Playing strategic tabletop games requires that individuals think like a computer. It requires that players follow a set of relatively uncomplicated rules with a few (relatively circumscribed) decision points for which players have voluminous data. However, humans have a difficult time writing succinct rules, understanding rules, and/or enacting the rules of these games (Berland & Lee, 2010). As game designers and players know, writing effective instructions is a key (often overlooked) challenge of game production. Even with effective instructions, novice players of almost any strategic boardgame will undoubtedly make several rule errors, (mis)communicate their idiosyncratic understandings, and generally create a mess of even a very well designed game.

This essay details the ways in which the thinking involved in playing strategic boardgames relates to computational thinking, both in terms of the algorithms that players must learn, and the types of bugs they generate and fix. These parallels suggest that players of boardgames are engaged in computational thinking, without ever touching a computer. I therefore posit that strategic tabletop games are a productive way for humans to learn the basic elements of computational thinking—that tabletop games can be computational-thinking training machines.

Computation With Or Without Computers: Computational Literacy

The ability to use computation as a ‘lens’ for thinking and learning is called ‘computational’ or ‘procedural’ literacy.89 Papert (1980) used the term procedural literacy to propose that all students be taught to program and to use computer programming as a means to think and learn about math, science, and even literature. More recently, Bogost (2007) and Mateas (2008) have taken up the mantle of procedural literacy as a way to express how students learn to think with videogames, by virtually inhabiting their rule-based worlds. This work builds upon that work by adapting it to an analog space. Both the work on procedural literacy and this work stem, in part, from work by constructionists such as diSessa (2001); diSessa argues that computational literacy can be understood in terms of three core components: the social, the cognitive, and the material. Material computational literacy is being able to use the tools of computer science and computation to solve problems. At this point, boardgames are unhelpful in teaching material computational literacy – no boardgame on the market will teach you computer programming or modeling. However, games are much better suited to social and cognitive computational literacies. Cognitive computational literacy is being able to think computationally. Social computational literacy is being able to communicate about computation. This essay explores the ways that strategic tabletop games engender the cognitive and social aspects of computational (or procedural) literacy.

Cognitive Computational Literacy: Computational Thinking

Cognitive computation focuses on creating, understanding, following and debugging rules. It is fundamentally about the enactment of explicit sets of instructions in a specified order. We call these instruction sets algorithms. An example of an algorithm might be:

Roll the dice

If you roll a seven, move the robber and enact the robber rules.

Otherwise, deal resources matching the rolled number to the appropriate player

The player who rolled can now enact one of three rule sets: building, trading, or playing cards.

These, not coincidentally, are the rules for Settlers of Catan (Teuber, 1995) phrased in a computer code idiom.90 In this example, at least five different computational concepts are required for understanding and following these instructions. Table 1 exemplifies these concepts by depicting the rules in both pseudocode (‘human readable program code’) and the rules of the game.

Computational Concept



Game Instructions (summarized)


A primitive is a basic action that the computer/game knows how to enact.


Roll the dice.

Branching Logic

If some specified set of logic is true, enact branch A, otherwise enact branch B

result = dice_roll();

if (result == 7):




If you roll a 7, move the robber and enact the robber rules; otherwise, deal resources matching the rolled number to the appropriate player.


Encapsulated and internally consistent set of instructions for a particular task.

function deal_resources(result):

foreach player:

foreach property of player:

if( number_value_of (property)

== result):


resource_value_of ( property ),

player );

To deal rsources, give each player the appropriate resoure(s) of any tile on which they have a property.



In both computer programming and tabletop games, the order of instructions is paramount.

(see any of the example above.)

1. Roll the dice

2. If you roll a 7, move the robber and enact the robber rules.


Do some specified sub-set of the instructions repeatedly, until some end condition is met.

while not done:

foreach player:


Go player by player, allowing each one a turn until the game is done.

