Imageability and Spatial Awareness
Coming back to life
As the final days of development of the project I am working on are nearing, I find myself, once again, interested in the design aspects of my trade.
For the last maybe 2 years, various elements of the industrial nature of the game industry have eroded my passion for level design/world design.
I have been sitting in my corner of this industry, looking out and looking in, and I have felt like the time of the world designer/level designer is nearing its unavoidable end.
I came across an article from Liu Xinyan, discussing the differences between linear levels and open-world outposts, and it nudged me out of hibernation to write some thoughts.
A few notes struck my interest, specifically regarding the relationship between function and imageability of spaces based on their intended purpose.
The article proposes that, during the design process, depending on how linear or nonlinear a level might have to be, prioritizing function over imageability or imageability over function might be the way to go.
For the uninitiated, imageability is:
“Imageability is a measure of how easily a physical object, word, or environment will evoke a clear mental image in the mind of any person observing it.”
Imageability is something I have been trying to use extensively in my work, much to the confusion of my colleagues. It’s not a hard thing to grasp, if you read the book; however, nudging predefined conceptions is hard business and coming up with solutions to problems that are not immediately apparent might make people shake their heads in a funny way.
As I mentioned to Liu Xinyan in a message to his post, I am not 100% that prioritization of function over imageability or imageability over function based on the linear/non-linear nature of your level is the way to go all the time.
Sometimes you might have to do both at the same time.
I also think it’s a choice you make, as a designer, voluntarily or involuntarily, as the product of your insight or education. It is a cultural thing.
It is also a business thing; the nature of your project dictates what sort of space you need for your content, be it a box to put your candy in or an open world for your assassin to stab people in.
But never the less, besides context, the subject matter deserves further exploration.
For me, this is interesting because looking at levels from an imageability perspective does make understanding the process of building levels, that, regardless of how linear/non-linear they might be, would evoke a clear mental model in the player mind, making them akin to real spaces.
This equates to John Romero's rule number 7 for designing levels:
- Making my levels flow so the player will revisit areas several times, so they will better understand the 3D space of the level.
And this is the cornerstone of learning anything. Repetition of patterns.
If you look at old video games from the 90s, specifically FPS games (Doom Games, Quake, Thief, Deus Ex, etc), there is an obsession in their levels, with loops. This comes on the part of the designer to showcase the 3d space, sometimes as a direct contrast to 2d games of the time or from the previous time.
This was a big deal; going from Commander Keen to Wolfenstein to Doom to Quake was a huge tech jump. It was a huge jump in imageability.
Players could suddenly imagine those virtual spaces in 3d.
And as a side note I don’t think this is purely related to just 3d games.
Any game level that encourages the use of any kind of spatial loop (including 2d games) has the capacity for great imageability.
A special shout-out to Metroidvania Games, who start out linear, and by opening up sections of the map using lock and keys, transform the space, create loops and gives the space the capacity to be mentally modeled.
I think these days we take that stuff for granted, and the average consumer trivializes the feat of playing a 3d game due to the availability of such products.
It doesn’t help that the amount of detail on screen and heavy reliance on UI forces you to stop looking at the 3d space and forces you to just follow icons on radars and other gadgets.
So, how do we improve imageability for the space we are making?
There are a bunch of factors that contribute:
- The shape of the edges constitutes the boundaries of the space.
- The depth of the field, the contrast between foreground and background.
- The amount of detail/environment, set dressing/canopy obscuring the edges
Being a level designer implies you have some skills in spatial awareness and putting those skills to work in 2d. Do you draw your layouts? You might also be the kind of person who draws other people's layouts from memory.
If you do, you might find this interesting.
Take a level:
- Disable all UI elements (Map Markers, Minimaps, etc.) if you can.
- Start in any location.
- What do you see?
- Start drawing what you see on a 2d map using simple lines. Focus on drawing the contours of the barriers and edges you see in front of you.
- Draw the landmarks you see in the distance. Notice that you might not be able to correctly judge the distance just by looking at them.
- Move a little further until your perspective changes
- Repeat the drawing process? Adjust your position on the map. Do you need to move the landmark? How does the shape of the edges and barriers in the environment get complemented and developed?
- Repeat this multiple times. How does the map you drew changed, how much iteration did you apply to it.
What you will eventually end up with is a very utilitarian version of the level you are drawing that contains only the key elements you need, obtained from a purely third-person/experiential standpoint.
Make a level:
- Determine your level parameters and intentions
- Build it in 3d
- Repeat the exercise above, but apply it to your level
- Notice how some of the elements you put in there are not easy to understand when analysing the level with the above method? How can you fix it? What can you simplify to make it easier to fix?
Did you get it?
This is a form of cartography specifically focusing on drawing what you see.
It is a form of virtual digital cartography, but in a sense, it serves the purpose of the designer, specifically to better understand the space that is being built, identify where the problems are and then iterate and make a better, easier to understand space.
Of course, we don’t have the tools to make complex tools, but we have eyes and can calculate distances and make judgement calls, and decisions on what we see and where it goes on the map.
If you are interested in imageability as an urban planning technique, considering the above-mentioned applicability to the design of levels, please read The Image of the City by Kevin A. Lynch.
What do you think of imageability?