Towards Lenses for Reviewing Playfulness in HCI

From playful design methods to artifacts that facilitate experiences of play, playfulness has been widely explored in HCI. However, this heterogeneous field of research lacks common ground – from a methodological, theoretical, as well as design perspective – which in turn creates problems that hinder related research outcomes. Towards common ground on playfulness, we propose these five lenses for reviewing, synthesizing, and critically engaging with literature: area of implementation, targeted audience, type of contribution, methodology, and motivation. To spark a discussion around the lenses, we probe them on a pilot review of CHI Play literature and discuss their relevance to future literature reviews that work towards making sense of the state of playfulness in HCI.

Given the growing heterogeneity of HCI's engagement with play and playfulness, we argue that there is a need for common grounding in this space.What is the state of play in HCI?What methodologies, methods, and tools do researchers employ?Why do they design for play, and for whom?Those and many other questions need to be answered if we are to understand what (and how) we are doing as a play(ful) research community.
Recent works have reviewed specific layers of playful HCI design and research, such as social play [52,60], active play [23], game research [14,27,39,48], gamification research [45], playful humanfood interaction [4,7], and methods at CHI Play [1,65].However, we have not found a comprehensive review that looks at the field holistically and from multiple lenses.We see that as a limitation that hinders our and other researchers' capacity to make an impact: identifying from where and how studies on playfulness draw knowledge can help us understand what are the theoretical and methodological foundations in the field.Additionally, it will help us identify what are the positions developed within playfulness in HCI and how they relate to other disciplines.We also see great potential in articulating to whom playfulness researchers address their work and their motivation for designing playfulness.
Thus, we have an agenda of overcoming those limitations and contributing to establishing productive common grounds in playful HCI.To that end, here we propose five lenses for analysing the literature on playfulness in HCI: area of implementation, targeted audience, type of contribution, methodology, and motivation (see details in section 2).Building on these lenses, we reviewed a body of playfulness-related research published at CHI Play as a first step towards establishing a base to inspire future broader sense-making efforts in playful HCI.As a pilot review for an initial test of the lenses, we scoped our definition of playfulness as an attitude without the activity of play, which means that we excluded most game-related work from the corpus (more details in section 3).To be clear, the goal of this paper is not to provide an analysis of the literature itself; instead, we aim to probe our in-progress lenses of analysis, discuss them with the games and play community within CHI, and strengthen them so they can be implemented in future work (which should include a broader definition of playfulness).This late-breaking-work aims to facilitate a discussion around which lenses could be used to review literature centered on playfulness in HCI and how to best employ them.In the next sections, we unpack our approach to the literature review, present the preliminary findings from our pilot review, and discuss the implications of the lenses to future studies in the same area.

DEFINING LENSES FOR REVIEWING PLAYFULNESS IN HCI
In our work, we faced issues such as making sense of multiple definitions of playfulness, understanding the different motivations to use playfulness in HCI, and identifying the methodological underpinnings of published work.We referred to the literature for inspiration to approach our motivating concerns, leading to five lenses for reviewing playfulness in HCI.In the next paragraphs we will describe the meaning and related work to each lens.

Area of Implementation
The area of implementation lens inquires into where playfulness is implemented.That is, it looks at its sub-field of application, e.g.health, education, or human-food interaction.We draw inspiration from studies that, despite researching specific applications of playfulness, identified such sub-fields of implementation.For example, Altarriba Bertran and colleagues listed application domains in Human-Food Interaction (HFI) [3,4,7], O'Donnell et al. classified the literature on gamification into areas of research to evaluate the multidisciplinary quality of the field [45], and Carter et al. grouped CHI publications on games with normative benefits towards a specific area (e.g.health or education) [14].In line with these studies, we used an open coding approach to allow unforeseen categories to emerge directly from the data.

Targeted Audience
This lens was inspired by recent work that discusses diversity in HCI research [16,24,28,37,46,57,58].Understanding for whom we design playfulness is a first step towards reflecting on diversity in the field.The targeted audience lens uses an open coding approach to examine the intended receivers of the proposed playful intervention; that is, the end-users of an application or system.For studies that involved study participants (interviews, focus groups, workshops, etc.), we also took note of those, since, in some cases, they differed from targeted users.

