Patterns of design thinking in playfixing broken games: an exploratory study

Abstract

Purpose

The purpose of this paper reporting an exploratory pilot study is to examine how participant engagement in design thinking varies when playing and fixing (playfixing) three partially complete games (broken games).

Design/methodology/approach

The data for this study consist of transcripts of five playfixing sessions with a total of 16 participants. Each session focused on one of three games. The authors used Winn’s (2009) design-play-experience framework to analyze features of each game that might relate to differences in design thinking. Next, the authors coded each playfixing session’s transcript to identify patterns of design thinking. Finally, these findings were used to make conjectures about how design features and flaws might encourage particular forms of design thinking.

Findings

The findings indicate how playfixing tabletop games with varied levels of complexity, playability and rule definition lead to different patterns of design thinking.

Originality/value

This is a first step toward understanding how the constraints associated with various elements of broken games might direct participants toward desired modes of design thinking and more broadly, contributes to the literature on the educational uses of game making.

Citation

Bernier, J., Gee, E.R., Gao, Y.(B)., Pérez Cortés, L.E. and Kessner, T.M. (2024), "Patterns of design thinking in playfixing broken games: an exploratory study", Information and Learning Sciences, Vol. 125 No. 11/12, pp. 1107-1125. https://doi.org/10.1108/ILS-02-2024-0017

Introduction

Over the past two decades, interest in the use of games for learning has grown exponentially. While the educational potential of playing games has received the most attention from educators and scholars, the value of making games for achieving desired learning outcomes has been explored in a wide range of settings. Making games as a learning activity aligns well with recent educational initiatives focused on constructionist approaches to learning, such as maker spaces (Kafai and Burke, 2015). Game making has been construed as an engaging context for introducing young people to computational skills, and the majority of existing studies have focused on these outcomes. Game making also has been used to enhance learners’ subject matter knowledge in other domains such as math and language arts (Gee and Tran, 2016).

Surprisingly, how game making might be an effective means of cultivating abilities and ways of thinking associated with design remains relatively unexplored. Game design is a potentially rich and multidimensional starting point for an introduction to design thinking. Making games requires identifying game goals, thinking through the relationship of game rules and mechanics to these goals, and anticipating how players will interact with these rules and mechanics; in short, the abilities commonly associated with what is popularly called design thinking (Razzouk and Shute, 2012). Perhaps the most widely used approach to using game design to cultivate aspects of design thinking is Gamestar Mechanic (Salen, 2007), a digital environment that introduces learners to game design without the need to learn to code through creating and playing platformer games of increasing complexity. Studies of Gamestar Mechanic and other efforts to use game making to support learning about design indicate its potential to engage learners in systems thinking, designerly discourse and complex problem solving, among other design-related skills (Games, 2010).

Despite these promising findings, game making presents challenges in educational settings. Designing entire games can take a considerable time, even when creating simple games. If a team approach is used, participants may need additional time and support for dividing responsibilities and negotiating decisions. Creating the actual game elements – digital or analog – can be time-consuming and take attention away from larger design efforts. One result is that participants may lose opportunities for design processes such as prototype testing, arguably a critical element of design. Educators who are not designers (or gamers) may find it difficult to assist participants in the design process.

One crucial aspect of introducing novices to design of any sort is the scaffolding provided for the design process. Scaffolding can make decision-making easier and less time-consuming by limiting choices, identifying essential tasks or providing models of successful designs. In game making, educators have operationalized scaffolding as game templates or step-by-step instructions for creating a standard game. In addition, educators have used game modification (modding) as an alternative to designing entire games (e.g. Bastani, 2022). In an interesting variation on game modding, Gamestar Mechanic (Salen, 2007) includes a series of missions that present the user with a game that is in some way broken and cannot be played. The user must identify the problem(s) and repair the game while also meeting requirements for the completed game. These missions are associated with a particular learning objective, such as using a tool or understanding game elements such as rules or mechanics.

More recently, Pérez Cortés et al. (2022) have explored the use of incomplete or “broken” tabletop games as a focal point for engaging participants in game design and design thinking. Broken tabletop games are games that are incomplete or ambiguous in ways “that require varying degrees of fixing in order to make them playable” (Pérez Cortés et al., 2022, p. 2). In this approach, pairs or small groups are asked to “playfix” these broken games through an iterative process of playing the game, identifying problems (such as incomplete rules), fixing the problems by modifying the game, playtesting the solution and continuing with gameplay. A distinctive feature of playfixing is how participants are situated as both players and designers. Analyses of the playfixing process indicate that participants with no prior design experience demonstrated quick, sustained and nonlinear engagement with designerly modes of thought, with a particular emphasis on problem identification (Pérez Cortés et al., 2022). Additional research suggests the potential value of playfixing broken games as a means of engaging participants in collaborative talk, a key element in successful collaborative design (Gao et al., 2022).

While these findings are intriguing on their own, questions remain regarding how the designs of playfixing activities may influence what forms of designerly thinking they stimulate and how they may prepare participants for future learning in the domain. One question with conceptual and educational implications is how features of broken games might contribute to different forms of engagement in design. In general, ambiguous rules and imperfect mechanics associated with broken games are key to prompting design thought and actions (Pérez Cortés et al., 2022). Nevertheless, rules and mechanics can vary considerably in the extent to which they are vague, incomplete or simply missing, and may contribute to different forms of designerly activity. Some games may simply be “too broken” or too complete to be useful, while others may have provocative combinations of design flaws that afford engagement with some forms of design thinking while constraining others.

