Cognitive load theory, developed by educational psychologist John Sweller in the 1980s, proposes that human working memory has a finite capacity, and that learning — and by extension, problem-solving — is most effective when cognitive demands are managed within that capacity. The theory distinguishes between intrinsic load (the inherent complexity of the task), extraneous load (unnecessary complexity introduced by the way information is presented), and germane load (mental effort that contributes to schema formation and genuine understanding).
This framework is not commonly applied in game design literature, but it is one of the most useful lenses we have found for thinking about puzzle design for players with cognitive accessibility needs — and for making better puzzle games in general.
How Puzzle Games Routinely Exceed Working Memory Limits
A puzzle game makes cognitive demands on its players in at least three dimensions. The first is state-tracking: the player must hold in mind the current state of the puzzle (which paths have been activated, which pieces are in which positions, which conditions have been met). The second is rule recall: the player must remember the game's mechanics and what is and is not permitted. The third is goal orientation: the player must maintain a model of what a solution looks like and evaluate current states against that model.
These three demands are simultaneous, and for any puzzle of meaningful complexity, they can together exceed the capacity of working memory. This is not a bug — it is often the intentional challenge of a puzzle game. The problem arises when extraneous load — complexity that is not part of the puzzle challenge but is present due to interface or design choices — adds to that demand without contributing to the satisfaction of solving.
Common sources of extraneous cognitive load in puzzle games include:
- Unclear feedback on which actions have been taken (players must retain past actions in memory rather than reading current state from the screen)
- Overlapping or ambiguous visual elements that require effort to parse before puzzle-solving can begin
- Menu and navigation systems that interrupt the puzzle context and require cognitive context-switching to use
- Unclear win conditions — players cannot evaluate whether they are progressing because the goal is underspecified
- Poorly designed hint systems that introduce more information than the player needs, increasing rather than reducing load
What Camins Taught Us About State Externalization
The central challenge in designing Camins for cognitive accessibility was state externalization — ensuring that the puzzle state was always fully visible in the game environment, so that players did not need to hold it in working memory. This sounds obvious stated plainly, but it has significant design implications.
In an early version of Camins, path history was stored internally and not shown visually in a persistent way — players could see the path they were currently tracing but not the full history of where they had been. Playtesters with attention difficulties consistently reported losing track of their progress and feeling uncertain about what they had and had not explored. The cognitive cost was not the puzzle itself — it was the memory overhead of tracking state that the game should have been tracking on their behalf.
The redesign showed the complete traversal history at all times as a persistent light trace on the landscape. The information was always available to read rather than recall. This change reduced extraneous load significantly and made the puzzle logic itself — the intrinsic challenge — more accessible, because players were no longer allocating cognitive resources to compensating for the missing external representation.
Difficulty Gradients and Cognitive Accessibility
A standard approach to difficulty scaling in puzzle games is to increase the complexity of the puzzle itself — more elements, more rules, more interdependencies. This is intrinsic load scaling: the challenge of the puzzle increases. This is generally what players sign up for and is not inherently an accessibility problem.
The cognitive accessibility issue arises when difficulty scaling also increases extraneous load — when harder puzzles are also visually busier, have more ambiguous feedback, or require more active memory management. This compounds the accessibility barrier: players who manage fine on early puzzles encounter not just harder challenges but harder interfaces as the game progresses.
In Camins, we made a deliberate choice to keep the visual vocabulary consistent across difficulty levels. Harder puzzles have more paths, more decision points, more nuanced logic — but the same visual language applies throughout. The layout principles that reduce extraneous load in the early puzzles are present in the most complex ones. This is harder to design — adding visual complexity is often the easiest way to signal that a puzzle is more advanced — but it means that the challenge scales in cognitive terms on the dimension we intend, rather than across both intended and unintended dimensions.
The Hint System Problem
Hint systems are meant to reduce cognitive load when players are stuck. They often increase it instead. The reason is that a hint presented as additional information — "here is something you may have missed" — requires the player to integrate that information with their existing model of the puzzle state, which may be partially incorrect or incomplete. The cognitive demand of integration can be higher than the demand of continuing to think independently.
We're not saying hint systems are a bad idea — they are not, and they are important for cognitive accessibility when players are stuck in a genuinely unproductive loop. We're saying that the design of the hint system matters as much as the decision to include one.
The hint approach in Camins shows rather than tells. Rather than displaying a text hint or highlighting a specific element, the hint mode briefly animates a possible next step — a short visual suggestion of direction — and then returns the game to its normal state. The player can replay the hint animation if needed, but it does not persist on screen. This approach reduces integration overhead: the player receives a procedural suggestion rather than propositional information, which maps more directly onto the visual reasoning the puzzle requires.
Cognitive Fatigue and Session Design
A dimension of cognitive accessibility that receives less design attention than it deserves is cognitive fatigue — the cumulative depletion of executive function resources over a play session. Players with ADHD, certain processing differences, chronic fatigue conditions, or depression can experience significant fatigue within play sessions that would not fatigue neurotypical players under similar circumstances.
Session design is a partial response to this. Camins is structured so that each puzzle is complete in itself — there are no puzzles that require information from previous puzzles to solve, and progress is saved automatically between puzzles so that stopping mid-session has no cost. Players can play one puzzle and stop without penalty. They can return at any point and pick up from exactly where they were.
The undo system also serves a cognitive fatigue function. Making mistakes in a state of fatigue is more likely, and the emotional cost of mistakes when fatigued is higher. A generous undo system — three levels in Camins, each reversing a complete decision rather than a single step — reduces the penalty for fatigue-induced errors and makes it safe to keep playing without accumulated frustration.
These design choices do not make Camins easier in terms of its puzzle challenge. They reduce the overhead that surrounds the challenge, so that the cognitive resources players bring to the game can be spent on the thing that makes puzzle games worthwhile: the genuine satisfaction of figuring something out.