Type of Contribution
The type of contribution lens looks at which type of contribution is provided by playful HCI publications, e.g.novel interaction modality, design implications, novel design method, etc.The categories are based on Wobbrock and and Kientz's seven research contribution types in HCI: empirical research, artifact, methodological, theoretical, dataset, survey, and opinion [68].During the analysis, we observed that empirical contributions could be divided into two other categories, similar to the classification used at CHI 2016: "empirical study that tells us about how people use a system" or "empirical study that tells us about people" [68].To understand if there is a difference in numbers between these two approaches, we divided the category empirical into empirical -artifact and empirical -people.Due to the attention given to the pitfalls and benefits of "implications for design" to research in HCI, we included it as an extra category [11,22,30,35,47].In this lens, it is likely that one same publication presents multiple types of contribution.To address that, the lens allows for sorting into multiple categories.

Methodology
The methodology lens looks at how researchers use playfulness not only as an object of design but also as a way of enacting the design process itself.By methodology we mean any identifiable set of grounding theories, lenses, frameworks, collection of methods, or procedures for a design process or study.Our motivation to include a classification of methodologies resonates with existing efforts to review methodological approaches in the field (e.g.review the use of the Game Experience Questionnaire [39], methods to research game streaming [27], Statistical Significance Testing [65], and the transparency of measurement reporting [1]).These studies used pre-coded terms based on existing literature for their reviews.In contrast, we decided to use an open coding approach to uncover the diversity of sub-areas present in the sample, which would be limited otherwise.

Motivation
The motivation lens looks at the underlying reasons driving the implementation of play.It uses pre-defined terms for classification: hedonic, eudaimonic, and problem-oriented.The first two terms are based on Sharp and Thomas' distinction of the hedonic and eudaimonic roles of games [54].Hedonic refers to playful experiences with no "additional forces or reasons beyond their enjoyment" [54].
In contrast to hedonic, eudaimonic experiences provide an extra outcome than the experience itself, they promote well-being.The problem-oriented category covers studies that provide a solution to a pre-defined problem, e.g., a tool to motivate people to exercise or a design that helps people reflect on sustainability.

APPROACH TOWARDS LENSES FOR REVIEWING PLAYFULNESS IN HCI
To test the efficiency of the lenses, we applied them in a pilot review of CHI Play proceedings.A pilot review means that the analysis performed here is not meant as a comprehensive literature review but rather as a stress-test of the lenses.In the next sections, we define our view on playfulness (which determined the scope of this pilot review), describe how we build the corpus, and explain how we analysed the data.

What do We Mean by Playfulness?
Although play and playfulness have been used interchangeably in the literature, we adhere to the definition of playfulness as a playful attitude -being fun, creative, social, engaging -without the actual activity of play [5,6,13,56].For example, telling jokes is a way to playfully engage with language, and asking Siri silly questions is a fun way to interact with the virtual assistant.Both examples do not have the activity of play as a goal but rather being social or having some fun by disrupting the use of technology.In that sense, playfulness can be present in the activity of play, for example, when kids get together in playgrounds or someone plays video games.However, here we can still make a distinction between playfulness as an attitude versus play as an activity.In this pilot review, we adhere to the use of playfulness as an attitude.

Database and Corpus
The body of research at CHI Play was chosen as a starting point for this pilot review, as it is a community that traditionally has been centered around games and play in HCI.Research at other conferences might also engage with playfulness, however, we decided not to include them to focus on probing our lenses within the scope of a pilot for a short paper.We searched proceedings from the establishment of the conference in 2014 until 2022.The ACM Digital Library was the database used to collect publications that fall into the area of playfulness as an attitude.To achieve this, we eliminated, for example, publications that explicitly focused on game mechanics or places specifically designed for play.Even though a few publications do implement games in their studies, the focus of their contribution (methods for playfulness) is still relevant to our area of interest.We stress the importance that future studies include a holistic view of playfulness by analyzing publications that focus both on playfulness as the attitude and as the activity.Nevertheless, in this pilot review we chose to work with a sample of publications to keep the analysis manageable and in accordance with our goal of testing the lenses.Our preliminary results help us reflect and iterate on the lenses, rather than provide a comprehensive overview of the literature.The terms in the search query were snowballed from an unpublished scoping review on playfulness in HCI performed by the first author on the ACM Digital Library and Google Scholar [69].This is not an exhaustive list but it serves the purpose of this pilot literature review.We used the following query to search the ACM Digital Library by publication title, filtering results by the proceeding series "CHI PLAY: Computer-Human Interaction in Play".
(playfulness OR play OR playthings OR "participatory play" OR playfication OR "playful experience design" OR "playful participatory practices" OR playable OR plex OR "playful experiences framework" OR "player-centered design" OR "player centered design") AND NOT game* It returned 195 results on April 21, 2023.