The purpose of this exploratory pilot study was to examine how participant engagement in design thinking varies across several different broken games. More specifically, we examined the design thinking processes of small teams who playfixed broken games, addressing the following research questions (RQs):

RQ1.

What design elements of broken games may be related to differences in design thinking?

RQ2.

How does evidence of design thinking vary over the playfixing session between groups who playfixed different broken games?

Background

Following previous studies (e.g. Pérez Cortés et al., 2022), we used broad conceptions of design and design thinking as a starting point for analyzing the design processes associated with playfixing broken games. In addition, we drew on frameworks specific to the design and analysis of games as a means of identifying game elements that might affect engagement in design thinking.

Design and design thinking

Design is a complex, flexible, situated and social process, as demonstrated by scholarly efforts to describe the complex processes of real world design (Johansson-Sköldberg et al., 2013). The term “designerly thinking” is often associated with studies of the design practices of professional designers. This body of scholarship stresses the complex, ill-defined nature of problems that are addressed by designers, the use of abductive reasoning and the role of context in evaluating design decisions (Laursen and Haase, 2019). A separate but related body of literature, associated with the term “design thinking,” focuses on translating design processes and methods for people without formal backgrounds in design. While this literature is based on similar assumptions about the nature of design problems and practice, the emphasis is on providing process models, techniques and tools that can be applied in varied settings by non-designers (Laursen and Haase, 2019). While design thinking models have been widely applied in professional settings, particularly in business management and education, they have been criticized for misrepresenting the design process as linear and decontextualized and from a pedagogical stance, focusing more on the use of tools than modes of thinking (Johansson-Sköldberg et al., 2013).

The broken game approach, as we will describe below, seeks to retain the nature of design as a situated, nonlinear and abductive process. However, the broad competencies associated with design thinking models – identifying needs, defining problems, generating solutions, creating prototypes and testing solutions – remain useful for characterizing participants’ design processes (Razzouk and Shute, 2012). Following previous work on playfixing broken games, we adapted these competencies as a basis for our analyses and use the term design thinking to reflect our focus on introducing design practices to non-designers.

Game design

Game design is no exception to the complexity associated with other design activities, or with efforts to simplify the process for novices. As Whitson (2020) points out, popular game design textbooks tend to present somewhat idealized portrayals of game development processes, mapping out discrete processes and tasks guided by generalized principles or guidelines. Whitman’s ethnographic research in game studios demonstrates the “messiness” of game development, including tensions between idealizations and actual experiences. The complexity of game design is reflected in Salen and Zimmerman’s (2003) argument that the goal of successful game design is what they call “meaningful play” (p. 37); designers must consider how players’ interactions with the game system, including tangible and intangible artifacts and mechanics, contribute to emotional and psychological experiences that are rewarding to players. Hunicke et al. (2004) codifed these dimensions into three main components of games: rules, system and “fun.” Their design counterparts – what designers must consider in designing a game – include mechanics, dynamics and aesthetics (MDA). In this MDA framework, player experiences (aesthetics) are elicited through interactions (dynamics) with the game rules and properties (mechanics).

Lankoski and Holopainen (2017) suggest that the MDA model may be the most commonly used framework for game design around the world. The framework does not offer a specific approach to the game design process, but rather was intended to be a “a formal approach to understanding games – one which attempts to bridge the gap between game design and development, game criticism, and technical game research” (Hunicke et al., 2004, Para. 4). Of particular value for our study of playfixing is the framework’s emphasis on experience-driven design and the perspectives of both players and designers.

While the MDA framework has been widely cited, other scholars have pointed out some limitations of this conceptualization and created alternative or more expansive models to address these limitations. We adopted one such model, the design-play-experience (DPE) framework (Winn, 2009), as a basis for our work with playfixing broken games. The DPE framework addresses several concerns with the MDA model, particularly its lack of attention to game elements beyond mechanics, such as the game narrative. Renaming the main elements of the framework reflects this inclusivity and avoids confusion over the interpretation of terms such as aesthetics. In addition, the DPE framework was explicitly created to address games for educational purposes: “The DPE framework presents a language to discuss design, a methodology to analyze a design, and a process to design a serious game for learning” (Winn, 2009, p. 6).

Following the MDA framework, the DPE framework depicts the relationships among player experiences, interactions (play) and game elements (design). However, the DPE framework articulates four additional subcomponents: learning, storytelling, game play and user experience. Each subcomponent cross-cuts design, play and experience, and is manifested in different ways. For example, learning is reflected in design as the content and pedagogy of the game, in play as the forms of teaching that occur during game play and in experience as the learning that takes place. Technology is another element of the model and is represented as a foundational layer, because technologies used to make and play a game will affect all aspects of game design, play and experience.

Playfixing broken games

The DPE framework is not a linear or prescriptive model of game design; we find it to be a useful means of understanding participant engagement throughout the playfixing process. As participants play the broken games, they can encounter problems in game elements that affect their experience (contributing to, e.g. confusion, boredom or frustration) and in turn, they must discern and hopefully resolve those problems through designing or redesigning elements of the game. Similar to the processes identified in the literature on designerly thinking, as described above, participants engage in an iterative process of identifying problems and solutions, potentially shifting frequently among different levels of problem analysis and design tasks (Laursen and Haase, 2019). In the current study, we draw primarily on the framework’s four components of design (content and pedagogy; character, setting and narrative; mechanics; and user interface) to characterize elements of the broken games that may relate to design thinking.

