GHItaly19: Research Perspectives on Game Human Interaction Maria De Marsico Davide Gadia Dario Maggiorini Sapienza University of Rome University of Milan University of Milan Rome, Italy Milan, Italy Milan, Italy demarsico@di.uniroma1.it gadia@di.unimi.it dario@di.unimi.it Ilaria Mariani Laura Anna Ripamonti Politecnico of Milan University of Milan Milan, Italy Milan, Italy ilaria1.mariani@polimi.it ripamonti@di.unimi.it ABSTRACT 1 INTRODUCTION This is a short introduction to the papers presented at the Playing has a fundamental role for animal development. Pup- 3rd Workshop on Games-Human Interaction - GHItaly19, pies learn adults’ actions by imitation, that in many cases that was held in connection with CHItaly 2019. This series works through playing with peers. As for human childhood, of workshops focuses on the multifaceted issues related to the attractiveness of games extends for an even longer time the design and development of human-game interfaces. This [16], and also spans adult life. This especially holds in our entails multidisciplinary competences and skills, and the fi- digital era, where technology multiplies and extends the pos- nal quality of the User eXperience depends on how consis- sibilities to design a huge variety of different categories of tently and smartly they are exploited. As a matter of fact, games. Increasingly sophisticated electronic devices, the pos- users’ engagement and satisfaction rely on the wise design sibility of online sharing and communication, and finally and skilled evaluation of the produced (multidimensional) virtual, augmented and mixed reality represent a set of ad- artifacts. This gains even more critical importance since the vanced building blocks for game design and development. application of video games has long overcome the borders At the same time, playing is not synonym for pure amuse- of amusement, to spur new possibilities for, e.g., continuous ment anymore. It finds a growing number of different ap- healthcare and education. plications. As a consequence, the design and development of video games can be considered as a fascinating and mul- CCS CONCEPTS tifaceted research field, that embeds the results from both • Human-centered computing → Human computer in- science and humanities regarding several technological, cog- teraction (HCI); Interaction paradigms; Interaction de- nitive, and social aspects. Topics from many computer sci- sign; Interaction design process and methods; Visual- ence areas are involved, from computer graphics to software ization; • Social and professional topics → User charac- engineering and usability evaluation. But also physics, psy- teristics; • Computing methodologies → Machine learn- chology and neurophysiology, industrial design, and, on the ing; Computer graphics; • Applied computing → Arts other side, literature, history, economy, visual arts, semiotics, and humanities; • Software and its engineering → In- etc., provide useful elements for an effective game design. teractive games. On one side, video games represent, now more than ever, a growing industrial field, as clearly testified by the increas- KEYWORDS ing revenues and the impact on the job market. On the other HCI; Game design; Usability; Brain Computer Interface; Adap- side, the deeper and deeper effect on different aspects of tation; Affective computing; Machine Learning; Procedural people’s everyday life calls for a true research commitment. Content Generation; Social interaction; Interaction Design; Gaming scope has gone well beyond the original boundaries Information visualization; Player experience; Storytelling; of enjoyment to extend to novel targets, from learning to Gamification rehabilitation and promotion of cultural heritage [6][8]. Un- fortunately, just due to the variety of involved knowledge and skill, the field still presents a high fragmentation. GHItaly19: 3rd Workshop on Games-Human Interaction, September 23, 2019, Padova, Italy In its first two editions [5][7], the GHItaly workshop aimed Copyright © 2019 for this paper by its authors. Use permitted under Cre- at constituting a bridge among the many different disciplinary ative Commons License Attribution 4.0 International (CC BY 4.0). areas involved, trying to decrease the still existing cultural for smart interactions. The progressive prototype develop- gaps and to establish a common ground and a crossroads ment suggests a series of guidelines and best practices, but for related research. Since the workshop series continues as also leaves open problems concerning the hybridisation of a space of interdisciplinary dialogue and exchange, also the digital means in an analog play experience. A hybrid board third edition has encouraged and welcomed the presence game could simplify the playing experience by entrusting of different and complementary perspectives. Some papers data to the game system itself. The players are not forced to dealt with User eXperience and preferences: the analysis of elaborate and remember the information necessary to set the users’ attitudes is a necessary condition to increase the up a gaming strategy, so that their enjoyment