Selection Criteria
The first author screened the corpus by title and abstract based on the criteria below.Borderline publications were skim-read in full and a decision was made accordingly.From 195 publications returned by the search at the ACM Digital Library, 109 were removed by title, 29 by abstract, and 20 after a full skim-read.Thirty seven publications remained for analysis.Exclusion criteria: • Focused only on game aspects: mechanics, players, application of games, or play in games; • Focus on places specifically designed for play (e.g.playgrounds) and doesn't address playfulness directly; • Session details, talks, doctoral consortium papers, workshop proposals, and interactivity session publications.
Inclusion criteria: • Implements or researches playfulness as an attitude of play; • New methods or evaluation of existing methods for measuring playfulness in the context of games, but that can be applied to other non-game-oriented scenarios.

Data Analysis Procedures
The 37 publications that remained from the selection were analyzed through an affinity diagram [29].The initial labels were generated by reading each article and annotating their content in relation to the five lenses (area of implementation, methodology, type of contribution, targeted users, and motivation).Since we started our analysis from the perspective of the five lenses, the top-most level categories referred to them.Considering our data comes from categorizing the content of research articles, we slightly adapted the affinity diagram recommendations to fit our data format (e.g., the annotations are not written in the first person).The first author performed the first round of analysis.The initial diagram was then presented to the other authors, who discussed and iterated the categories one more time.The authors used the online tool Miro1 to categorize the annotations digitally.

PRELIMINARY RESULTS: REFLECTING ON THE LENSES FOR REVIEWING PLAYFULNESS IN HCI
The main contribution of this paper is to establish and reflect on a set of lenses to review playfulness in HCI literature.These lenses can be applied in future work to synthesize the literature on playfulness in HCI, working towards establishing a productive common ground in this area of research.Furthermore, we invite the research community to engage with the findings discussed below to move towards a more conscious, reflexive field of research.
It would be beneficial for future studies to provide an additional comparative analysis of the sub-fields.This alternative organization of the findings can showcase how sub-areas of research perform in comparison to each other, stressing opportunities for collaboration.
Even for the publications that explicitly described the group of users or participants they engage with, we noticed a lack of reflection on the implications of designing for these communities or even how their characteristics (physical, social, economic, etc.) influenced the proposed work.Here we see connections to Spiel's [58] provocations for embodied interaction, which is very relevant to be investigated in the context of play research.They provide a collection of points for reflection that researchers should engage in their work to "[...] rattle and challenge our complicity with dominant power structures and as a starting point towards dismantling the oppressive norms implicitly and explicitly guiding our designs." [58, p. 10] Himmelsbach et al. [28] also noted that simply understanding the user group of a project is a step that might lead to more diversityawareness in the first place.Such reflections would make the work around playfulness more conscious of current discussions in the broader HCI community regarding representation and inclusion.
Based on these findings, we are confident in saying that CHI Play needs more theoretical and survey contributions on playfulness as an attitude.Nevertheless, it is pressing that future studies review and compare all applications of playfulness (as an attitude and as an activity).This lens can also highlight possible imbalances in types of contribution published in different conferences and journals.For example, it can help identifying if theoretical or methodological contributions for playfulness are published in other conferences or even other fields of research (in contrast to CHI and CHI Play, or the field of HCI).It would be interesting to engage with similar questions to the ones made by Carter et al. [14, p. 35]: "Is CHI, as its name suggests, a 'parent' conference for CHI Play?And if so, what kind of parent is it?Is CHI Play for research marginalized from mainstream HCI that does not sufficiently reflect the 'values' of HCI (or at least, CHI reviewers)?Is the purpose of CHI Play to legitimate the discipline of PCI as a subset of HCI, or as an independent field of work in a similar fashion to CSCW?"As a step towards a common ground for playfulness in HCI, we side with Seaborn and colleagues' manifesto for research synthesis in HCI [49,53].Particularly her argument for game user research at CHI Play is valid for playfulness: research synthesis will validate the field of study and enrich primary research; it will add to both primary and secondary research; and the community should adopt established tools and protocols [53].