Methods

Our data for this study consist of transcripts of five playfixing sessions with a total of 16 participants. Participants in the study were university students recruited through flyers on campus. Nine women and seven men from varied majors participated in the study; Table 1 provides more information on the backgrounds and prior experiences of each participant.

Game selection

Each session focused on one of three games: Pollutaplop, Budget and The Rescue (see Figure 1). Pollutaplop is a game where players try to clear out pollution from water-based biomes within an interconnected system. Budget is a turn-based racing game where players manage a limited amount of funds to increase their speed or avoid traps. The Rescue is a roleplay-inspired game where players try to defeat monsters on the way to rescue a child who is scared of the dark. The games were developed by youngsters (ages 10–16) and differ greatly in their complexity and brokenness, as we discuss below.

We selected these games from a larger set of tabletop games created by youngsters (ages 10–16) in afterschool game making workshops hosted by a local library in collaboration with the research team. The game topics were chosen by the youth as issues that they felt were relevant to other teens or children, and were not limited to a particular subject area. We used several criteria to select the games used in the current study:

• each game’s rules and player actions were somewhat defined but incomplete;

• the games differed in complexity and genre; and

• each game addressed a different topic.

While our goal in this analysis was not to examine content learning, we hoped that the varied game topics might illustrate the relevance of design thinking and playfixing in relation to quite different content areas.

Data collection

Two researchers facilitated 45-min playfixing sessions at a university research lab. Groups of two to four participants were formed based on participant availability, with each group being assigned a broken game to playfix. When these participants arrived, they were asked to complete a short survey regarding their background and experiences with games and design (see Table 1). Then, the participating students were informed that the games were made by kids with an educational goal pertaining to a social or personal issue, but that the games were in need of adjustment. Participants were given existing game rules and materials as well as additional game materials, and were told that they could make any changes to the game. Importantly, the interviewers stated:

There are probably a lot of different ways to complete the game, and there isn’t any one “correct” way to make it playable. We’re interested in the process of whatever decisions you make.

These playfixing sessions were audio and videorecorded, and the audio recordings were transcribed. At the conclusion of the session, participants were asked to respond to a few reflective questions on a survey and then to consider whether they had participated in design thinking according to the d.school (2018) model in a reflective interview. For the purposes of this manuscript, we use only the pre-survey and playfixing data.

Data analysis

Our RQs required two approaches to data analysis. To answer our first question about how design elements of broken games may relate to differences in design thinking, we focused our analysis on the three broken games, using the four elements of design in the DPE framework. While we have some insight from our gameplay as well as the gameplay of the playfixers, this was insufficient to fully characterize the components of play and experience for each game. Because play (and accordingly, playfixing) is mediated by the players, our arguments as to differences in potential opportunities for design thinking in these games rest in consideration of the design elements. From this analysis of design elements, we constructed descriptions of each game and then identified key design differences across the games.

To address our second question about variations in design thinking across groups, we analyzed each playfixing session’s transcript using a coding scheme based on the Stanford d.school’s design thinking strategies – Empathize, Define, Ideate, Prototype and Test. This coding scheme, as reported by Pérez Cortés et al. (2022), consisted of modified versions of these five design modes that relate more specifically to game design. Our definitions for these codes alongside examples are presented in Table 2. Pairs of researchers coded transcripts separately and discussed coding disagreements until consensus was achieved. We used MAXQDA’s data visualization features to generate charts and document portraits reflecting the distribution of design thinking codes. Visualizations were generated for each individual transcript, for transcripts grouped by game and for all the transcripts together. Finally, we used the findings from both sets of analyses to make conjectures about how differences in design features and flaws might encourage particular forms of design thinking.

Findings

In the following sections, we discuss findings related to each RQ. We begin with a description of each game’s design elements (RQ1: What design elements of broken games may be related to differences in design thinking?).

Game design elements

Table 3 presents a breakdown of design elements for each of our three broken games: Pollutaplop, Budget and The Rescue. Key differences can be seen in their relative complexity and completeness. Here we describe each game’s “brokenness” in more detail in relation to their design according to the DPE framework.

Pollutaplop.

Pollutaplop is a game about eliminating water pollution that takes inspiration, both aesthetically and mechanically, from resource management games like Settlers of Catan. Pollutaplop is a substantially more complex game than Budget and The Rescue, with a greater number of user interface components, greater player choice in actions during turns, complications in the form of disaster and event cards and a more complicated win condition.

As designed, Pollutaplop is a fairly complete game. There are a few small ambiguities in the rules; for example, players draw between one to six action cards each turn for use during the turn, but the rules do not specify if they keep those cards in their hand or if they discard them at the end of their turn. As might be expected given the number of mechanics and user interface components, the brokenness of Pollutaplop is found in that some of the elements of the game’s design do not seem to form a cohesive whole.

The role of factories and the recycling plant in the game’s design are problematic. At the start of the game, there are no factories on the board, which means there is no ongoing generation of pollution apart from disaster cards and event cards. Factories can only be added to the game through disaster cards or through an action card that generates factories in random regions. This contrasts with the “real-world” case where factories produce items of value, generating money for the owner and tax revenue for the local community in exchange for the pollution generated. Moreover, because there are no factories in play when the game starts and the means to put them in play are rare, the mechanics of factories can be seen as extraneous. The recycling plant works somewhat better as a real-life analogue, as it removes pollution from a communal area. But, four total actions are needed to build the recycling plant, requiring a player to use multiple turns or to collaborate. For this effort, it removes only one pollution cube from the common area per round, advancing the game slightly closer to an end state without helping any individual player.

