaction using eyesight, the first thing that comes into my mind is this SCP-173, an imaginary monster designed by netizens, which cannot move in direct eyesight.
Our team believes this mechanism is thrilling and efficient enough to build a small and entertaining project in merely one week.

Still need an objective

Now we have a basic game mechanism. The current problem is that players don't have a game objective. If the monster won't harm you when you are staring at it, the only thing that players need to do is keeping looking at the monster. Thus, we still need a reason for players to look at other places on the screen.

An outlet is a very natural answer to the question. If you are in the same room with such a horrible creature, you probably want to escape. So I think making the game map like a maze where players try to find the exit while keeping the monster away from them is reasonable and complete.


Problems with 2D

Great! Now we have a complete game mechanism. Due to time limitations, we decide to implement the mechanism in a 2D project with a top view. Now a new problem emerges: it's very easy to find the exit in a top-view 2D game since objects in the scene don't occlude each other.

I immediately think of the Pokemon series, which uses shadow to create mazes in 2D. Players' sight can illuminate a certain part of the map. So players have to search around, finding the exit while freezing the monster and keeping it away from the character. 

In this way, a tactic for our game is automatically established: players have to balance their decisions between staring at the monster and searching around the map; without either one of them, players cannot reach their goal.


Future improvements

  • Polishing level design to make players be quickly familiar with the game map. Currently, playtesters reflect that the game map is confusing, and they don't know where they are when moving on the map. The main reason is that the map doesn't provide any conspicuous elements(such as a huge statue standing in the middle of the dungeon) helping the player reference their position. 

Future improvements

  • Add a battery system to make the game more thrilling. The light circle controlled by the player's sight is actually emitted from the flashlight the character is holding. We can make a radius of the light circle slowly shrink and add a UI object on the screen showing the current flashlight's battery level. Also, batteries will be placed on the ground can be picked up and replenished flashlight's battery.

  • I love the battery system idea because currently, if players are patient enough, moving the character step by step, they will eventually achieve their goal, even if it takes a considerable amount of time. Since this is a horror game and supposed to be thrilling, I'm not too fond of a safe way for hacking the game to exist.

bottom of page