can be in- quality of artifacts whose purpose is both to entertain and creased via a lower effort. The developed prototype relies elicit fun [3][9][13][19], and also to support serious activ- on the strategy of “putting knowledge” in the world [15]. ities, e.g., education. The emerging ideas can inspire new Among the problems underlined by the experiments, it is ways of researching, teaching, and working on the design worth underlining a series of implications able to reduce the and development of video games, in their broadest role: both feasibility of designing hybrid artefacts. Establishing share entertainment and applied goals. and exchange of data between different electronic compo- nents, each with very specific tasks, might lead to problems 2 SUMMARY OF PAPERS PRESENTED AT concerning the efficiency, noise and latency of the commu- GHITALY19 nication. GHItaly19 included contributions that can be divided into two broad groups. The papers in the first group deal with The paper “A.T.L.A.S.: Automatic Terrain and Labels Assem- game design issues. The papers in the second group focus bling Software” by De Francesco et al. [4] presents a tool for on both static and dynamic analysis of player’s preferences. the automatic creation of complex imaginary worlds, that make up the basis for video games. Game designers or game The paper “Analysis of Advertising in E-Sports Broadcasts” writers can testify that making an imaginary world “credi- by Kareinen, et al. [12] proposes to develop a new strategy ble” and convincing for the player, requires it to be “con- for advertisement in e-sports broadcasts. It uses game data, sistent”. In other words, no aspect should be perceived as to increase the audience interest. The video game Counter- “weird” or “out-of-place”. The proposed tool is methodolog- Strike represents the case-study. In Global Offensive (CSGO), ically different from other imaginary world generators pro- two teams of five players try to reach 16 round wins to posed in the Procedural Content Generation (PCG) field [18]. achieve a map win. The “commentators” discuss the actions It was developed as one of the components of a more com- of the previous round in a “freezetime” at the beginning of plex story-driven approach to the generation of video games: every round. The proposal is in a preliminary stage and de- in fact, A.T.L.A.S. smoothly integrates its output with that of velops a new analytic tool for commentators that uses the GHOST [11]. The latter tool semi-automatically produces Game State Integration of CSGO. In this way, their analy- the main narrative structure of a story and its characters. sis is better supported in order to provide a new advertising This provides a solid backbone to be filled by the game/level element for more attractive broadcasts. Viewers get infor- designer. A.T.L.A.S. follows a top-down approach to gen- mation in the form of game statistics. At the same time, the erate the features of the imaginary worlds. It moves from display shows a possible new spot for advertising the spon- the general/outer level (i.e., the generation of the Earth’s sors of the broadcast in the place of the more traditional crust) to the particular/local (roads, villages, specific build- static advertising. Unfortunately, a survey carried out with ings and locations). The generation process is divided into some volunteers revealed that, though the new strategy was two phases. The first phase creates the orography of the appreciated, the new kind of advertisement was not better environment, according to elements of physical geography, remembered than the traditional static one. that can be controlled by the game/level designer. The sec- ond phase follows political geography principles, and adds Cavicchini and Mariani, in their paper “Hybrid board game: points of interest to the map (i.e., cities, villages, roads, etc.). Possibilities and implications from an interaction design per- Most of them can be imported directly from a structure for spective” [14], deal with interaction design. They investi- a story produced by GHOST. gate the combination of digital and analog media in the con- text of IoT (Internet of Things). The case study is represented In the second group, the paper “A BCI-based Assessment of by boardgames. The authors consider that an object-based a Player’s State of Mind for Game Adaptation” by Carofiglio, system such as a tabletop represents an effective support et al. [2] discusses how a passive Brain Computer Interface (BCI) can be used to enhance the gaming experience through https://blog.counter-strike.net/index.php/about/ adaptation by assessing the state of player’s mind. The au- and enjoyment of video game players, based on the Self- thors assume that both cognitive and emotional factors drive Determination Theory (SDT). SDT turned out to be a highly the process of playing videogames [20]. Therefore, they fo- successful tool for the investigations on video games un- cus on the efficient recognition and classification of states der several perspectives. It predicts that people tend to be of the players during gameplay, including boredom, flow