Another problematic design element of Pollutaplop is its competitive nature. A player wins if they have cleaned the most net pollution from the board at the end of the game. This creates some odd incentives – a player could make choices early in the game to pollute their own region to more easily clean up more pollution later. While there is some verisimilitude, as a company might make cleaning up its own mess into a public relations win, and might refuse to help clean any pollution that is “not their problem,” this is a case of ludonarrative dissonance within Pollutaplop. Cleaning water pollution is a collective problem for society rather than an individual one, and so a cooperative game would seem more thematically fitting.

Budget.

Budget is a turn-based racing game meant to help players learn basic financial literacy through how they spend money in the game. As can be seen in Table 3, the game’s mechanics are reminiscent of classic childhood board games like Life, Chutes and Ladders and Candyland. Compared to Pollutaplop and The Rescue, Budget is a much simpler game, with far fewer user interface components than either game, a clear win condition and a minimalistic setting and narrative.

In accord with its simplicity, Budget can be a very brief game – players can read the rules and play through a complete game in about 30 min. While brevity in board games can lead to highly replayable experiences, for Budget, the brevity and simplicity of its rules combine to define its brokenness. The movement in the game is highly randomized, minimizing the effect of strategic play. A player who rolls unluckily could hit a spike token early on, meaning that they effectively have nothing to do for two full cycles in the game. A player who rolls luckily can acquire multiple speed tokens (that accelerate movement) and get to the end of the road in just a few turns.

These problems dovetail into a broader, thematic problem. Budget is meant to be about budgeting, yet it offers players minimal opportunities to manage their money. Players largely earn money by random chance. As a consequence, players may not receive any money until their second or third turn. When they do, there may not be many options remaining for spending money, depending on other players’ luck and decisions. Because players are limited in their ability to know what amount of money they will have on future turns, they are incentivized to spend as much as they can as soon as they can, which is not an ideal financial strategy in “real life.”

The Rescue.

The Rescue is a board game themed around helping children overcome their fear of the dark, inspired by tabletop roleplaying games like Dungeons & Dragons. Players choose a class and a weapon and take turns traveling on a board from the top to the bottom along branching pathways, encountering monsters along the way and trying to reach the scared child at the bottom. While this overarching goal is defined, The Rescue leaves many rules unspecified or underspecified, making it in some ways the “most” broken game we used in this study. In particular, while each character and monster has hit points and weapons have damage amounts, no rules for combat are provided to give meaning to these values and how they interact with each other. A playfixer familiar with role playing games or other games with fantasy combat might have some ideas on this front, but the design problem is still larger in scope than Budget or Pollutaplop.

A particularly challenging monster to design combat rules for is the old spider. Unlike the other monsters in the game, no hit point amount is provided for the spider. The only information is that its unnamed default attack does 30 damage, but it sometimes spits venom to deal 60 damage. A playfixer who encounters the old spider has to make a number of decisions to make the combat work: how much health does the spider have? How often should the player and the spider attack each other? How often should the spider spit venom? Should attacks be 100% accurate, or should there be chance involved?

Ultimately, The Rescue, Pollutaplop and Budget offer varying kinds of brokenness along the lines of their design elements.

Variations in design thinking across broken games

Our RQ2 asked “How does evidence of design thinking vary over the playfixing session between groups who playfixed different broken games?” For the combined playfixing sessions, we coded 526 instances of design thinking, with ideate and then define identified most frequently, comprising more than half of coded segments (see Figure 2). Instances of empathize and test were much less common.

To identify variations in these general patterns, we looked at the distribution of codes within groups playing the same game (see Figure 2) as well as examined the “document portraits” rendered by MAXQDA for each group. Below we describe several key patterns of design thinking associated with each game, illustrated through brief excerpts from the groups’ conversations.

Pollutaplop – defining problems in the system.

Among groups who played Pollutaplop, Define was the most frequently identified design thinking mode, as represented in Figure 2. Instances of problem definition appeared throughout both groups’ playfixing sessions. The complexity of the game required participants to spend time reading and referring back to the written rules before and during gameplay. As described above, Pollutaplop’s rules were relatively complete, so most problems identified by participants were related to game play and player experience. These included whether the game would take too long to win, how recycling plants and factories might be incorporated in more meaningful ways and the alignment of game play with real-world pollution mitigation strategies, as in the following excerpt:

Ivy: Yeah, I want to keep a hand at all times. Because then, that’s strategy, right? You wouldn’t play all your good cards at the beginning. You would think about it. Where I don’t have anything to play and for the last two turns, I’ve only been able to play whatever.

Jay: Well, maybe that’s the point. Instead of being able to long term strategize, it’s an improvisational strategy game […].

Ivy: If you’re dealing with pollution, you should do long term strategies. Just saying.

In addition, the complexity of the game contributed to extended sequences of Defining among both groups, as they considered multiple perspectives on a problem or returned to it as they played. Group C, for example, returned to the purpose of the recycling plant several times. About 15 min into game play, when Adam began to build a recycling plant in the common lake area, he remarked “I don’t necessarily see the point of building a recycling plant if it’s not in your region, because it’s always built in the lake region.” This prompted an extended conversation that touched on a broader question of whether the game is competitive or collaborative:

Adam: And so in the scheme of it, it’s useful that the pollution gets reduced, but in terms of winning the game that you want to clean out more than your opponents.

