Thoughts on Movement Patterns

Iuliu-Cosmin Oniscu
7 min readOct 8, 2021

I've been talking a lot about the importance of molecular design patterns in level design.

(It's imperative to consider molecular structure when working with open-world content, various degrees of variety in thematic and gameplay); however, I have not talked about why working with them is essential.

What is the core reason behind molecular design works, and more importantly, why should you use it in your work?


Deterministic movement patterns

I've been reading a lot lately about determinism. We don't get to think about that enough; we tend to ignore the core reasons behind why things happen. This makes things "seem" relatively easy to fix or complain about and hard for people actually to set them.

This leads to somewhat "magical" interpretations of the world as if things happen spontaneously and without reason, and we, as participants, are nothing but surprised by their instant manifestation.

Things, however, don't just "manifest". There is a reason for them to exist, and regardless if we like it or not, everything is preceded by a cause.

Movement patterns are (like all things) deterministic as well. There are reasons why we move, how we move, where we move and when we move.

These reasons might be based on our will, ambition, etc. or could be independent of our choice, and could happen for a variety of other reasons: survival, boredom, muscle spasms, etc.

But regardless, they are all the effect of an anterior cause.

In the case of movement, the cause is a change in the state of a subject. When the subject transitions from a form of "Dwelling" to a state of "Action",.

To make it clear, we always seem to "move" from places of "low action" (where we might dwell) to areas of "high action" (where we might act) and then back into "low action".

This could also be a reversed pattern.

But regardless of how we frame it, we are experiencing a cycle.

This alternate between dwelling and acting (or reacting) leads to the generation of rhythm.

Continuos movement

Movement may or may not be continuous, but it is most undoubtedly linear. Linearity is how we get from one state to the other. We might try to argue that that's not the case in open-world games because you can take multiple paths, abandon some routes, get distracted, etc.

That might be correct when looking at the situation from a very macro perspective; however, that is inaccurate.

Yes, we do have multiple paths, distractions, etc. And we want them to be used to generate routes and ideas and plans and whatnot. We want the player's agency to go through the roof, but ultimately the path you choose will be linear.

You will go from A->B->C->D, but the difference is you get to define what A, B, C, D are. And that is what triggers the deterministic nature of things.

You will follow points in space in a linear fashion, even though they are not linearly positioned for you, mainly because we are constantly experiencing things.

Events preceded one another rather than happening spontaneously at the same time.

Breaking the chain of causality and allowing for things to happen spontaneously, simultaneously, without primary cause would be crazy, and that would be what we would call "magic".

Player Paths

All this materialises itself in the form of player paths, fictitious assumptions we (as designers) attempt to apply our cowboyistic "design experience" to.

Player movement can be interpreted generally either by "metric" assumptions or by game mechanic assumptions.

If a player can "jump", they might "jump" in spaces where the "metric" allows it.

Working from the bottom-up, you might avoid such spaces by reflex (designer drives the experience).

In more realistic scenarios where "high functional art" meats "freedom of choice", the player paths will still be linear to some extent, but avoiding such spaces will become very hard (player-driven experience).

In scenarios where "many" paths become possible, and you face the genuine possibility of not determining where the player will be, we might be tempted to use "zones" as an abstract construct of where a player might be.

However, suppose you run this as a simulation and subsequently produce a heat map, either by raw data or syntax analysis. In that case, you will realise that despite the many possible variations that would result, the overlap consists of many linear paths created between critical points of interest.


A theory describes the amount of choice available in a level due to a player's reaction to "attractive" or "repulsive" things.

Mainly used to articulate how a player might interpret space in a stealth game, it seems to be something that we can apply to many avenues of level design.

Attractors and Repulsors can also be viewed as magnetic cores that push/pull the player away from the "player path".

In sandbox games, POI (points of interest) can act as both attractors or repulsors based on the oscillating needs of the player. Suppose you manage to build a system that alternately generates and drains player resources with a bit of care. In that case, it might turn into a never-ending cycle of challenge and reward, as the player depletes their resource and then gets attracted to POI where such resources can be re-acquired.

Drain some of these POI entirely or give them a cool down, and then the priorities would shift dynamically.

Ex. If you play as a Woodcutter, you will leave your dealing to cut some trees, trim those trees, fill your lumberyard with that would return to the trees only to find them still cut, therefore being forced to look for other spots where the trees grow. Etc

Traversal is reactionary.

Ultimately traversal is reactionary. It's not necessarily dictated by the geometry that encapsulates it (that's a very pedestrian way of looking at it); instead, it's defined by the variety of attractors and repulsors that populate that space.

The reaction to those elements is what generates the path.

Thinking about how those elements dictate movement eventually leads to generating an agreeable path through the level.

It allows for those elements to exist while conserving player agency makes the player reactive rather than just active.

Building skill progression on top of that framework is how you ensure the player learns to play the game.

Movement Patterns

Primary goals will be prioritised first, and reaching them repeatedly will create the need for "Primary Routes".

Less important goals will allow for "Secondary Routes", "Thirtiarry routes and so on.

This leads us to the core of the issue.

Urban or Civilised movement can be classified as either Organised or Disorganised. It can also be classified as Monocentric or Pluricentric.


Organised movement dictates that individuals will move between centres of interests linearly along the "primary or secondary routes".

Disorganised movement means that the movement will happen between the centres of interest, but paths can intersect without generating a point of interest. This generally leads to a more "disorganised" chaotic and challenging to interpret space.


Social spaces can be monocentric or pluricentric. A substantial monocentric space could lead to a sizeable identifiable intersection/landmark that would serve as an easy way to enhance orientation.

Pluricentric spaces could lead to disorientation if there is little distinction between the centres. Modifiers applied to such a framework would dictate what spots might be an attractor or a repulsor.

Clusters of similar centres can probably lead to visitors interpreting that space as a "district" of sorts where these kinds of "centres" live.

Excerpt from A. Bertaud (2001) Metropolis: A Measure of the Spatial Organization of 7 Large Cities.


I think it's a good idea to keep in mind these issues and not take the standard factoids that emerge from game development for granted. Always look at the bigger picture and try to anchor your design in parameters that make the player move rather than just building things so they can look pretty.

That's it for me! I will get Back into my lair for now.



Iuliu-Cosmin Oniscu

As a level designer I am not a creator, I am a facilitator. Senior Open World Designer. #Leveldesign #Open World #Design @notimetoulose.