Methodology
Table 1 illustrates the diversity of methodologies we identified.During the analysis, the lack of detailed descriptions for methodologies was very prominent.This led the authors to perform a second step in the analysis to understand if there was a correlation to the publications' format (work-in-progress/spotlight track, short paper, or full paper).We expected that, in some cases, work-inprogress (WIP) publications would not provide a robust description of their methodology due to the in-progress state of the work and the restricted number of pages.However, in our findings WIPs and full papers performed similarly in relation to their methodologymeaning that several full papers missed aspects in that area.Table 2 provides an overview of the publications divided by the type of publication and how in-depth the methodology is described.We kept the categories study procedure (N=12) and not identified (N=3) from the previous step and re-organized the remaining publications into no dedicated section (N=11) and fully described (N=11).For publications in the no dedicated section category, we identified their approach from theories, concepts, or methods mentioned in the introduction, related work, or discussion sections.For example, [67] describes its design process where they were inspired by the dimensions of the Playful Experience Framework (PLEX).In that case, the PLEX framework was noted as their methodology, even though the authors didn't explicitly state it.
Our results stress the need for dedicated literature reviews that analyse the types of methodologies used in the context of playfulness in HCI.Such in-depth analysis can identify the plurality of approaches in the field and help researchers build a common language for reporting methodologies.We also want to emphasize that the results presented here would benefit from a cross-comparison with other lenses.For example, we did not account for the fact that some of the publications where no methodology was identified are opinion papers, which can mislead into thinking that they lack methodological rigor.The publications' metadata should be considered by future studies implementing the lenses proposed here.Additionally, we suggest including an overview of the results of the lenses altogether to provide a means of comparison.We leave a few questions for future work that could not be answered in the scope of our preliminary analysis: do different types of contributions require different levels of methodological reflection?Are study procedures strong enough methodological grounding?Is there a lack of theoretical positioning?

Motivation
The most interesting finding of this lens is not how many papers are in each category but rather how difficult it was to define their motivation.Most publications did not explicitly discuss their motivation to use playfulness.The publications were classified as eudaimonic (N=7): [7,8,18,20,34,38,52]; hedonist (N=13): [17, 32, 33, 36, 40-44, 51, 63, 66, 67]; and problem-oriented (N=17): [2,9,10,12,15,19,21,25,26,31,50,55,59,61,62,64,70].We encourage future reviews of the literature to reflect on studies' motivation for using playfulness by critically analyzing the content of the publications (e.g., using content analysis in addition to categorization methods).Such analysis could provide examples from the literature and propose a framework to help authors clearly describe their motivations for researching playfulness.Reflecting on motivation has an impact on making sense of the multitude of approaches in HCI by relating studies' use of playfulness to their methodology (which often comes from a field outside of HCI, e.g., education, health, etc.).Moreover, it would contribute to an interesting debate around the meaning of playfulness, particularly in relation to its distinction from playfulness as the activity.We believe that the lack of reflection on the motivation to use playfulness is partially related to the fact that the concept of playfulness itself is very plural, with multiple interpretations across the literature.In relation to the use of playfulness, we could also not easily identify where in the design process playfulness was used.For example, if it was used in the process of designing (e.g., playful ideation methods), if it was the object of study (e.g., observing play), or if it was a goal of the final design (e.g., creating an artifact that invites play or a playful experience).For future work, an extra category could be used to prompt: where is playfulness implemented in the design process?Which elements of the design elicit playfulness and how?

FUTURE WORK AND CONCLUSION
We invite other researchers to discuss, expand, and question the reflections shared here, as we plan to continue this work in the future by incorporating critical and diverse views from the community.We see a great opportunity to investigate how the many WIPs evaluated in this pilot review follow up their discussions, since these publications alone give only a partial view of the studies.We are also intrigued about how results would differ if the five lenses we proposed were applied to other conferences, such as DIS, TEI, IDC, or CHI.Furthermore, future studies should consider looking into SIGs and workshops, since those can offer an overview of which communities exist and how they evolve throughout the years.We also encourage future work to provide a cross comparison between the lenses, opposed to investigating them individually.Although both approaches offer benefits, we understood that the lenses complement each other to highlight the many correlations between different aspects of playfulness.More importantly, there is the opportunity to map out related concepts, theories, methodologies, and methods to enrich the understanding of playfulness in play studies.Other studies should also consider including game-related publications, since those also deal with the topic of playfulness.This can help us make a parallel between the scope of research on playfulness in games and outside that context.Apart from the focus on reviewing the literature on playfulness, we see a lot of opportunity to strengthen the framework proposed here.Our goal is to open a conversation to critically analyze and iterate the lenses.This can be done by proposing new directions to the content of the lenses and engaging the community by, for example, interviewing playfulness researchers to enhance the credibility of the framework.By sharing these initial considerations, we hope to spark ideas for work that investigates and discusses these issues in-depth to contribute to a more conscious, value-driven, and reflexive approach to playful HCI.The lenses proofed to work for us in unveiling the aspects in the literature discussed.Further, this shows what they can offer as a tool for reviewing existing research in the field and positioning one's own work, ultimately contributing to common ground in regards to using the term playfulness.

Table 1 :
Identified methodologies and respective sources.

Table 2 :
Number of publications divided by type and how in-depth their methodology is described.