Betsy: Yeah.

Adam: Yeah, it’s weird. And I think the base concern is that it’s a competitive thing as opposed to a cooperative thing.

Betsy: Yeah. I was thinking it’s cooperative where you play with everything, together.

Adam: Where it’s like Pandemic or something.

Betsy: Yeah.

David: What’s the point of the game if there’s no competition?

Betsy: Well, I mean, you’re competing against the game.

Adam: Yeah, you’re competing against the systems of the game.

In this example, the participants are engaged Defining mode of design thinking because they are articulating and clarifying a game-specific design problem: whether a game about fighting pollution should be a competitive or a cooperative game. Adam suggests this as a concern with how the current game is designed, while David demands clarification on what a cooperative approach to the game would look like.

Both groups identified design problems that were not always straightforward and prompted alternative interpretations. However, the groups engaged in the playfixing process in different ways. Group C spent 10 min reading the rules aloud before starting the game, while Group B read enough to begin the game and then read further in the rules as gameplay progressed. Group B spent much more time questioning various aspects of the game than Group C, including whether game play seemed meaningful in relation to their assumptions about real world solutions to water pollution. Ivy, a Group B participant pointedly asked: “Do you just want the game to work, or do you want it to make sense?” Occasionally they identified an issue that they discovered was already addressed by a rule they had not read, but incrementally reading the rules also seemed to prompt more discussion of their logic and impact on game play. They brought up some issues repeatedly as they played, without making any actual changes in the game rules to address them. Group C did not identify as many potential problems in the game, though Define was still by far the most frequent design thinking mode in their discussion. In contrast to Group B, Group C Tested several game modifications, such as allowing every player to build their own recycling plant.

Overall, Pollutaplop’s varied game elements and rules required participants to spend time both in discussion and in actual play to understand the existing mechanics and their impact on player experience. This contributed to the prevalence of problem definition as a design thinking mode, as participants clarified their understanding of the mechanics and their rationale. The game represented the complexity of addressing an important environmental issue, prompting participants to focus in part on whether the game mechanics “made sense” in relation to real-world dynamics, adding to the emphasis on problem definition. Interestingly, neither group gave any attention to the accuracy of “content” as represented in the event cards or disaster cards. The game was relatively complete and playable, so participants did not necessarily have to create new rules or revise existing ones to play, reducing the extent to which they engaged in ideating, prototyping or testing.

Budget – play, then ideate.

Ideation comprised nearly three-quarters of the design thinking modes identified in the group who playfixed Budget (see Figure 2). As described above, this game is relatively simple and has few significant issues that interfere with game play. This group played through the entire game during the playfixing session, apparently because they had little need to Prototype or Test new game elements to keep playing, They identified only minor problems in game play mechanics and interface during game play, including lack of clarity about the initial setup of the game board, and rules about the purchase and use of speed tokens. They Prototyped and Tested a new rule for speed tokens, but otherwise were able to play the game with no other modifications. However, about 20 min into the session, the group identified a broader problem in how game mechanics did not align with the stated purpose of the game, to help players learn about budgeting:

Luna: I guess I might as well buy my speed token. I’m confused how this is a budgeting game if there’s no incentive for-

Merryl: For saving more.

Luna: Saving and having more money at the end […].

Kori: Yeah, that’s a question as well. The only way to win is to finish first.

As they finished playing, participants began to Ideate ways to incorporate budgeting into the game. Kori, for example, stated:

I think for a budget focused game, it would be interesting if kind of like Life, but not exactly like Life, where you can have multiple different roads and you can only go in a certain road if you have a certain amount of money. So it makes you more conscious about spending money.

The majority of this Ideation occurred after Kori won and the game ended. In the remaining 15 min, the group engaged in an extended brainstorming session, focusing on adding choice and complexity to the game, as in this excerpt:

Kori: Maybe allow returning a speed token or not using it so there’s more choice about avoiding hazards and maybe more hazards or diamonds.

Merryl: Yeah. It would still cost you money to return a speed token. It would just be you had already invested in something and were giving it up and would have to pay it again. And that would be even more onerous when we’re up to the 600 one. So I think that would make it more interesting, because then I would have been able to choose more times […].

In this final discussion, the group returned repeatedly to the alignment of game play and budgeting, ending with the following idea as time ran out:

Merryl: Yeah. I feel like there wasn’t much choice about what to buy really […] I don’t know what it teaches you about budgeting.

Kori: Yeah. I think with the monetary incentives we’ve put in–

Luna: The real budgeting would be at the beginning, everyone got the same amount of money and you never had an opportunity to get more money the rest of the game […].

Kori: So it would be payoff between do you want to buy the speed tokens versus do you want to have money to save yourself from the spikes? Yeah. That would be interesting. Having a set amount of money [crosstalk].

Merryl: Yeah. You’d have to make all your purchases at the beginning of the game and then see it play out. That would be good.

Luna: Or every fourth turn, when you get money, that’s the only opportunity to buy things. Not every turn. You get money and buy things.

These examples represent the Ideate mode of design thinking because they consist of a sequence of design ideas for changes to the game. Though there are implicit and sometimes explicit goals presented as purposes for these changes, the primary action of the playfixers here was to suggest as many changes as possible. As most of the suggested design changes are not refined through iteration but instead simply stated once before moving on, we view this as an extended Ideation session.