proactive and engaged in activities that can satisfy three and stress. Features extracted from EEG signals should al- specific intrinsic needs: the need for competence (the hu- low state classification. The authors performed an experi- man innate desire to grow one’s own abilities), the need for ment collecting data from 35 subjects playing at a horror autonomy (the human innate desire to be the causal agent video game, for a total of 240 EEG signals recordings. The of one’s own life), the need for relatedness (the human in- horror video game presented specific features and mechan- nate need for meaningful interactions with peers) [17]. In ics causing emotional impact on the player, e.g., the use of the paper, the authors test the SDT predictions by manip- a measure of “mental health”, an adaptive use of rendering ulating the amount of reward given to the player within effects and sounds, the behaviour of Non-Player Characters Torchlight II commercial video game, in opposition to ap- (NPCs). At the end, the players had to answer to a question- plied games and gamified application used in previous stud- naire to assess their perceived emotions, engagement, and ies. The collected data include game metrics, video record- the appropriateness of their skills with respect to the game ings, and self-reported feedback. The analysis was carried challenges. The answers of the questionnaire allowed the in- out on information from two groups of video game players terpretation of the physiological data, and the annotation of that participated in a 60-minutes play session. The control the collected data. Results showed that the emotional states group played the standard version of the game, while the ex- extracted by BCI are coherent with the self-evaluated mea- perimental group played a version of the game that provided sures. In particular, the level of engagement was mainly co- five times the amount of rewards. Results showed that the herent with experimental conditions. speed of player character growth affected the participants’ perceived competence and their enjoyment of the game, al- The paper “Towards a model to meet players’ preferences though the game metrics indicated that the two gameplay in games” by Bellini [1] presents a different approach to the sessions were almost identical. adaptation of a game to players’ attitude. The author analy- ses the approaches based on Procedural Content Generation 3 CONCLUSION via Machine Learning (PCGML) [21]. PCGML represents a Video games are still almost universally considered as pure new paradigm for the self-driven creation of new contents. amusement artifacts, and assigned a secondary role in both With respect to Procedural Content Generation, the qual- editorial, artistic and scientific worlds. However, a growing ity of the created content is generally higher. It is achieved number of studies and experiences demonstrate that video by integrating automatic creation techniques with Machine game design and development is a technically and cultur- Learning models. The latter models are trained on existing ally rich and challenging area. It is continuously expanding content. PCGML has been applied to create video game con- to “serious” applications like healthcare and education. In tents (e.g., levels structure, story progression), but not yet to this context, it exploits the results from several different re- generate the complete video games. The author applies Ma- search fields. In a complementary way, it can contribute to chine Learning for the recognition of a player’s attitudes, the research in neighboring areas. The works presented at and proposes a PCGML model able to adapt a game to play- GHItaly19, as well as those in the previous workshop edi- ers’ preferences. The game content is procedurally customized tions, clearly demonstrate the interest and relevance of this according to the single player’s profile and to a Player Model fascinating field for both researchers and practitioners. The built during an in-game opaque survey, This entails that, results obtained include new approaches, techniques, and in- during playing, the model records the player’s path, made terdisciplinary exchanges characterizing game design and up by each chosen action, and updates the player’s profile development. However, they are significant to the extent accordingly. At each new level, the algorithm picks a sub- that they can support an improved and more engaging User sequent level according to a probabilistic distribution, that eXperience. It is important to achieve this goal whether the takes into account what is more likely to be interesting for players are people playing for fun or for more serious rea- the player, following the current Player Model. sons, since it is the final expected outcome of any technical as well as theoretical investigation. However, as already ob- The last paper presented at the workshop is “Faster is Bet- served in the previous editions, the complexity implicit in ter: The Speed of Player Character Growth affects Enjoy- ment and Perceived Competence” by Guardini et al. [10]. It presents an interesting study on the perceived competence https://www.torchlight2.com video game design and implementation can stimulate differ- 2246 - Proceedings of GHItaly18 - 2nd International Workshop on Games- ent investigations on several game aspects. These include, Human Interaction. e.g., design strategies and techniques, playability and en- [8] Maria De Marsico, Laura Anna Ripamonti, Davide Gadia, Dario Mag- giorini, and Ilaria Mariani. 