However, it can be difficult for humans to enact these rules. In earlier work (Berland & Lee, 2010; Berland, Lee, & DuMont, 2010), we showed how novices playing a game of Pandemic (Leacock, 2008) encountered several fundamentally computational problems, and how they solved those problems using debugging techniques relatively similar to those in computer programming. These bugs shed light on how players understand the game. The table below presents potential bugs resulting from the active player rolling a seven in Settlers of Catan:



Rule in game (summarized)

Robber is placed back on the ‘desert’ title.

Misapprehension of a (sub-) rule

The robber must be moved upon rolling a 7, but it cannot be placed on the desert tile.

Players with 9 cards keep 5, rather than 4

“Off by one” -- a common error in computer programming.

Any player with more than 7 cards can only keep half, rounded down.

Players who did not roll are trading.

Not following rules that seem unnecessarily runitive or “un-fun.”

The active player must be involved in any trade on his/her turn.

The problems are inherent to the context; humans are not explicitly rule-comprehension or rule-following machines. The main reason that a human would work to follow a rule without error when playing a game for fun is if s/he believes that the game will be more fun when the rules are perfectly followed. If this is not the case, players will change or ignore the rules. These are (perhaps paradoxically) the core strengths of the environment:

Computational thinking is socially reinforced.

Rule modification (or ‘computer programming’) is socially reinforced

The following section describes the ways that boardgames socially reinforce computational thinking by all players.

Social Computational Thinking: Boardgames As A Socially-Mediated, Collaborative Computational-Learning Environment

Not only do tabletop boardgames encourage or require that players engage in computational thinking, but they require that players talk about their computational thinking and engage in the social aspects of computational literacy. In fact, boardgames, rather than being self-learning environments, can be excellent collaborative learning environments (Berland & Lee, 2010; Zagal, Rick, & Hsi, 2006). They share myriad characteristics with environments that “foster communities of learners” (Brown, 1992); such environments have been shown to be an effective way to support learning complex content. That is, they:

Engage a group of learners in solving a joint task

Encourage learners to share information to move towards a unified goal

Engage in a consequential independent task serving the unified goal

Engage learners in reflection about the viability of their contribution

Furthermore, the experiences around strategic tabletop games share many characteristics with the types of discussion boards that we see around multiplayer online games. (Steinkuehler & Duncan, 2008) detail the literacy benefits and the scientific argumentation styles that occur with frequency on videogame message boards. Boardgames have a feature that much of the literature on games and learning has not addressed in significant depth: this collaboration is happening in real-time, face-to-face, and it is self-motivated.

In a game of Red November (Faidutti & Gontier, 2008) that I observed recently, there existed an exchange in which:

A player proposed a problem with a specific rule.

Another player proposed a resolution.

The third player commenced reading the rules through the instruction booklet, pausing for collective analysis, in which:

She read a variable number of words, pausing on any complex construction.

Each person offered an interpretation of those words, based on prior evidence.

The reader summarized the collective decision of the meaning of those words, often revisiting prior rule understanding decisions.

This looks remarkably similar to the exploratory talk described by Mercer (1996) “in which partners engage critically but constructively with each other’s ideas…” (p. 98) while making their knowledge public and working toward agreement. This is one of the ideal ways that participants can engage in collaborative knowledge construction.

Not coincidentally, it also looks like a hardware resource conflict resolution algorithm (i.e., how a computer decides which resources need addressing). When motivated by problems requiring an algorithm to solve, humans can spontaneously generate algorithmic resolutions to those problems.

Learning & Implications

Much of the games or informal learning environment research has, contained within itself, a question that appeals to learning scientists: how much of what people learn in an experience can they take away from it? So, what are people taking away from these experiences? What do they learn when playing tabletop games that they can use in other aspects of their lives?