While the Budget’s game mechanics led to a rather brief game play experience, the advantage of a playable game was that participants had the opportunity to reflect on the game as a whole. This allowed them to assess the overall experience of game play (not very interesting, per participants’ own words) as well as how it did or did not achieve its intended pedagogical goals. However, the group had little or no chance to actually Prototype and Test new game mechanics; “fixing” is supposed to be a key affordance of using broken games to facilitate design thinking. In this case, a relatively simple solution would be to allow more time for the playfixing session so that participants could modify the game and play again, potentially confronting additional issues as they experimented with new mechanics.

The Rescue – prototyping rules.

Figure 2 indicates that ideating was also the most frequently coded design thinking mode in groups playfixing The Rescue, but with a higher proportion of prototyping than groups playfixing the other two games. The Rescue’s ill-defined game rules for the character classes, monsters and weapons, along with the complexity of the fighting mechanic, required participants to Prototype more extensively to make the game playable. In both groups, Prototyping was prompted when a participant first encountered a monster on the game board. For example, in Group F, Omar moved into a space with the Old Spider, initiating the group’s creation of rules for combat:

Nicole: Here I’ll write it down so we can keep track. So when you get to […]. a monster, what should the rule be? When you get to one, you roll to see what it does and you roll to see what your power will be? […].

Omar: I need to, yeah, figure out if it’s going to spit venom […].

Nicole: Oh, even it will attack and odd it doesn’t?

Pam: Yeah. And then you can determine your damage.

This conversation continued for ten minutes as the group created various rules for damage points and dice rolls. While this session included the Ideation mode of design thinking as the participants suggested ideas for how combat should work, these ideas were quickly subject to adjustment and refinement through the Prototyping process. Once they had a sufficiently complete Prototype, Omar fought the Spider Testing the new rules. The group then further Prototyped the combat experience by proposing additional rules to address questions such as whether players could move on if they survived one round of fighting. Omar was able to move on, leading the group to discover a larger problem with the game: the monsters were so few and dispersed along different paths that defeating just one monster was all that was necessary to reach the scared child. Some monsters could be avoided altogether, rendering them apparently meaningless. The group spent the last 15 min of their session Ideating and Prototyping new rules to incentivize fighting monsters and make the game more interesting.

Group A took a slightly different approach to fixing the problem of incomplete rules. When Grace encountered the first monster, in this case the Ogre, the group created Prototype rules for fighting it. However, instead of Testing these rules and continuing to play, Fred proposed that they further develop their Prototype for combat by dealing with all monsters: “Players roll […] Monsters hit […] as would be dictated by the monster card. So, let’s look at all the monsters, and then just assign rules now.” This group devoted about 15 min to Ideating and Prototyping rules for fighting all monsters. When they finished this task and started to play again, Eva asked a question that shifted the group’s focus back to Prototyping, in this case to clarify abilities of different character classes, and give a function to “secrets” on the game board. This Ideation and Prototyping episode lasted for an additional 11 min. When the group finally returned to game play, killing the ogre took another 8 min due to rules they created for fighting (one round comprised a player’s entire turn), reflected in the following example of ideation and prototyping:

Eva: As well, you can also think about how long does the battle occur. Is it to death? Or do you have multiple turns that you can-

Fred: I would say you roll once-

Eva: I think multiple turns.

Fred: Yeah, it would be multiple turns, so like she’ll roll and deal her attack-

Eva: And then she just can’t get past it until it’s dead.

Fred: Yeah, until it’s defeated. And then of course he’ll roll and then it’ll go to your turn, then my turn, and then back to her. And she’ll have to roll again

Eva: And then, she’s still in battle while we’re progressing.

Here, Fred and Eva develop a Prototype for “how long does the battle occur,” by taking the initial idea (battles take multiple turns) and further fleshing it out as a system. During the battle with the Ogre, another player began to fight a different monster, the Summoner. Due in part to these multiple lengthy combats, in the limited time remaining, the group did not reach the end of the game and did not have an opportunity to reflect on the meaningfulness or challenges of the game as a whole.

In comparison to the other two games, The Rescue’s genre seemed particularly important to participants’ design thinking processes. Both groups had participants familiar with dice-based role-playing games with a combat focus, reflected in how readily they discussed damage points and health points and how they used dice rolling as the fighting mechanic. The players made a few references to game content, such as questioning the nature of Robin as a character or asserting that a spider should do less damage than a fire wizard, but neither group discussed whether the game might achieve its pedagogical goal of helping children overcome a fear of the dark. This might be due to the complexity of creating game mechanics for multiple characters and monsters, a task that kept their focus primarily on discrete game elements. It might also be related to the goal; overcoming fear of the dark involves addressing emotional states that might not seem as straightforward as providing opportunities to calculate how to spend in-game money or demonstrate how a combination of human and natural causes affect water pollution.

Discussion

Extending findings from previous research (Pérez Cortés et al., 2022), this study suggests how playfixing tabletop games can engage participants in design thinking in brief design sessions with little preparation. While the three games in this study had varied levels of complexity, playability and rule definition, our findings indicate similarities as well as differences in the opportunities they offered for design thinking. As this was a pilot study, our data are insufficient to explain the similarities or differences conclusively. However, our analysis at least suggests that the attributes of the broken games played some role in these outcomes, as well as how we organized the playfixing activity.