2018. GHItaly18: Game-Human Interac- gagement, and user evaluation, as well as specific applica- tion in Research. In CEUR Workshop Proceedings 2246 - Proceedings of tion contexts. Applied games extend their target from educa- GHItaly18 - 2nd International Workshop on Games-Human Interaction. tion to cultural heritage to healthcare. Albeit their diversity, [9] Tracy Fullerton. 2018. Game design workshop: a playcentric approach each contribution presented at the GHItaly19 workshop un- to creating innovative games (Fourth Edition). CRC press. [10] Pietro Guardini, Dalila De Simone, and Rossana Actis-Grosso. 2019. derlines that games are artifacts by far more complex and Faster is Better: The Speed of Player Character Growth affects Enjoy- cross-disciplinary than generally deemed. Moreover, the ex- ment and Perceived Competence. In CEUR Workshop Proceedings - Pro- perience lived by the player has deep implications for their ceedings of GHItaly19 - 3rd International Workshop on Games-Human fruition, whatever is the category they belong to, and espe- Interaction. cially if their goal is not pure amusement. As a consequence, [11] Andrea Guarneri, Laura Anna Ripamonti, Francesco Tissoni, Marco Trubian, Dario Maggiorini, and Davide Gadia. 2017. GHOST: A they can no longer be classified as only “simple” software ap- GHOst STory-writer. In Proceedings of the 12th Biannual Conference plications. A more frequent, intense and fruitful exchange on Italian SIGCHI Chapter (CHItaly ’17). ACM, 24:1–24:9. of ideas, experiences and results among the different disci- [12] Kalle Kareinen, Antti Knutas, and Jussi Kasurinen. 2019. Analysis plines involved should take place, as we hope this will hap- of Advertising in E-Sports Broadcasts. In CEUR Workshop Proceed- pen in the next future. ings - Proceedings of GHItaly19 - 3rd International Workshop on Games- Human Interaction. [13] Raph Koster. 2013. Theory of fun for game design. O’Reilly Media, Inc. REFERENCES [14] Ilaria Mariani and Saverio Cavicchini. 2019. Hybrid board game: Pos- [1] Mattia Bellini. 2019. Towards a model to meet players’ preferences in sibilities and implications from an interaction design perspective. In games. In CEUR Workshop Proceedings - Proceedings of GHItaly19 - 3rd CEUR Workshop Proceedings - Proceedings of GHItaly19 - 3rd Interna- International Workshop on Games-Human Interaction. tional Workshop on Games-Human Interaction. [2] Valeria Carofiglio, Berardina Nadja De Carolis, and Francesca [15] Don Norman. 2013. The design of everyday things: Revised and ex- D’Errico. 2019. A BCI-based Assessment of a Player’s State of Mind panded edition. Basic books. for Game Adaptation. In CEUR Workshop Proceedings - Proceedings of [16] Anthony D Pellegrini and Peter K Smith. 2005. The nature of play: GHItaly19 - 3rd International Workshop on Games-Human Interaction. Great apes and humans. Guilford Press. [3] Mihaly Csikszentmihalyi and Jeanne Nakamura. 1979. The concept [17] Scott Rigby and Richard M. Ryan. 2011. Glued to games: How video of flow. Play and learning (1979), 257–274. games draw us in and hold us spellbound. Praeger/ABC-CLIO. [4] Alessandro De Francesco, Laura Anna Ripamonti, Davide Gadia, and [18] Noor Shaker, Julian Togelius, and Mark J. Nelson. 2016. Procedural Dario Maggiorini. 2019. A.T.L.A.S.: Automatic Terrain and Labels As- Content Generation in Games: A Textbook and an Overview of Current sembling Software. In CEUR Workshop Proceedings - Proceedings of Research. Springer. GHItaly19 - 3rd International Workshop on Games-Human Interaction. [19] Lee Sheldon. 2004. Character Development and Storytelling for Games. [5] Maria De Marsico, Laura Anna Ripamonti, Davide Gadia, Dario Mag- Thomson Course Technology PTR. giorini, and Ilaria Mariani (Eds.). 2017. CEUR Workshop Proceedings [20] Tan Eduard Sioe-Hao. 2008. Entertainment is Emotion: The Func- 1956 - Proceedings of GHItaly 2017 - 1st International Workshop on tional Architecture of the Entertainment Experience. Media Psychol- Games-Human Interaction. ogy 11, 1 (2008), 28–51. [6] Maria De Marsico, Laura Anna Ripamonti, Davide Gadia, Dario Mag- [21] Adam Summerville, Sam Snodgrass, Matthew Guzdial, Christoffer giorini, and Ilaria Mariani. 2017. GHItaly 2017: Game-Computer Inter- Holmgård, Amy K. Hoover, Aaron Isaksen Isaksen, Andy Nealen, and action in Research. In CEUR Workshop Proceedings 1956 - Proceedings Julian Togelius. 2018. Procedural Content Generation via Machine of GHItaly 2017 - 1st International Workshop on Games-Human Inter- Learning (PCGML). IEEE Transactions on Games 10, 3 (2018), 257– action. 270. [7] Maria De Marsico, Laura Anna Ripamonti, Davide Gadia, Dario Mag- giorini, and Ilaria Mariani (Eds.). 2018. CEUR Workshop Proceedings