Players are not only learning to think computationally, but they are learning tools with which to communicate computational concepts. Social and cognitive computational literacy skills are most likely an effective “preparation for future learning” of material computational literacy (Bransford & Schwartz, 1999). That is, while students may not be able to program upon gaining boardgame skills, they have a framework on which to build those skills, and; as such, will likely do so more robustly.

Expert boardgamers become boardgame designers. Harel & Papert (1990) show that when engaging students/players as teachers/designers, the students learn much more effectively. They are forced to think through their design decisions and the implications of rule changes. I suggest that strategic tabletop games encourage this type of decision-making. In fact, expert boardgamers often edit the rules of a game fluidly, communicate those changes to other expert gamers, and understand how to best implement those changes. As there is no overhead to changing a boardgame’s rules – you simply say the changes aloud – it has become very common among expert gamers. In an informal survey of frequent gamers, most of them adapt the rules – if sometimes only slightly – of every single strategic boardgame that they routinely play. As such, boardgamers routinely trend towards game design, which is itself a key mode of learning computational literacy.


Though I have no evidence for it, I would suggest that the spread of the modern strategic boardgame in the US became possible only after the spread of broadband access and computer use. All material technology for these games has existed for upwards of two hundred years at a relatively low cost; it is the form of cognition that was not immediately available. These games only became fun to Americans when they became better prepared to “think like computers,” if only at a surface level.

This essay has shown how playing through tabletop boardgames can turn us into more expert computational thinkers, and how it can make us more computationally literate (without ever touching a computer). The benefit of deeper computational thinking is relatively clear: it gives us a new way to think about hard problems. Having these new ways to think about data, to work through problems, and to understand the world is undoubtedly important; learning to think in this way by enjoying oneself could only be positive.


Berland, M., & Lee, V. L. (2010). Collaborative Strategic Boardgames as a Site for Distributed Computational Thinking. International Journal of Game-Based Learning.

Berland, M., Lee, V. L., & DuMont, M. (2010). Small Groups, Big Mistakes: The Emergence of Faulty Rules During a Collaborative Boardgame. Proceedings of the International Conference of the Learning Sciences (ICLS-10).

Bogost, I. (2007). Persuasive games : the expressive power of videogames. Cambridge MA: MIT Press.

Bransford, J. D., & Schwartz, D. L. (1999). Rethinking transfer: A simple proposal with multiple implications. Review of research in education, 24(1), 61.

Brown, A. L. (1992). Design Experiments: Theoretical and Methodological Challenges in Creating Complex Interventions in Classroom Settings. The Journal of the Learning Sciences, 2(2), 141-178.

DiSessa, A. A. (2001). Changing minds: computers, learning, and literacy. MIT Press.

Faidutti, B., & Gontier, J. (2008). Red November. Fantasy Flight Games.

Harel, I., & Papert, S. (1990). Software design as a learning environment. Interactive Learning Environments, 1(1), 1–32.

Leacock, M. (2008). Pandemic. Z-Man Games.

Mateas, M. (2008). Procedural literacy: educating the new media practitioner. In Beyond Fun (pp. 67–83).

Mercer, N. (1996). The quality of talk in children’s collaborative activity in the classroom. Learning and Instruction, 6(4), 359-377. doi:doi: DOI: 10.1016/S0959-4752(96)00021-7

Papert, S. (1980). Mindstorms : children, computers, and powerful ideas. New York: Basic Books.

Steinkuehler, C., & Duncan, S. (2008). Scientific Habits of Mind in Virtual Worlds. Journal of Science Education and Technology, 17(6), 530-543. doi:10.1007/s10956-008-9120-8

Teuber, K. (1995). Settlers of Catan. Mayfair Games.

Zagal, J. P., Rick, J., & Hsi, I. (2006). Collaborative games: Lessons learned from boardgames. Simulation & Gaming, 37(1), 24 -40. doi:10.1177/1046878105282279

89 In this work, we use computational and procedural literacy interchangeably.

90 The rules of Settlers of Catan are addressed in more detail in “Settlers of Catan,” p.93.