Define and Ideate were by far the most frequently observed forms of design thinking across all playfixing sessions, with the exception of The Rescue session in which Ideate and Prototyping were most frequent. Attributes of the playfixing sessions likely contributed to these patterns. For example, participants had no prior experience with these games or with playfixing as an activity. Participants were told that these games were broken and asked to fix the games as they played. Thus, participants may have been oriented toward identifying problems and ideating design alternatives, even when those problems did not necessitate actual changes. Conversely, Testing and Empathizing were infrequently identified in any playfixing sessions. Limited Testing could be explained in part by the 45-min timeframe. While we set this limit to test an activity that might fit within a typical class period, longer or multiple sessions with the same game might lead to more extensive Prototyping as well as Testing.

Enhancing the (observable) Empathizing in playfixing, or game design in general, is a challenge with conceptual and practical dimensions. Design thinking “toolkits” typically include exercises aimed at increasing understanding of user perspectives. These exercises are completed as a preliminary step, prior to problem definition, ideation and so forth. A clear distinction is assumed between designer and user, and empathy is reflected in a designer’s explicit reference to users’ feelings or beliefs. Playfixing broken games as a design activity assumes that empathizing might occur throughout the design process, and does not make clear distinctions between designer and users. In our sessions, participants were relatively similar in age and educational background, and they were playfixing games intended for children or teens. Adding participants who represent the intended players of the games, or involving more diverse groups in playfixing games together, may enhance participants’ expressions of empathy for players who might be quite different in their preferences, abilities or backgrounds.

Along with these similarities, we observed distinct patterns of design thinking associated with different broken games. Pollutaplop, with its multiple elements and rules, along with its representation of a complex environmental issue, placed more demands on participants to understand the existing game as well as potential issues in game play, contributing to an emphasis on problem definition. In comparison, Budget’s game mechanics were relatively simple as well as complete; rather than pausing to Define potential problems in game design, participants were able to play through the game. They were able to reflect in a more global sense on player experience, particularly about whether the game was enjoyable and achieved its learning objectives, and to Ideate potential design modifications. Finally, The Rescue’s incomplete mechanics prompted participants to Prototype rules; Testing these Prototypes depended on whether they attempted to “fix” all incomplete elements at once or continue with the game.

An important pedagogical strategy for cultivating design thinking is using constraints to reduce the complexity of the design situation and guide learners toward successful outcomes (Goncher and Johri, 2015). Our findings offer a preliminary step toward more detailed understandings of how constraints associated with elements of broken games might direct participants toward desired forms of thinking. At the same time, we also must consider the constraints of our data collection process. We collected data from participants with no prior relationship to one another in a one-shot, 45-min activity isolated from any coursework or professional experience in design. As such, we have not presented an activity that is ripe for implementation in the classroom, but rather, the seed of such an activity. To transition this work into formal classrooms, a follow-up study with a revised version of the playfixing activity is needed.

In such a study, the first element to consider would be integration into a particular course context. Playfixing in isolation may give students some experience with design thinking, but it is only through deliberate placement within a larger curriculum that this experience could enable learning. While the 45-min playfixing experience is timed as such to fit into typical time constraints for lessons at both the secondary and postsecondary levels, a larger design could consider how the prior and posterior class sessions set up and build on the playfixing activity. Another significant element to consider would be the use of more deliberate scaffolding to foster more extensive engagement with design thinking across all broken games. Such scaffolding might consist of providing guiding questions, additional resources and structure during playfixing activities, consistent with the educational goals of a particular context.

Another element to consider in future scholarship is to further explore how more specific design patterns (Bjork and Holopainen, 2005) in broken games might be particularly conducive to eliciting certain forms of design thinking. Our findings indicated some general relationships, such as how increased design complexity prompted more problem definition, while simplicity and relative completeness in design increased participant focus on the game’s larger goals. The identification of specific patterns may be aided by expanded use of the DPE framework. For example, we examined only broken game elements in the design category of the framework. Analyzing instances of play and experience elicited by broken games might provide additional insight into how playfixing stimulates participant engagement in design thinking. Finally, many other contextual factors may affect how groups of novice designers engage with playfixing broken games, including the participants’ prior knowledge of games and how participants interact with each other; these factors can be additional focal points for future scholarship.

Methodologically, future studies might use different strategies for analyzing design thinking in the context of playfixing. For the purpose of this study, we adopted a coding approach based on simplified descriptions of design thinking modes. This approach was useful for broad comparisons but did not reveal more nuanced variations in, for example, how problems were identified and explored, or how design decisions were made. Analytic techniques such as discourse analysis, interaction analysis and multimodal techniques offer alternative ways to investigate design processes.

In conclusion, our work contributes to the growing body of scholarship that emphasizes understanding how specific mechanisms, tools and tasks are critical to successful educational design activities (Li et al., 2019). Kafai and Burke (2015) noted the difficulty of drawing conclusions from the literature on educational uses of game making, given the great diversity in contexts, tools, desired learning outcomes and so forth. We hope that the findings of our present study contribute to lay the groundwork for more nuanced and extensive investigation of the educational potential of playfixing broken games – and of making games more broadly.

Figures

Figure 1.

Photographs of Our Broken Games. From left to right: Pollutaplop, Budget, and The Rescue

Figure 2.

Distribution of Design Thinking Codes by Game

Table 1.

Summary of groups and participants

Group	A	Game	The Rescue
Participant pseudonym	Gender	Degree and major	Gaming experience	Design experience
Eva	Female	Environmental sciences, PhD	Never	Some
Fred	Male	Economics, bachelor’s	At least once a week	Some
Grace	Female	Geological sciences, PhD	Less than once a week	None
Group	B	Game	Pollutaplop
Participant pseudonym	Gender	Degree and major	Gaming experience	Design experience
Hugh	Male	Astrophysics, bachelor’s	Less than once a week	Some
Ivy	Female	Geological sciences, PhD	Once a week	Some
Jay	Male	Exploration systems design, PhD	At least once a week	Some
Group	C	Game	Pollutaplop
Participant pseudonym	Gender	Degree and major	Gaming experience	Game design experience
Adam	Male	Digital culture, PhD	Once a week	Some
Betsy	Female	Computer systems engineering, bachelor’s	Never	Some
Carlos	Male	Geology, PhD	Once a week	Some
David	Male	English and computer science, bachelor’s	Never	None
Group	E	Game	Budget
Participant pseudonym	Gender	Degree and major	Gaming experience	Game design experience
Kori	Female	Geology, masters	Less than once a week	None
Merryl	Female	Geological sciences, PhD	Less than once a week	None
Luna	Female	Civil engineering, bachelor’s	Never	Some
Group	F	Game	The Rescue
Participant pseudonym	Gender	Degree and major	Gaming experience	Prior design experience
Nicole	Female	Business administration, masters	Never	None
Omar	Male	Biomedical engineering, bachelors	Once a week	Some
Pam	Female	Non-profit leadership, masters	Less than once a week	None

Source: Created by authors

Table 2.

Adaptation of design thinking modes

Design mode	d.school description	Playfixing-specific adaptation	Example of coded speech segments
Empathize	Observe, engage and immerse self with end users; wear the users’ shoes	Seeking to understand and be responsive to the perspectives, needs, motives, preferences, intentions or emotions of end-users and/or original designers	“I think it's ambiguous to me which one is intended [by the original designers], but if you did a multiplier, then the game would be over far too quickly”
Define	Identify and scope a meaningful design challenge	Articulating, clarifying or interpreting a game-specific design problem	“I'm trying to imagine how the enemy attacks, how the monster would attack”
Ideate	Explore a wide solution space by generating radical design alternatives	Suggesting ways to modify or change game elements	“Maybe instead of losing if you have money, can you pay to not lose a turn if you're on a spike? We could add that maybe?”
Prototype	Using physical objects to get ideas out of a person’s head and into the world so others may interact with them	Getting ideas out of a person’s head and into the world in either physical or abstract ways so others may interact with the ideas	“But also getting to draw more cards isn't actually a benefit if you only have two points to spend every round. You just have more options. You can't do more”
Test	Iteratively gather feedback, refine solutions and continue to learn about users in authentic scenarios	(Re)play the game with newly prototyped changes to the previous iteration to collect information on how the change works or does not work with the existing and/or projected version of the game	“Yeah, we can do that. So let’s restart our whole thing. So I’ll roll twice and see what I get”

Source: Created by authors

Table 3.

Analysis of our broken games – key elements

	Pollutaplop	Budget	The rescue
Learning Content and Pedagogy	Processes, challenges and importance of cleaning up water pollution	Financial literacy and budgeting	Overcoming fear of the dark
Storytelling Character, Setting and Narrative	Characters: N/A Setting: Waterways (lake, beach, canals, coast, river, outskirts, swamp) Narrative: “Players work together to clear pollution from every region’s waterways, gaining recognition for being the best at water conservation efforts”	Characters: N/A Setting: A linear race track with spikes Narrative: “Players try to make wise choices about how to spend money, while still racing to the end as fast as possible”	Characters: Classes (warrior, thief, mage, robin), monsters (summoner, ogre, fire wizard, old spider), and the scared child Setting: A dark stone dungeon with branching pathways Narrative: “Players work together to reach the end of the maze and rescue the scared child in the dark”
Gameplay Mechanics	Rounds of gameplay with three phases Upkeep phase: pollution cubes get added or removed based on the presence of factory and recycling plant tokens Action: Players take turns. Each player rolls a die, draws that many action cards, plays cards up to a total cost of two actions. If they roll an even number, a disaster card is played Event: At the end of the round, an event card is played Winning: The game ends either after a set number of rounds or when there are no more pollution cubes on the board. The winner is the player who clears the most pollution cubes off the board	Players take turns rolling a die to move along the track Spending: Players can spend money at the start of their turn. Players can buy speed tokens and spike remover tokens Spikes: If a player lands on a square with spikes, they lose their next two turns and return all of their speed tokens Movement: Players roll a die and move that many spaces, then do so again for every speed token they have Winning: The winner is the first player to reach the final square	Players take turns rolling a die and choosing a path through the dungeon Monsters are scattered throughout the board, and it is suggested that players battle these monsters; however, no rules for combat are provided Winning: If any player makes it to the scared kid, the players win collectively
User Experience User Interface	Board: Seven hexagonal pieces, each representing one of the waterways Tokens: Pollution cubes, factory tokens and recycling plant tokens Dice: One six-sided die Cards: Action cards, disaster cards, and event cards, all in separate decks	Board: Rectangular board with a linear, switchback pathway Tokens: Car tokens, speed tokens, spike remover tokens and play money Dice: One six-sided die Cards: None	Board: Rectangular board with a branching pathway Tokens: Character tokens Dice: One six-sided die Cards: Four class cards, five weapon cards, four monster cards Other: Sheet of paper that partially covers the board

Source: Created by authors